{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.302042404201517,
  "eval_steps": 3000,
  "global_step": 81000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.742576789855956,
      "epoch": 0.00038902937171756465,
      "grad_norm": 5.21875,
      "learning_rate": 2e-06,
      "loss": 10.7444,
      "mean_token_accuracy": 0.0,
      "num_tokens": 8755.0,
      "step": 5
    },
    {
      "entropy": 10.742605495452882,
      "epoch": 0.0007780587434351293,
      "grad_norm": 5.375,
      "learning_rate": 4.5e-06,
      "loss": 10.7346,
      "mean_token_accuracy": 0.0,
      "num_tokens": 17641.0,
      "step": 10
    },
    {
      "entropy": 10.742611026763916,
      "epoch": 0.001167088115152694,
      "grad_norm": 5.46875,
      "learning_rate": 7e-06,
      "loss": 10.6932,
      "mean_token_accuracy": 0.0,
      "num_tokens": 26293.0,
      "step": 15
    },
    {
      "entropy": 10.742596912384034,
      "epoch": 0.0015561174868702586,
      "grad_norm": 5.03125,
      "learning_rate": 9.5e-06,
      "loss": 10.6732,
      "mean_token_accuracy": 0.0003551477799192071,
      "num_tokens": 35627.0,
      "step": 20
    },
    {
      "entropy": 10.742521381378173,
      "epoch": 0.0019451468585878235,
      "grad_norm": 4.71875,
      "learning_rate": 1.2e-05,
      "loss": 10.5441,
      "mean_token_accuracy": 0.009159033582545818,
      "num_tokens": 43861.0,
      "step": 25
    },
    {
      "entropy": 10.742165279388427,
      "epoch": 0.002334176230305388,
      "grad_norm": 4.1875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.4398,
      "mean_token_accuracy": 0.03247077781707049,
      "num_tokens": 52816.0,
      "step": 30
    },
    {
      "entropy": 10.740872764587403,
      "epoch": 0.0027232056020229526,
      "grad_norm": 3.375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.3142,
      "mean_token_accuracy": 0.03825325649231672,
      "num_tokens": 61370.0,
      "step": 35
    },
    {
      "entropy": 10.737211132049561,
      "epoch": 0.0031122349737405172,
      "grad_norm": 2.765625,
      "learning_rate": 1.95e-05,
      "loss": 10.1536,
      "mean_token_accuracy": 0.044245490059256556,
      "num_tokens": 69444.0,
      "step": 40
    },
    {
      "entropy": 10.730943870544433,
      "epoch": 0.003501264345458082,
      "grad_norm": 2.40625,
      "learning_rate": 2.2e-05,
      "loss": 10.0549,
      "mean_token_accuracy": 0.044146507233381274,
      "num_tokens": 77892.0,
      "step": 45
    },
    {
      "entropy": 10.725597763061524,
      "epoch": 0.003890293717175647,
      "grad_norm": 2.078125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.9865,
      "mean_token_accuracy": 0.04212083183228969,
      "num_tokens": 86154.0,
      "step": 50
    },
    {
      "entropy": 10.72157793045044,
      "epoch": 0.004279323088893211,
      "grad_norm": 2.0625,
      "learning_rate": 2.7e-05,
      "loss": 9.9514,
      "mean_token_accuracy": 0.04536668732762337,
      "num_tokens": 95589.0,
      "step": 55
    },
    {
      "entropy": 10.716520404815673,
      "epoch": 0.004668352460610776,
      "grad_norm": 1.984375,
      "learning_rate": 2.95e-05,
      "loss": 9.898,
      "mean_token_accuracy": 0.04543796852231026,
      "num_tokens": 105268.0,
      "step": 60
    },
    {
      "entropy": 10.711576366424561,
      "epoch": 0.0050573818323283405,
      "grad_norm": 1.890625,
      "learning_rate": 3.2e-05,
      "loss": 9.7831,
      "mean_token_accuracy": 0.04743263386189937,
      "num_tokens": 113347.0,
      "step": 65
    },
    {
      "entropy": 10.705528163909912,
      "epoch": 0.005446411204045905,
      "grad_norm": 1.9609375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.7233,
      "mean_token_accuracy": 0.04614104889333248,
      "num_tokens": 121235.0,
      "step": 70
    },
    {
      "entropy": 10.699932765960693,
      "epoch": 0.00583544057576347,
      "grad_norm": 1.828125,
      "learning_rate": 3.7e-05,
      "loss": 9.674,
      "mean_token_accuracy": 0.045549771189689635,
      "num_tokens": 129529.0,
      "step": 75
    },
    {
      "entropy": 10.695240116119384,
      "epoch": 0.0062244699474810344,
      "grad_norm": 1.84375,
      "learning_rate": 3.95e-05,
      "loss": 9.6104,
      "mean_token_accuracy": 0.046765000373125074,
      "num_tokens": 137918.0,
      "step": 80
    },
    {
      "entropy": 10.686996364593506,
      "epoch": 0.006613499319198599,
      "grad_norm": 1.921875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.5017,
      "mean_token_accuracy": 0.05623045340180397,
      "num_tokens": 146445.0,
      "step": 85
    },
    {
      "entropy": 10.671263790130615,
      "epoch": 0.007002528690916164,
      "grad_norm": 1.84375,
      "learning_rate": 4.45e-05,
      "loss": 9.4967,
      "mean_token_accuracy": 0.057636797428131104,
      "num_tokens": 155904.0,
      "step": 90
    },
    {
      "entropy": 10.655223369598389,
      "epoch": 0.007391558062633729,
      "grad_norm": 1.8125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.4006,
      "mean_token_accuracy": 0.05863568969070911,
      "num_tokens": 164436.0,
      "step": 95
    },
    {
      "entropy": 10.634439468383789,
      "epoch": 0.007780587434351294,
      "grad_norm": 1.7421875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3022,
      "mean_token_accuracy": 0.06445249132812023,
      "num_tokens": 172985.0,
      "step": 100
    },
    {
      "entropy": 10.591667461395264,
      "epoch": 0.008169616806068859,
      "grad_norm": 1.6953125,
      "learning_rate": 5.2e-05,
      "loss": 9.1936,
      "mean_token_accuracy": 0.06430471092462539,
      "num_tokens": 182084.0,
      "step": 105
    },
    {
      "entropy": 10.53823823928833,
      "epoch": 0.008558646177786422,
      "grad_norm": 1.640625,
      "learning_rate": 5.45e-05,
      "loss": 9.0792,
      "mean_token_accuracy": 0.0625245351344347,
      "num_tokens": 190742.0,
      "step": 110
    },
    {
      "entropy": 10.481583213806152,
      "epoch": 0.008947675549503988,
      "grad_norm": 1.7109375,
      "learning_rate": 5.7e-05,
      "loss": 8.9948,
      "mean_token_accuracy": 0.06432581804692745,
      "num_tokens": 199600.0,
      "step": 115
    },
    {
      "entropy": 10.431952953338623,
      "epoch": 0.009336704921221552,
      "grad_norm": 1.671875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.8995,
      "mean_token_accuracy": 0.06441468745470047,
      "num_tokens": 207883.0,
      "step": 120
    },
    {
      "entropy": 10.365525436401366,
      "epoch": 0.009725734292939117,
      "grad_norm": 1.5,
      "learning_rate": 6.2e-05,
      "loss": 8.8018,
      "mean_token_accuracy": 0.06325564533472061,
      "num_tokens": 216216.0,
      "step": 125
    },
    {
      "entropy": 10.309541988372803,
      "epoch": 0.010114763664656681,
      "grad_norm": 1.46875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.6768,
      "mean_token_accuracy": 0.06168311797082424,
      "num_tokens": 225250.0,
      "step": 130
    },
    {
      "entropy": 10.241107082366943,
      "epoch": 0.010503793036374246,
      "grad_norm": 1.3515625,
      "learning_rate": 6.7e-05,
      "loss": 8.5623,
      "mean_token_accuracy": 0.059869692847132686,
      "num_tokens": 233748.0,
      "step": 135
    },
    {
      "entropy": 10.114632701873779,
      "epoch": 0.01089282240809181,
      "grad_norm": 1.2265625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.4505,
      "mean_token_accuracy": 0.06234452910721302,
      "num_tokens": 242629.0,
      "step": 140
    },
    {
      "entropy": 10.001305103302002,
      "epoch": 0.011281851779809376,
      "grad_norm": 1.46875,
      "learning_rate": 7.2e-05,
      "loss": 8.3258,
      "mean_token_accuracy": 0.05941925346851349,
      "num_tokens": 251458.0,
      "step": 145
    },
    {
      "entropy": 9.833595371246338,
      "epoch": 0.01167088115152694,
      "grad_norm": 1.203125,
      "learning_rate": 7.45e-05,
      "loss": 8.2078,
      "mean_token_accuracy": 0.06866976507008075,
      "num_tokens": 260557.0,
      "step": 150
    },
    {
      "entropy": 9.655412101745606,
      "epoch": 0.012059910523244505,
      "grad_norm": 1.09375,
      "learning_rate": 7.7e-05,
      "loss": 8.1497,
      "mean_token_accuracy": 0.06398723609745502,
      "num_tokens": 269262.0,
      "step": 155
    },
    {
      "entropy": 9.519824028015137,
      "epoch": 0.012448939894962069,
      "grad_norm": 1.0703125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.0264,
      "mean_token_accuracy": 0.06563923060894013,
      "num_tokens": 277804.0,
      "step": 160
    },
    {
      "entropy": 9.293780612945557,
      "epoch": 0.012837969266679634,
      "grad_norm": 1.1015625,
      "learning_rate": 8.2e-05,
      "loss": 7.9388,
      "mean_token_accuracy": 0.0646277904510498,
      "num_tokens": 287450.0,
      "step": 165
    },
    {
      "entropy": 9.156445980072021,
      "epoch": 0.013226998638397198,
      "grad_norm": 0.96875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.9051,
      "mean_token_accuracy": 0.06273124553263187,
      "num_tokens": 296497.0,
      "step": 170
    },
    {
      "entropy": 8.887872219085693,
      "epoch": 0.013616028010114764,
      "grad_norm": 0.93359375,
      "learning_rate": 8.7e-05,
      "loss": 7.7677,
      "mean_token_accuracy": 0.06659091413021087,
      "num_tokens": 305414.0,
      "step": 175
    },
    {
      "entropy": 8.771506786346436,
      "epoch": 0.014005057381832327,
      "grad_norm": 0.78125,
      "learning_rate": 8.95e-05,
      "loss": 7.7563,
      "mean_token_accuracy": 0.06347964778542518,
      "num_tokens": 313639.0,
      "step": 180
    },
    {
      "entropy": 8.585464668273925,
      "epoch": 0.014394086753549893,
      "grad_norm": 0.8125,
      "learning_rate": 9.2e-05,
      "loss": 7.7352,
      "mean_token_accuracy": 0.06393053829669952,
      "num_tokens": 322855.0,
      "step": 185
    },
    {
      "entropy": 8.50670690536499,
      "epoch": 0.014783116125267459,
      "grad_norm": 0.85546875,
      "learning_rate": 9.45e-05,
      "loss": 7.8169,
      "mean_token_accuracy": 0.06778145208954811,
      "num_tokens": 332071.0,
      "step": 190
    },
    {
      "entropy": 8.431924152374268,
      "epoch": 0.015172145496985022,
      "grad_norm": 0.8984375,
      "learning_rate": 9.7e-05,
      "loss": 7.6122,
      "mean_token_accuracy": 0.07193873748183251,
      "num_tokens": 339990.0,
      "step": 195
    },
    {
      "entropy": 8.313230800628663,
      "epoch": 0.015561174868702588,
      "grad_norm": 0.8515625,
      "learning_rate": 9.95e-05,
      "loss": 7.6533,
      "mean_token_accuracy": 0.06932461187243462,
      "num_tokens": 348581.0,
      "step": 200
    },
    {
      "entropy": 8.288263511657714,
      "epoch": 0.01595020424042015,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000102,
      "loss": 7.6415,
      "mean_token_accuracy": 0.07069163881242276,
      "num_tokens": 357032.0,
      "step": 205
    },
    {
      "entropy": 8.254517364501954,
      "epoch": 0.016339233612137717,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.607,
      "mean_token_accuracy": 0.06844000443816185,
      "num_tokens": 366562.0,
      "step": 210
    },
    {
      "entropy": 8.228306484222411,
      "epoch": 0.016728262983855283,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000107,
      "loss": 7.5965,
      "mean_token_accuracy": 0.07050794810056686,
      "num_tokens": 374555.0,
      "step": 215
    },
    {
      "entropy": 8.206252288818359,
      "epoch": 0.017117292355572845,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001095,
      "loss": 7.6274,
      "mean_token_accuracy": 0.06776856631040573,
      "num_tokens": 383583.0,
      "step": 220
    },
    {
      "entropy": 8.121715068817139,
      "epoch": 0.01750632172729041,
      "grad_norm": 0.78125,
      "learning_rate": 0.000112,
      "loss": 7.6405,
      "mean_token_accuracy": 0.06792488619685173,
      "num_tokens": 392932.0,
      "step": 225
    },
    {
      "entropy": 8.169992160797118,
      "epoch": 0.017895351099007976,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0001145,
      "loss": 7.4836,
      "mean_token_accuracy": 0.07281714007258415,
      "num_tokens": 401653.0,
      "step": 230
    },
    {
      "entropy": 8.105243301391601,
      "epoch": 0.01828438047072554,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.5741,
      "mean_token_accuracy": 0.07232600301504136,
      "num_tokens": 410432.0,
      "step": 235
    },
    {
      "entropy": 8.124885082244873,
      "epoch": 0.018673409842443103,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.5359,
      "mean_token_accuracy": 0.07120791971683502,
      "num_tokens": 419380.0,
      "step": 240
    },
    {
      "entropy": 8.092760372161866,
      "epoch": 0.01906243921416067,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000122,
      "loss": 7.4537,
      "mean_token_accuracy": 0.07131325155496597,
      "num_tokens": 427682.0,
      "step": 245
    },
    {
      "entropy": 8.064310121536256,
      "epoch": 0.019451468585878234,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0001245,
      "loss": 7.5297,
      "mean_token_accuracy": 0.06986031346023083,
      "num_tokens": 437708.0,
      "step": 250
    },
    {
      "entropy": 8.066622257232666,
      "epoch": 0.0198404979575958,
      "grad_norm": 1.0,
      "learning_rate": 0.000127,
      "loss": 7.49,
      "mean_token_accuracy": 0.07588860243558884,
      "num_tokens": 446480.0,
      "step": 255
    },
    {
      "entropy": 8.031266736984254,
      "epoch": 0.020229527329313362,
      "grad_norm": 0.953125,
      "learning_rate": 0.0001295,
      "loss": 7.524,
      "mean_token_accuracy": 0.06815474294126034,
      "num_tokens": 455273.0,
      "step": 260
    },
    {
      "entropy": 8.06600604057312,
      "epoch": 0.020618556701030927,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000132,
      "loss": 7.4511,
      "mean_token_accuracy": 0.07497221603989601,
      "num_tokens": 464306.0,
      "step": 265
    },
    {
      "entropy": 7.944579219818115,
      "epoch": 0.021007586072748493,
      "grad_norm": 0.984375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.4544,
      "mean_token_accuracy": 0.07208218537271023,
      "num_tokens": 472812.0,
      "step": 270
    },
    {
      "entropy": 7.977314424514771,
      "epoch": 0.02139661544446606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.437,
      "mean_token_accuracy": 0.07208085916936398,
      "num_tokens": 481326.0,
      "step": 275
    },
    {
      "entropy": 8.039547538757324,
      "epoch": 0.02178564481618362,
      "grad_norm": 0.921875,
      "learning_rate": 0.0001395,
      "loss": 7.377,
      "mean_token_accuracy": 0.0771931130439043,
      "num_tokens": 489147.0,
      "step": 280
    },
    {
      "entropy": 7.921053743362426,
      "epoch": 0.022174674187901186,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.3711,
      "mean_token_accuracy": 0.08139716684818268,
      "num_tokens": 497455.0,
      "step": 285
    },
    {
      "entropy": 7.94639024734497,
      "epoch": 0.02256370355961875,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001445,
      "loss": 7.3084,
      "mean_token_accuracy": 0.08105910569429398,
      "num_tokens": 506735.0,
      "step": 290
    },
    {
      "entropy": 7.939362382888794,
      "epoch": 0.022952732931336317,
      "grad_norm": 0.953125,
      "learning_rate": 0.000147,
      "loss": 7.3807,
      "mean_token_accuracy": 0.07725862711668015,
      "num_tokens": 515976.0,
      "step": 295
    },
    {
      "entropy": 7.8592973232269285,
      "epoch": 0.02334176230305388,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001495,
      "loss": 7.3728,
      "mean_token_accuracy": 0.0786518905311823,
      "num_tokens": 524986.0,
      "step": 300
    },
    {
      "entropy": 7.871623373031616,
      "epoch": 0.023730791674771445,
      "grad_norm": 1.03125,
      "learning_rate": 0.000152,
      "loss": 7.3857,
      "mean_token_accuracy": 0.08207501620054244,
      "num_tokens": 533146.0,
      "step": 305
    },
    {
      "entropy": 7.966270780563354,
      "epoch": 0.02411982104648901,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.2788,
      "mean_token_accuracy": 0.08436923623085021,
      "num_tokens": 541704.0,
      "step": 310
    },
    {
      "entropy": 7.803447914123535,
      "epoch": 0.024508850418206576,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000157,
      "loss": 7.2596,
      "mean_token_accuracy": 0.08413073793053627,
      "num_tokens": 550088.0,
      "step": 315
    },
    {
      "entropy": 7.808595514297485,
      "epoch": 0.024897879789924138,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001595,
      "loss": 7.2537,
      "mean_token_accuracy": 0.0833888128399849,
      "num_tokens": 558763.0,
      "step": 320
    },
    {
      "entropy": 7.894071006774903,
      "epoch": 0.025286909161641703,
      "grad_norm": 0.96875,
      "learning_rate": 0.000162,
      "loss": 7.3379,
      "mean_token_accuracy": 0.08004550337791443,
      "num_tokens": 567538.0,
      "step": 325
    },
    {
      "entropy": 7.824001407623291,
      "epoch": 0.02567593853335927,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.3535,
      "mean_token_accuracy": 0.07598577216267585,
      "num_tokens": 577626.0,
      "step": 330
    },
    {
      "entropy": 7.784535598754883,
      "epoch": 0.026064967905076834,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.299,
      "mean_token_accuracy": 0.07876510173082352,
      "num_tokens": 587163.0,
      "step": 335
    },
    {
      "entropy": 7.756780767440796,
      "epoch": 0.026453997276794396,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.2511,
      "mean_token_accuracy": 0.08080902062356472,
      "num_tokens": 595896.0,
      "step": 340
    },
    {
      "entropy": 7.8993690490722654,
      "epoch": 0.026843026648511962,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.2821,
      "mean_token_accuracy": 0.08391863107681274,
      "num_tokens": 603815.0,
      "step": 345
    },
    {
      "entropy": 7.769748687744141,
      "epoch": 0.027232056020229527,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.3017,
      "mean_token_accuracy": 0.08513520881533623,
      "num_tokens": 612208.0,
      "step": 350
    },
    {
      "entropy": 7.818035459518432,
      "epoch": 0.027621085391947093,
      "grad_norm": 0.984375,
      "learning_rate": 0.000177,
      "loss": 7.2697,
      "mean_token_accuracy": 0.08533527255058289,
      "num_tokens": 621840.0,
      "step": 355
    },
    {
      "entropy": 7.816836357116699,
      "epoch": 0.028010114763664655,
      "grad_norm": 0.984375,
      "learning_rate": 0.0001795,
      "loss": 7.2659,
      "mean_token_accuracy": 0.08314265757799148,
      "num_tokens": 630770.0,
      "step": 360
    },
    {
      "entropy": 7.714138650894165,
      "epoch": 0.02839914413538222,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000182,
      "loss": 7.2661,
      "mean_token_accuracy": 0.08392885178327561,
      "num_tokens": 640313.0,
      "step": 365
    },
    {
      "entropy": 7.856315994262696,
      "epoch": 0.028788173507099786,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0001845,
      "loss": 7.2949,
      "mean_token_accuracy": 0.08515013977885247,
      "num_tokens": 648594.0,
      "step": 370
    },
    {
      "entropy": 7.847173547744751,
      "epoch": 0.02917720287881735,
      "grad_norm": 1.3125,
      "learning_rate": 0.000187,
      "loss": 7.2775,
      "mean_token_accuracy": 0.08574092835187912,
      "num_tokens": 656478.0,
      "step": 375
    },
    {
      "entropy": 7.662766361236573,
      "epoch": 0.029566232250534917,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001895,
      "loss": 7.2441,
      "mean_token_accuracy": 0.08885113224387169,
      "num_tokens": 665405.0,
      "step": 380
    },
    {
      "entropy": 7.820247316360474,
      "epoch": 0.02995526162225248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000192,
      "loss": 7.2536,
      "mean_token_accuracy": 0.08792344853281975,
      "num_tokens": 674301.0,
      "step": 385
    },
    {
      "entropy": 7.737472772598267,
      "epoch": 0.030344290993970045,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001945,
      "loss": 7.2531,
      "mean_token_accuracy": 0.09053674265742302,
      "num_tokens": 683159.0,
      "step": 390
    },
    {
      "entropy": 7.638981580734253,
      "epoch": 0.03073332036568761,
      "grad_norm": 1.03125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0875,
      "mean_token_accuracy": 0.09513727799057961,
      "num_tokens": 690992.0,
      "step": 395
    },
    {
      "entropy": 7.6433281898498535,
      "epoch": 0.031122349737405176,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.1734,
      "mean_token_accuracy": 0.09246476516127586,
      "num_tokens": 699934.0,
      "step": 400
    },
    {
      "entropy": 7.765408372879028,
      "epoch": 0.03151137910912274,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000202,
      "loss": 7.2187,
      "mean_token_accuracy": 0.08869696706533432,
      "num_tokens": 708844.0,
      "step": 405
    },
    {
      "entropy": 7.719944286346435,
      "epoch": 0.0319004084808403,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.2784,
      "mean_token_accuracy": 0.08538637533783913,
      "num_tokens": 718148.0,
      "step": 410
    },
    {
      "entropy": 7.750207042694091,
      "epoch": 0.03228943785255787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000207,
      "loss": 7.2089,
      "mean_token_accuracy": 0.08420872502028942,
      "num_tokens": 726859.0,
      "step": 415
    },
    {
      "entropy": 7.683270502090454,
      "epoch": 0.032678467224275434,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002095,
      "loss": 7.2145,
      "mean_token_accuracy": 0.09039615541696548,
      "num_tokens": 735045.0,
      "step": 420
    },
    {
      "entropy": 7.59128589630127,
      "epoch": 0.033067496595993,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000212,
      "loss": 7.0827,
      "mean_token_accuracy": 0.09434614479541778,
      "num_tokens": 743538.0,
      "step": 425
    },
    {
      "entropy": 7.6708005428314205,
      "epoch": 0.033456525967710565,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002145,
      "loss": 7.1281,
      "mean_token_accuracy": 0.0933873251080513,
      "num_tokens": 752518.0,
      "step": 430
    },
    {
      "entropy": 7.578746604919433,
      "epoch": 0.033845555339428124,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.1528,
      "mean_token_accuracy": 0.08902024403214455,
      "num_tokens": 760354.0,
      "step": 435
    },
    {
      "entropy": 7.675575685501099,
      "epoch": 0.03423458471114569,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002195,
      "loss": 7.2022,
      "mean_token_accuracy": 0.0862238198518753,
      "num_tokens": 769522.0,
      "step": 440
    },
    {
      "entropy": 7.607626724243164,
      "epoch": 0.034623614082863255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000222,
      "loss": 7.1284,
      "mean_token_accuracy": 0.09797925055027008,
      "num_tokens": 778548.0,
      "step": 445
    },
    {
      "entropy": 7.591102743148804,
      "epoch": 0.03501264345458082,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002245,
      "loss": 7.143,
      "mean_token_accuracy": 0.09332999736070632,
      "num_tokens": 787666.0,
      "step": 450
    },
    {
      "entropy": 7.638665723800659,
      "epoch": 0.035401672826298386,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.0542,
      "mean_token_accuracy": 0.09499929621815681,
      "num_tokens": 796550.0,
      "step": 455
    },
    {
      "entropy": 7.587637615203858,
      "epoch": 0.03579070219801595,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.1221,
      "mean_token_accuracy": 0.09292486310005188,
      "num_tokens": 804549.0,
      "step": 460
    },
    {
      "entropy": 7.547951364517212,
      "epoch": 0.03617973156973352,
      "grad_norm": 1.0,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.0937,
      "mean_token_accuracy": 0.09402075335383415,
      "num_tokens": 813328.0,
      "step": 465
    },
    {
      "entropy": 7.606256341934204,
      "epoch": 0.03656876094145108,
      "grad_norm": 1.0,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.1404,
      "mean_token_accuracy": 0.0924923688173294,
      "num_tokens": 821733.0,
      "step": 470
    },
    {
      "entropy": 7.556232976913452,
      "epoch": 0.03695779031316864,
      "grad_norm": 1.1875,
      "learning_rate": 0.000237,
      "loss": 7.0732,
      "mean_token_accuracy": 0.09500073790550231,
      "num_tokens": 830243.0,
      "step": 475
    },
    {
      "entropy": 7.638236665725708,
      "epoch": 0.03734681968488621,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002395,
      "loss": 7.0519,
      "mean_token_accuracy": 0.09168064296245575,
      "num_tokens": 839059.0,
      "step": 480
    },
    {
      "entropy": 7.385732889175415,
      "epoch": 0.03773584905660377,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000242,
      "loss": 6.9442,
      "mean_token_accuracy": 0.09834211766719818,
      "num_tokens": 846932.0,
      "step": 485
    },
    {
      "entropy": 7.602040719985962,
      "epoch": 0.03812487842832134,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002445,
      "loss": 7.1423,
      "mean_token_accuracy": 0.08770201802253723,
      "num_tokens": 855333.0,
      "step": 490
    },
    {
      "entropy": 7.395545721054077,
      "epoch": 0.0385139078000389,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000247,
      "loss": 6.9209,
      "mean_token_accuracy": 0.09375111684203148,
      "num_tokens": 863571.0,
      "step": 495
    },
    {
      "entropy": 7.454263353347779,
      "epoch": 0.03890293717175647,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002495,
      "loss": 6.992,
      "mean_token_accuracy": 0.10193705037236214,
      "num_tokens": 872145.0,
      "step": 500
    },
    {
      "entropy": 7.494772291183471,
      "epoch": 0.039291966543474034,
      "grad_norm": 1.171875,
      "learning_rate": 0.000252,
      "loss": 7.0283,
      "mean_token_accuracy": 0.09722969233989716,
      "num_tokens": 880634.0,
      "step": 505
    },
    {
      "entropy": 7.492091417312622,
      "epoch": 0.0396809959151916,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002545,
      "loss": 6.9729,
      "mean_token_accuracy": 0.09642753973603249,
      "num_tokens": 889178.0,
      "step": 510
    },
    {
      "entropy": 7.418828535079956,
      "epoch": 0.04007002528690916,
      "grad_norm": 1.09375,
      "learning_rate": 0.000257,
      "loss": 6.9655,
      "mean_token_accuracy": 0.09765015915036201,
      "num_tokens": 898255.0,
      "step": 515
    },
    {
      "entropy": 7.432187509536743,
      "epoch": 0.040459054658626724,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002595,
      "loss": 7.0224,
      "mean_token_accuracy": 0.09072192162275314,
      "num_tokens": 907369.0,
      "step": 520
    },
    {
      "entropy": 7.4239561557769775,
      "epoch": 0.04084808403034429,
      "grad_norm": 1.15625,
      "learning_rate": 0.000262,
      "loss": 6.9475,
      "mean_token_accuracy": 0.0962302841246128,
      "num_tokens": 915946.0,
      "step": 525
    },
    {
      "entropy": 7.447218513488769,
      "epoch": 0.041237113402061855,
      "grad_norm": 1.03125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.9879,
      "mean_token_accuracy": 0.09441638067364692,
      "num_tokens": 925579.0,
      "step": 530
    },
    {
      "entropy": 7.409813165664673,
      "epoch": 0.04162614277377942,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.977,
      "mean_token_accuracy": 0.09423855319619179,
      "num_tokens": 934732.0,
      "step": 535
    },
    {
      "entropy": 7.454560852050781,
      "epoch": 0.042015172145496986,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.0283,
      "mean_token_accuracy": 0.09172707796096802,
      "num_tokens": 943371.0,
      "step": 540
    },
    {
      "entropy": 7.407678842544556,
      "epoch": 0.04240420151721455,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.8926,
      "mean_token_accuracy": 0.09576850831508636,
      "num_tokens": 951977.0,
      "step": 545
    },
    {
      "entropy": 7.412649726867675,
      "epoch": 0.04279323088893212,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002745,
      "loss": 6.886,
      "mean_token_accuracy": 0.10018825083971024,
      "num_tokens": 960474.0,
      "step": 550
    },
    {
      "entropy": 7.447184467315674,
      "epoch": 0.04318226026064968,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000277,
      "loss": 6.9998,
      "mean_token_accuracy": 0.09603087753057479,
      "num_tokens": 969823.0,
      "step": 555
    },
    {
      "entropy": 7.450275087356568,
      "epoch": 0.04357128963236724,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002795,
      "loss": 6.9175,
      "mean_token_accuracy": 0.10505582615733147,
      "num_tokens": 977417.0,
      "step": 560
    },
    {
      "entropy": 7.354970455169678,
      "epoch": 0.04396031900408481,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.9189,
      "mean_token_accuracy": 0.11012696176767349,
      "num_tokens": 985519.0,
      "step": 565
    },
    {
      "entropy": 7.33652982711792,
      "epoch": 0.04434934837580237,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002845,
      "loss": 6.9209,
      "mean_token_accuracy": 0.10035700276494026,
      "num_tokens": 994136.0,
      "step": 570
    },
    {
      "entropy": 7.332489347457885,
      "epoch": 0.04473837774751994,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000287,
      "loss": 6.8909,
      "mean_token_accuracy": 0.09913867935538292,
      "num_tokens": 1002896.0,
      "step": 575
    },
    {
      "entropy": 7.440053081512451,
      "epoch": 0.0451274071192375,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002895,
      "loss": 6.968,
      "mean_token_accuracy": 0.10127009153366089,
      "num_tokens": 1011032.0,
      "step": 580
    },
    {
      "entropy": 7.2483752250671385,
      "epoch": 0.04551643649095507,
      "grad_norm": 1.140625,
      "learning_rate": 0.000292,
      "loss": 6.9354,
      "mean_token_accuracy": 0.09478853791952133,
      "num_tokens": 1019901.0,
      "step": 585
    },
    {
      "entropy": 7.367902326583862,
      "epoch": 0.045905465862672634,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002945,
      "loss": 6.9199,
      "mean_token_accuracy": 0.09609107375144958,
      "num_tokens": 1029049.0,
      "step": 590
    },
    {
      "entropy": 7.297397232055664,
      "epoch": 0.0462944952343902,
      "grad_norm": 1.03125,
      "learning_rate": 0.000297,
      "loss": 6.88,
      "mean_token_accuracy": 0.11009425967931748,
      "num_tokens": 1037580.0,
      "step": 595
    },
    {
      "entropy": 7.358423042297363,
      "epoch": 0.04668352460610776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002995,
      "loss": 6.9401,
      "mean_token_accuracy": 0.10255178958177566,
      "num_tokens": 1045804.0,
      "step": 600
    },
    {
      "entropy": 7.327436733245849,
      "epoch": 0.047072553977825324,
      "grad_norm": 1.15625,
      "learning_rate": 0.000302,
      "loss": 6.8943,
      "mean_token_accuracy": 0.0945185400545597,
      "num_tokens": 1054719.0,
      "step": 605
    },
    {
      "entropy": 7.400276851654053,
      "epoch": 0.04746158334954289,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003045,
      "loss": 6.8877,
      "mean_token_accuracy": 0.10351422131061554,
      "num_tokens": 1063373.0,
      "step": 610
    },
    {
      "entropy": 7.205434131622314,
      "epoch": 0.047850612721260455,
      "grad_norm": 1.078125,
      "learning_rate": 0.000307,
      "loss": 6.8601,
      "mean_token_accuracy": 0.1029354140162468,
      "num_tokens": 1073021.0,
      "step": 615
    },
    {
      "entropy": 7.366950464248657,
      "epoch": 0.04823964209297802,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003095,
      "loss": 6.9232,
      "mean_token_accuracy": 0.09619447886943817,
      "num_tokens": 1081547.0,
      "step": 620
    },
    {
      "entropy": 7.274023580551147,
      "epoch": 0.048628671464695586,
      "grad_norm": 1.0625,
      "learning_rate": 0.000312,
      "loss": 6.8907,
      "mean_token_accuracy": 0.09834453091025352,
      "num_tokens": 1090695.0,
      "step": 625
    },
    {
      "entropy": 7.250863170623779,
      "epoch": 0.04901770083641315,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003145,
      "loss": 6.7816,
      "mean_token_accuracy": 0.10651093125343322,
      "num_tokens": 1099563.0,
      "step": 630
    },
    {
      "entropy": 7.316005992889404,
      "epoch": 0.04940673020813072,
      "grad_norm": 1.09375,
      "learning_rate": 0.000317,
      "loss": 6.8534,
      "mean_token_accuracy": 0.1008830115199089,
      "num_tokens": 1109021.0,
      "step": 635
    },
    {
      "entropy": 7.2779614448547365,
      "epoch": 0.049795759579848276,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003195,
      "loss": 6.8518,
      "mean_token_accuracy": 0.10254788249731064,
      "num_tokens": 1118281.0,
      "step": 640
    },
    {
      "entropy": 7.271533203125,
      "epoch": 0.05018478895156584,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000322,
      "loss": 6.8932,
      "mean_token_accuracy": 0.09681643918156624,
      "num_tokens": 1127513.0,
      "step": 645
    },
    {
      "entropy": 7.177491044998169,
      "epoch": 0.05057381832328341,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.766,
      "mean_token_accuracy": 0.10509725958108902,
      "num_tokens": 1136140.0,
      "step": 650
    },
    {
      "entropy": 7.26359543800354,
      "epoch": 0.05096284769500097,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.7939,
      "mean_token_accuracy": 0.10301651507616043,
      "num_tokens": 1144297.0,
      "step": 655
    },
    {
      "entropy": 7.281031465530395,
      "epoch": 0.05135187706671854,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.8864,
      "mean_token_accuracy": 0.09698515310883522,
      "num_tokens": 1153639.0,
      "step": 660
    },
    {
      "entropy": 7.245992708206177,
      "epoch": 0.0517409064384361,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.8538,
      "mean_token_accuracy": 0.09941518083214759,
      "num_tokens": 1162825.0,
      "step": 665
    },
    {
      "entropy": 7.20550389289856,
      "epoch": 0.05212993581015367,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.8767,
      "mean_token_accuracy": 0.10294063836336136,
      "num_tokens": 1171887.0,
      "step": 670
    },
    {
      "entropy": 7.228694581985474,
      "epoch": 0.052518965181871234,
      "grad_norm": 1.171875,
      "learning_rate": 0.000337,
      "loss": 6.7878,
      "mean_token_accuracy": 0.10553507208824157,
      "num_tokens": 1179963.0,
      "step": 675
    },
    {
      "entropy": 7.170407724380493,
      "epoch": 0.05290799455358879,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003395,
      "loss": 6.7633,
      "mean_token_accuracy": 0.10891782119870186,
      "num_tokens": 1188408.0,
      "step": 680
    },
    {
      "entropy": 7.168772220611572,
      "epoch": 0.05329702392530636,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000342,
      "loss": 6.7986,
      "mean_token_accuracy": 0.09899375736713409,
      "num_tokens": 1197751.0,
      "step": 685
    },
    {
      "entropy": 7.197608995437622,
      "epoch": 0.053686053297023924,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.8312,
      "mean_token_accuracy": 0.1030447706580162,
      "num_tokens": 1207101.0,
      "step": 690
    },
    {
      "entropy": 7.218799304962158,
      "epoch": 0.05407508266874149,
      "grad_norm": 1.0625,
      "learning_rate": 0.000347,
      "loss": 6.829,
      "mean_token_accuracy": 0.10361301824450493,
      "num_tokens": 1216395.0,
      "step": 695
    },
    {
      "entropy": 7.224749946594239,
      "epoch": 0.054464112040459055,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003495,
      "loss": 6.8108,
      "mean_token_accuracy": 0.10502151176333427,
      "num_tokens": 1225496.0,
      "step": 700
    },
    {
      "entropy": 7.232848119735718,
      "epoch": 0.05485314141217662,
      "grad_norm": 1.0625,
      "learning_rate": 0.000352,
      "loss": 6.8746,
      "mean_token_accuracy": 0.10201601535081864,
      "num_tokens": 1233934.0,
      "step": 705
    },
    {
      "entropy": 7.18028507232666,
      "epoch": 0.055242170783894186,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003545,
      "loss": 6.7279,
      "mean_token_accuracy": 0.1090910330414772,
      "num_tokens": 1242165.0,
      "step": 710
    },
    {
      "entropy": 7.153624200820923,
      "epoch": 0.05563120015561175,
      "grad_norm": 1.296875,
      "learning_rate": 0.000357,
      "loss": 6.7967,
      "mean_token_accuracy": 0.1042649894952774,
      "num_tokens": 1250021.0,
      "step": 715
    },
    {
      "entropy": 7.190495872497559,
      "epoch": 0.05602022952732931,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003595,
      "loss": 6.8155,
      "mean_token_accuracy": 0.10334040820598603,
      "num_tokens": 1258397.0,
      "step": 720
    },
    {
      "entropy": 7.170076513290406,
      "epoch": 0.056409258899046875,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000362,
      "loss": 6.8121,
      "mean_token_accuracy": 0.10407566651701927,
      "num_tokens": 1266764.0,
      "step": 725
    },
    {
      "entropy": 7.1744321346282955,
      "epoch": 0.05679828827076444,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003645,
      "loss": 6.8094,
      "mean_token_accuracy": 0.09987704753875733,
      "num_tokens": 1275643.0,
      "step": 730
    },
    {
      "entropy": 7.19658842086792,
      "epoch": 0.057187317642482006,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000367,
      "loss": 6.7852,
      "mean_token_accuracy": 0.10181980207562447,
      "num_tokens": 1284379.0,
      "step": 735
    },
    {
      "entropy": 7.100433397293091,
      "epoch": 0.05757634701419957,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003695,
      "loss": 6.6451,
      "mean_token_accuracy": 0.11134169921278954,
      "num_tokens": 1293115.0,
      "step": 740
    },
    {
      "entropy": 7.243102264404297,
      "epoch": 0.05796537638591714,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000372,
      "loss": 6.8351,
      "mean_token_accuracy": 0.09663488790392875,
      "num_tokens": 1301583.0,
      "step": 745
    },
    {
      "entropy": 7.155255460739136,
      "epoch": 0.0583544057576347,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003745,
      "loss": 6.7361,
      "mean_token_accuracy": 0.10636070072650909,
      "num_tokens": 1310721.0,
      "step": 750
    },
    {
      "entropy": 7.158748054504395,
      "epoch": 0.05874343512935227,
      "grad_norm": 1.234375,
      "learning_rate": 0.000377,
      "loss": 6.8794,
      "mean_token_accuracy": 0.09966575875878333,
      "num_tokens": 1319056.0,
      "step": 755
    },
    {
      "entropy": 7.131784152984619,
      "epoch": 0.059132464501069834,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003795,
      "loss": 6.7662,
      "mean_token_accuracy": 0.1062756285071373,
      "num_tokens": 1327259.0,
      "step": 760
    },
    {
      "entropy": 7.098258781433105,
      "epoch": 0.05952149387278739,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000382,
      "loss": 6.6695,
      "mean_token_accuracy": 0.10549260452389717,
      "num_tokens": 1335906.0,
      "step": 765
    },
    {
      "entropy": 7.124299383163452,
      "epoch": 0.05991052324450496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003845,
      "loss": 6.7193,
      "mean_token_accuracy": 0.10388574972748757,
      "num_tokens": 1344743.0,
      "step": 770
    },
    {
      "entropy": 7.124751281738281,
      "epoch": 0.060299552616222524,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.732,
      "mean_token_accuracy": 0.10936428904533387,
      "num_tokens": 1353211.0,
      "step": 775
    },
    {
      "entropy": 7.119222116470337,
      "epoch": 0.06068858198794009,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.8463,
      "mean_token_accuracy": 0.09924818351864814,
      "num_tokens": 1361849.0,
      "step": 780
    },
    {
      "entropy": 7.19121618270874,
      "epoch": 0.061077611359657655,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.8264,
      "mean_token_accuracy": 0.10749334692955018,
      "num_tokens": 1369804.0,
      "step": 785
    },
    {
      "entropy": 7.045189571380615,
      "epoch": 0.06146664073137522,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.6156,
      "mean_token_accuracy": 0.10870056599378586,
      "num_tokens": 1378478.0,
      "step": 790
    },
    {
      "entropy": 7.11121244430542,
      "epoch": 0.061855670103092786,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.7107,
      "mean_token_accuracy": 0.10817317888140679,
      "num_tokens": 1387225.0,
      "step": 795
    },
    {
      "entropy": 7.035085678100586,
      "epoch": 0.06224469947481035,
      "grad_norm": 1.25,
      "learning_rate": 0.0003995,
      "loss": 6.6882,
      "mean_token_accuracy": 0.10929678976535798,
      "num_tokens": 1395338.0,
      "step": 800
    },
    {
      "entropy": 7.0744215965271,
      "epoch": 0.06263372884652792,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000402,
      "loss": 6.6625,
      "mean_token_accuracy": 0.10991387590765953,
      "num_tokens": 1403794.0,
      "step": 805
    },
    {
      "entropy": 6.982004165649414,
      "epoch": 0.06302275821824548,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004045,
      "loss": 6.738,
      "mean_token_accuracy": 0.10092620700597763,
      "num_tokens": 1412522.0,
      "step": 810
    },
    {
      "entropy": 7.081152534484863,
      "epoch": 0.06341178758996305,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.7227,
      "mean_token_accuracy": 0.10850374028086662,
      "num_tokens": 1421347.0,
      "step": 815
    },
    {
      "entropy": 7.046439456939697,
      "epoch": 0.0638008169616806,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004095,
      "loss": 6.7041,
      "mean_token_accuracy": 0.11018050685524941,
      "num_tokens": 1429788.0,
      "step": 820
    },
    {
      "entropy": 7.077642393112183,
      "epoch": 0.06418984633339817,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000412,
      "loss": 6.6955,
      "mean_token_accuracy": 0.1097219817340374,
      "num_tokens": 1437442.0,
      "step": 825
    },
    {
      "entropy": 6.984000587463379,
      "epoch": 0.06457887570511574,
      "grad_norm": 1.125,
      "learning_rate": 0.0004145,
      "loss": 6.6325,
      "mean_token_accuracy": 0.10700507164001465,
      "num_tokens": 1445604.0,
      "step": 830
    },
    {
      "entropy": 7.009225702285766,
      "epoch": 0.0649679050768333,
      "grad_norm": 1.140625,
      "learning_rate": 0.000417,
      "loss": 6.6963,
      "mean_token_accuracy": 0.10920493751764297,
      "num_tokens": 1454342.0,
      "step": 835
    },
    {
      "entropy": 7.046473693847656,
      "epoch": 0.06535693444855087,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004195,
      "loss": 6.6777,
      "mean_token_accuracy": 0.11271065473556519,
      "num_tokens": 1463189.0,
      "step": 840
    },
    {
      "entropy": 7.0349955558776855,
      "epoch": 0.06574596382026843,
      "grad_norm": 1.1875,
      "learning_rate": 0.000422,
      "loss": 6.751,
      "mean_token_accuracy": 0.1046958401799202,
      "num_tokens": 1471608.0,
      "step": 845
    },
    {
      "entropy": 7.108201360702514,
      "epoch": 0.066134993191986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004245,
      "loss": 6.7468,
      "mean_token_accuracy": 0.10504127219319344,
      "num_tokens": 1480452.0,
      "step": 850
    },
    {
      "entropy": 7.051074695587158,
      "epoch": 0.06652402256370356,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000427,
      "loss": 6.6688,
      "mean_token_accuracy": 0.10868829116225243,
      "num_tokens": 1489118.0,
      "step": 855
    },
    {
      "entropy": 6.962509250640869,
      "epoch": 0.06691305193542113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004295,
      "loss": 6.6703,
      "mean_token_accuracy": 0.10638387426733971,
      "num_tokens": 1498961.0,
      "step": 860
    },
    {
      "entropy": 7.081515598297119,
      "epoch": 0.06730208130713869,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000432,
      "loss": 6.6661,
      "mean_token_accuracy": 0.10934964269399643,
      "num_tokens": 1506772.0,
      "step": 865
    },
    {
      "entropy": 6.947929859161377,
      "epoch": 0.06769111067885625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004345,
      "loss": 6.5534,
      "mean_token_accuracy": 0.11734423041343689,
      "num_tokens": 1515095.0,
      "step": 870
    },
    {
      "entropy": 6.930423736572266,
      "epoch": 0.06808014005057382,
      "grad_norm": 1.078125,
      "learning_rate": 0.000437,
      "loss": 6.593,
      "mean_token_accuracy": 0.11688333824276924,
      "num_tokens": 1523477.0,
      "step": 875
    },
    {
      "entropy": 7.0028105735778805,
      "epoch": 0.06846916942229138,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004395,
      "loss": 6.6287,
      "mean_token_accuracy": 0.10662296116352081,
      "num_tokens": 1531786.0,
      "step": 880
    },
    {
      "entropy": 6.932631206512451,
      "epoch": 0.06885819879400895,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000442,
      "loss": 6.7189,
      "mean_token_accuracy": 0.10187280923128128,
      "num_tokens": 1540737.0,
      "step": 885
    },
    {
      "entropy": 6.950942516326904,
      "epoch": 0.06924722816572651,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004445,
      "loss": 6.5705,
      "mean_token_accuracy": 0.11004639267921448,
      "num_tokens": 1548923.0,
      "step": 890
    },
    {
      "entropy": 6.921659803390503,
      "epoch": 0.06963625753744408,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000447,
      "loss": 6.5273,
      "mean_token_accuracy": 0.11898597478866577,
      "num_tokens": 1557043.0,
      "step": 895
    },
    {
      "entropy": 6.893300342559814,
      "epoch": 0.07002528690916164,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.5837,
      "mean_token_accuracy": 0.11473687514662742,
      "num_tokens": 1564994.0,
      "step": 900
    },
    {
      "entropy": 6.921711587905884,
      "epoch": 0.0704143162808792,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5688,
      "mean_token_accuracy": 0.11160103157162667,
      "num_tokens": 1573859.0,
      "step": 905
    },
    {
      "entropy": 6.986482381820679,
      "epoch": 0.07080334565259677,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.5448,
      "mean_token_accuracy": 0.1087655171751976,
      "num_tokens": 1581947.0,
      "step": 910
    },
    {
      "entropy": 6.883696842193603,
      "epoch": 0.07119237502431433,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.5205,
      "mean_token_accuracy": 0.11201854199171066,
      "num_tokens": 1589971.0,
      "step": 915
    },
    {
      "entropy": 6.8449201583862305,
      "epoch": 0.0715814043960319,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4817,
      "mean_token_accuracy": 0.11529209017753601,
      "num_tokens": 1598612.0,
      "step": 920
    },
    {
      "entropy": 6.935947418212891,
      "epoch": 0.07197043376774946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000462,
      "loss": 6.621,
      "mean_token_accuracy": 0.10783407837152481,
      "num_tokens": 1607378.0,
      "step": 925
    },
    {
      "entropy": 6.844717597961425,
      "epoch": 0.07235946313946703,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004645,
      "loss": 6.617,
      "mean_token_accuracy": 0.1145533099770546,
      "num_tokens": 1615987.0,
      "step": 930
    },
    {
      "entropy": 6.833771276473999,
      "epoch": 0.07274849251118459,
      "grad_norm": 1.078125,
      "learning_rate": 0.000467,
      "loss": 6.4242,
      "mean_token_accuracy": 0.11947599574923515,
      "num_tokens": 1625366.0,
      "step": 935
    },
    {
      "entropy": 6.898650407791138,
      "epoch": 0.07313752188290217,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004695,
      "loss": 6.6015,
      "mean_token_accuracy": 0.11430608332157136,
      "num_tokens": 1633020.0,
      "step": 940
    },
    {
      "entropy": 6.870701551437378,
      "epoch": 0.07352655125461972,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000472,
      "loss": 6.5056,
      "mean_token_accuracy": 0.11246326416730881,
      "num_tokens": 1641518.0,
      "step": 945
    },
    {
      "entropy": 6.800650787353516,
      "epoch": 0.07391558062633728,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004745,
      "loss": 6.5851,
      "mean_token_accuracy": 0.10907077416777611,
      "num_tokens": 1650472.0,
      "step": 950
    },
    {
      "entropy": 6.919282150268555,
      "epoch": 0.07430460999805485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000477,
      "loss": 6.6162,
      "mean_token_accuracy": 0.10955818668007851,
      "num_tokens": 1659037.0,
      "step": 955
    },
    {
      "entropy": 6.930318880081177,
      "epoch": 0.07469363936977241,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004795,
      "loss": 6.6848,
      "mean_token_accuracy": 0.1120296597480774,
      "num_tokens": 1667897.0,
      "step": 960
    },
    {
      "entropy": 6.849212980270385,
      "epoch": 0.07508266874148999,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000482,
      "loss": 6.5291,
      "mean_token_accuracy": 0.1111655130982399,
      "num_tokens": 1676535.0,
      "step": 965
    },
    {
      "entropy": 6.944094228744507,
      "epoch": 0.07547169811320754,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004845,
      "loss": 6.5692,
      "mean_token_accuracy": 0.11485065668821334,
      "num_tokens": 1684415.0,
      "step": 970
    },
    {
      "entropy": 6.882363891601562,
      "epoch": 0.07586072748492512,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000487,
      "loss": 6.5195,
      "mean_token_accuracy": 0.1194272443652153,
      "num_tokens": 1692443.0,
      "step": 975
    },
    {
      "entropy": 6.744445562362671,
      "epoch": 0.07624975685664268,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895,
      "loss": 6.5166,
      "mean_token_accuracy": 0.11746397837996483,
      "num_tokens": 1700814.0,
      "step": 980
    },
    {
      "entropy": 6.767728662490844,
      "epoch": 0.07663878622836025,
      "grad_norm": 1.09375,
      "learning_rate": 0.000492,
      "loss": 6.3699,
      "mean_token_accuracy": 0.11762674301862716,
      "num_tokens": 1709591.0,
      "step": 985
    },
    {
      "entropy": 6.861118412017822,
      "epoch": 0.0770278156000778,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004945,
      "loss": 6.6015,
      "mean_token_accuracy": 0.10945382416248321,
      "num_tokens": 1718951.0,
      "step": 990
    },
    {
      "entropy": 6.884693145751953,
      "epoch": 0.07741684497179536,
      "grad_norm": 1.265625,
      "learning_rate": 0.000497,
      "loss": 6.566,
      "mean_token_accuracy": 0.11041911020874977,
      "num_tokens": 1727481.0,
      "step": 995
    },
    {
      "entropy": 6.846743249893189,
      "epoch": 0.07780587434351294,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995,
      "loss": 6.5609,
      "mean_token_accuracy": 0.11417475417256355,
      "num_tokens": 1736985.0,
      "step": 1000
    },
    {
      "entropy": 6.892654371261597,
      "epoch": 0.0781949037152305,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999999989075146,
      "loss": 6.6256,
      "mean_token_accuracy": 0.11123185381293296,
      "num_tokens": 1745470.0,
      "step": 1005
    },
    {
      "entropy": 6.851777601242065,
      "epoch": 0.07858393308694807,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999999944692927,
      "loss": 6.5357,
      "mean_token_accuracy": 0.10865053832530976,
      "num_tokens": 1754185.0,
      "step": 1010
    },
    {
      "entropy": 6.776595211029052,
      "epoch": 0.07897296245866563,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499999986617054,
      "loss": 6.546,
      "mean_token_accuracy": 0.11072391048073768,
      "num_tokens": 1762100.0,
      "step": 1015
    },
    {
      "entropy": 6.961122608184814,
      "epoch": 0.0793619918303832,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999999753507986,
      "loss": 6.5689,
      "mean_token_accuracy": 0.10964735001325607,
      "num_tokens": 1771070.0,
      "step": 1020
    },
    {
      "entropy": 6.746126365661621,
      "epoch": 0.07975102120210076,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999606705267,
      "loss": 6.5597,
      "mean_token_accuracy": 0.1127906545996666,
      "num_tokens": 1780085.0,
      "step": 1025
    },
    {
      "entropy": 6.842060327529907,
      "epoch": 0.08014005057381832,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999999425762385,
      "loss": 6.5404,
      "mean_token_accuracy": 0.11701353117823601,
      "num_tokens": 1789316.0,
      "step": 1030
    },
    {
      "entropy": 6.778835821151733,
      "epoch": 0.08052907994553589,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999999210679344,
      "loss": 6.475,
      "mean_token_accuracy": 0.11409812644124032,
      "num_tokens": 1797975.0,
      "step": 1035
    },
    {
      "entropy": 6.787038946151734,
      "epoch": 0.08091810931725345,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999998961456145,
      "loss": 6.5602,
      "mean_token_accuracy": 0.11720341742038727,
      "num_tokens": 1806129.0,
      "step": 1040
    },
    {
      "entropy": 6.956608533859253,
      "epoch": 0.08130713868897102,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999998678092794,
      "loss": 6.5564,
      "mean_token_accuracy": 0.1117042988538742,
      "num_tokens": 1813956.0,
      "step": 1045
    },
    {
      "entropy": 6.857865571975708,
      "epoch": 0.08169616806068858,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999998360589293,
      "loss": 6.5845,
      "mean_token_accuracy": 0.11212223619222642,
      "num_tokens": 1822572.0,
      "step": 1050
    },
    {
      "entropy": 6.7688240051269535,
      "epoch": 0.08208519743240615,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499999800894565,
      "loss": 6.4791,
      "mean_token_accuracy": 0.11256719455122947,
      "num_tokens": 1831870.0,
      "step": 1055
    },
    {
      "entropy": 6.68245153427124,
      "epoch": 0.08247422680412371,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999997623161866,
      "loss": 6.5135,
      "mean_token_accuracy": 0.1109384387731552,
      "num_tokens": 1840414.0,
      "step": 1060
    },
    {
      "entropy": 6.897469997406006,
      "epoch": 0.08286325617584128,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999997203237951,
      "loss": 6.5357,
      "mean_token_accuracy": 0.1112027607858181,
      "num_tokens": 1850385.0,
      "step": 1065
    },
    {
      "entropy": 6.6467287063598635,
      "epoch": 0.08325228554755884,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499999674917391,
      "loss": 6.4407,
      "mean_token_accuracy": 0.12009736895561218,
      "num_tokens": 1858479.0,
      "step": 1070
    },
    {
      "entropy": 6.7970844268798825,
      "epoch": 0.0836413149192764,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999996260969748,
      "loss": 6.5109,
      "mean_token_accuracy": 0.11019074022769929,
      "num_tokens": 1867357.0,
      "step": 1075
    },
    {
      "entropy": 6.754551267623901,
      "epoch": 0.08403034429099397,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999995738625475,
      "loss": 6.4736,
      "mean_token_accuracy": 0.11016227900981904,
      "num_tokens": 1876211.0,
      "step": 1080
    },
    {
      "entropy": 6.769411373138428,
      "epoch": 0.08441937366271153,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999995182141099,
      "loss": 6.3942,
      "mean_token_accuracy": 0.11268195807933808,
      "num_tokens": 1884301.0,
      "step": 1085
    },
    {
      "entropy": 6.7784522533416744,
      "epoch": 0.0848084030344291,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999994591516626,
      "loss": 6.5555,
      "mean_token_accuracy": 0.11373023912310601,
      "num_tokens": 1893191.0,
      "step": 1090
    },
    {
      "entropy": 6.673826599121094,
      "epoch": 0.08519743240614666,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999993966752066,
      "loss": 6.4283,
      "mean_token_accuracy": 0.11703478917479515,
      "num_tokens": 1901434.0,
      "step": 1095
    },
    {
      "entropy": 6.760957384109497,
      "epoch": 0.08558646177786423,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499999330784743,
      "loss": 6.4054,
      "mean_token_accuracy": 0.12261063754558563,
      "num_tokens": 1910054.0,
      "step": 1100
    },
    {
      "entropy": 6.764858293533325,
      "epoch": 0.08597549114958179,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999992614802725,
      "loss": 6.5472,
      "mean_token_accuracy": 0.11576480343937874,
      "num_tokens": 1918607.0,
      "step": 1105
    },
    {
      "entropy": 6.76983675956726,
      "epoch": 0.08636452052129936,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999991887617966,
      "loss": 6.4969,
      "mean_token_accuracy": 0.11678617969155311,
      "num_tokens": 1927610.0,
      "step": 1110
    },
    {
      "entropy": 6.654897880554199,
      "epoch": 0.08675354989301692,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999991126293158,
      "loss": 6.4455,
      "mean_token_accuracy": 0.11631680354475975,
      "num_tokens": 1936174.0,
      "step": 1115
    },
    {
      "entropy": 6.732142686843872,
      "epoch": 0.08714257926473448,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999990330828318,
      "loss": 6.3693,
      "mean_token_accuracy": 0.12286917939782142,
      "num_tokens": 1944228.0,
      "step": 1120
    },
    {
      "entropy": 6.709880685806274,
      "epoch": 0.08753160863645205,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999989501223456,
      "loss": 6.4821,
      "mean_token_accuracy": 0.11934061720967293,
      "num_tokens": 1953144.0,
      "step": 1125
    },
    {
      "entropy": 6.783202028274536,
      "epoch": 0.08792063800816961,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999988637478584,
      "loss": 6.5277,
      "mean_token_accuracy": 0.11476651057600976,
      "num_tokens": 1961947.0,
      "step": 1130
    },
    {
      "entropy": 6.779601001739502,
      "epoch": 0.08830966737988719,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999987739593716,
      "loss": 6.5028,
      "mean_token_accuracy": 0.1122349627315998,
      "num_tokens": 1970674.0,
      "step": 1135
    },
    {
      "entropy": 6.816339492797852,
      "epoch": 0.08869869675160474,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999986807568865,
      "loss": 6.4793,
      "mean_token_accuracy": 0.11963360384106636,
      "num_tokens": 1979173.0,
      "step": 1140
    },
    {
      "entropy": 6.641682291030884,
      "epoch": 0.08908772612332232,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999985841404045,
      "loss": 6.4536,
      "mean_token_accuracy": 0.11661232337355613,
      "num_tokens": 1988134.0,
      "step": 1145
    },
    {
      "entropy": 6.796137905120849,
      "epoch": 0.08947675549503988,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999984841099271,
      "loss": 6.4027,
      "mean_token_accuracy": 0.11723847314715385,
      "num_tokens": 1996977.0,
      "step": 1150
    },
    {
      "entropy": 6.740950250625611,
      "epoch": 0.08986578486675743,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499998380665456,
      "loss": 6.4805,
      "mean_token_accuracy": 0.11753954589366913,
      "num_tokens": 2005946.0,
      "step": 1155
    },
    {
      "entropy": 6.676974439620972,
      "epoch": 0.090254814238475,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999982738069924,
      "loss": 6.2987,
      "mean_token_accuracy": 0.12011792436242104,
      "num_tokens": 2014882.0,
      "step": 1160
    },
    {
      "entropy": 6.609836864471435,
      "epoch": 0.09064384361019256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999981635345382,
      "loss": 6.4888,
      "mean_token_accuracy": 0.11186635941267013,
      "num_tokens": 2024858.0,
      "step": 1165
    },
    {
      "entropy": 6.676596641540527,
      "epoch": 0.09103287298191014,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999980498480949,
      "loss": 6.4003,
      "mean_token_accuracy": 0.11673102378845215,
      "num_tokens": 2034072.0,
      "step": 1170
    },
    {
      "entropy": 6.680279922485352,
      "epoch": 0.0914219023536277,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999979327476644,
      "loss": 6.4054,
      "mean_token_accuracy": 0.11940634250640869,
      "num_tokens": 2042649.0,
      "step": 1175
    },
    {
      "entropy": 6.699150514602661,
      "epoch": 0.09181093172534527,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999978122332485,
      "loss": 6.4254,
      "mean_token_accuracy": 0.12093920707702636,
      "num_tokens": 2050866.0,
      "step": 1180
    },
    {
      "entropy": 6.7128864288330075,
      "epoch": 0.09219996109706283,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999976883048487,
      "loss": 6.4421,
      "mean_token_accuracy": 0.11782675310969352,
      "num_tokens": 2058914.0,
      "step": 1185
    },
    {
      "entropy": 6.6109076023101805,
      "epoch": 0.0925889904687804,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999975609624673,
      "loss": 6.3777,
      "mean_token_accuracy": 0.11382007896900177,
      "num_tokens": 2068098.0,
      "step": 1190
    },
    {
      "entropy": 6.656971406936646,
      "epoch": 0.09297801984049796,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999974302061059,
      "loss": 6.3773,
      "mean_token_accuracy": 0.11357153728604316,
      "num_tokens": 2077299.0,
      "step": 1195
    },
    {
      "entropy": 6.678202056884766,
      "epoch": 0.09336704921221552,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999972960357666,
      "loss": 6.4757,
      "mean_token_accuracy": 0.11248122900724411,
      "num_tokens": 2087868.0,
      "step": 1200
    },
    {
      "entropy": 6.6500016212463375,
      "epoch": 0.09375607858393309,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999971584514516,
      "loss": 6.3967,
      "mean_token_accuracy": 0.12456627488136292,
      "num_tokens": 2096553.0,
      "step": 1205
    },
    {
      "entropy": 6.7588951110839846,
      "epoch": 0.09414510795565065,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999970174531627,
      "loss": 6.3727,
      "mean_token_accuracy": 0.11930682808160782,
      "num_tokens": 2104809.0,
      "step": 1210
    },
    {
      "entropy": 6.5227766036987305,
      "epoch": 0.09453413732736822,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999968730409022,
      "loss": 6.3513,
      "mean_token_accuracy": 0.11753158569335938,
      "num_tokens": 2113869.0,
      "step": 1215
    },
    {
      "entropy": 6.59004921913147,
      "epoch": 0.09492316669908578,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999967252146723,
      "loss": 6.1845,
      "mean_token_accuracy": 0.11563206911087036,
      "num_tokens": 2121749.0,
      "step": 1220
    },
    {
      "entropy": 6.602604007720947,
      "epoch": 0.09531219607080335,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999965739744752,
      "loss": 6.3919,
      "mean_token_accuracy": 0.11459906175732612,
      "num_tokens": 2130828.0,
      "step": 1225
    },
    {
      "entropy": 6.719199275970459,
      "epoch": 0.09570122544252091,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999964193203132,
      "loss": 6.3664,
      "mean_token_accuracy": 0.12008291706442834,
      "num_tokens": 2138928.0,
      "step": 1230
    },
    {
      "entropy": 6.622300052642823,
      "epoch": 0.09609025481423847,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999962612521888,
      "loss": 6.4103,
      "mean_token_accuracy": 0.11864980831742286,
      "num_tokens": 2147643.0,
      "step": 1235
    },
    {
      "entropy": 6.690579223632812,
      "epoch": 0.09647928418595604,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999960997701041,
      "loss": 6.3286,
      "mean_token_accuracy": 0.12595243826508523,
      "num_tokens": 2156766.0,
      "step": 1240
    },
    {
      "entropy": 6.594869613647461,
      "epoch": 0.0968683135576736,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999959348740618,
      "loss": 6.3937,
      "mean_token_accuracy": 0.11976057812571525,
      "num_tokens": 2165517.0,
      "step": 1245
    },
    {
      "entropy": 6.534668731689453,
      "epoch": 0.09725734292939117,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999957665640641,
      "loss": 6.2614,
      "mean_token_accuracy": 0.12245074957609177,
      "num_tokens": 2174812.0,
      "step": 1250
    },
    {
      "entropy": 6.62302975654602,
      "epoch": 0.09764637230110873,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499995594840114,
      "loss": 6.3965,
      "mean_token_accuracy": 0.11696946769952773,
      "num_tokens": 2183968.0,
      "step": 1255
    },
    {
      "entropy": 6.6896251201629635,
      "epoch": 0.0980354016728263,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999954197022138,
      "loss": 6.4257,
      "mean_token_accuracy": 0.12401301488280296,
      "num_tokens": 2191947.0,
      "step": 1260
    },
    {
      "entropy": 6.606826972961426,
      "epoch": 0.09842443104454386,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999952411503661,
      "loss": 6.271,
      "mean_token_accuracy": 0.12751914858818053,
      "num_tokens": 2199951.0,
      "step": 1265
    },
    {
      "entropy": 6.564306163787842,
      "epoch": 0.09881346041626143,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999950591845738,
      "loss": 6.4197,
      "mean_token_accuracy": 0.12145831063389778,
      "num_tokens": 2208714.0,
      "step": 1270
    },
    {
      "entropy": 6.571636629104614,
      "epoch": 0.09920248978797899,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999948738048397,
      "loss": 6.2797,
      "mean_token_accuracy": 0.12155360877513885,
      "num_tokens": 2217250.0,
      "step": 1275
    },
    {
      "entropy": 6.634035968780518,
      "epoch": 0.09959151915969655,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999946850111663,
      "loss": 6.3297,
      "mean_token_accuracy": 0.12339223623275757,
      "num_tokens": 2225468.0,
      "step": 1280
    },
    {
      "entropy": 6.528422164916992,
      "epoch": 0.09998054853141412,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999944928035569,
      "loss": 6.2904,
      "mean_token_accuracy": 0.12815140709280967,
      "num_tokens": 2233893.0,
      "step": 1285
    },
    {
      "entropy": 6.6666259765625,
      "epoch": 0.10036957790313168,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499994297182014,
      "loss": 6.3539,
      "mean_token_accuracy": 0.12206242978572845,
      "num_tokens": 2242487.0,
      "step": 1290
    },
    {
      "entropy": 6.534954357147217,
      "epoch": 0.10075860727484925,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999940981465408,
      "loss": 6.1874,
      "mean_token_accuracy": 0.12534248530864717,
      "num_tokens": 2250518.0,
      "step": 1295
    },
    {
      "entropy": 6.555987071990967,
      "epoch": 0.10114763664656681,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999938956971404,
      "loss": 6.3579,
      "mean_token_accuracy": 0.11568900793790818,
      "num_tokens": 2259024.0,
      "step": 1300
    },
    {
      "entropy": 6.613235092163086,
      "epoch": 0.10153666601828439,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999936898338156,
      "loss": 6.4011,
      "mean_token_accuracy": 0.11064468100666999,
      "num_tokens": 2267832.0,
      "step": 1305
    },
    {
      "entropy": 6.580302953720093,
      "epoch": 0.10192569539000194,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999934805565698,
      "loss": 6.4644,
      "mean_token_accuracy": 0.11155277863144875,
      "num_tokens": 2276834.0,
      "step": 1310
    },
    {
      "entropy": 6.698448944091797,
      "epoch": 0.10231472476171952,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499993267865406,
      "loss": 6.3886,
      "mean_token_accuracy": 0.11738630458712578,
      "num_tokens": 2285736.0,
      "step": 1315
    },
    {
      "entropy": 6.663760614395142,
      "epoch": 0.10270375413343708,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999930517603275,
      "loss": 6.3469,
      "mean_token_accuracy": 0.12696315422654153,
      "num_tokens": 2294177.0,
      "step": 1320
    },
    {
      "entropy": 6.5038152694702145,
      "epoch": 0.10309278350515463,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999928322413374,
      "loss": 6.1338,
      "mean_token_accuracy": 0.13216883465647697,
      "num_tokens": 2302784.0,
      "step": 1325
    },
    {
      "entropy": 6.468310213088989,
      "epoch": 0.1034818128768722,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999926093084394,
      "loss": 6.2448,
      "mean_token_accuracy": 0.12271884679794312,
      "num_tokens": 2311650.0,
      "step": 1330
    },
    {
      "entropy": 6.5902739524841305,
      "epoch": 0.10387084224858976,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999923829616365,
      "loss": 6.3573,
      "mean_token_accuracy": 0.12029535993933678,
      "num_tokens": 2319828.0,
      "step": 1335
    },
    {
      "entropy": 6.606714200973511,
      "epoch": 0.10425987162030734,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999921532009323,
      "loss": 6.3635,
      "mean_token_accuracy": 0.1170631617307663,
      "num_tokens": 2328636.0,
      "step": 1340
    },
    {
      "entropy": 6.487597703933716,
      "epoch": 0.1046489009920249,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999919200263304,
      "loss": 6.2424,
      "mean_token_accuracy": 0.12811223715543746,
      "num_tokens": 2336771.0,
      "step": 1345
    },
    {
      "entropy": 6.611960554122925,
      "epoch": 0.10503793036374247,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999916834378342,
      "loss": 6.388,
      "mean_token_accuracy": 0.11745835095643997,
      "num_tokens": 2346106.0,
      "step": 1350
    },
    {
      "entropy": 6.6681524276733395,
      "epoch": 0.10542695973546003,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999914434354472,
      "loss": 6.3915,
      "mean_token_accuracy": 0.11755762621760368,
      "num_tokens": 2354943.0,
      "step": 1355
    },
    {
      "entropy": 6.3937695026397705,
      "epoch": 0.10581598910717759,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999912000191733,
      "loss": 6.2812,
      "mean_token_accuracy": 0.12295414507389069,
      "num_tokens": 2363699.0,
      "step": 1360
    },
    {
      "entropy": 6.576402616500855,
      "epoch": 0.10620501847889516,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499990953189016,
      "loss": 6.4118,
      "mean_token_accuracy": 0.11689253002405167,
      "num_tokens": 2373316.0,
      "step": 1365
    },
    {
      "entropy": 6.740978050231933,
      "epoch": 0.10659404785061272,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999907029449791,
      "loss": 6.4907,
      "mean_token_accuracy": 0.11326492577791214,
      "num_tokens": 2381550.0,
      "step": 1370
    },
    {
      "entropy": 6.545381116867065,
      "epoch": 0.10698307722233029,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999904492870663,
      "loss": 6.3081,
      "mean_token_accuracy": 0.12226687595248223,
      "num_tokens": 2389546.0,
      "step": 1375
    },
    {
      "entropy": 6.559665966033935,
      "epoch": 0.10737210659404785,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999901922152816,
      "loss": 6.2522,
      "mean_token_accuracy": 0.13022437021136285,
      "num_tokens": 2398852.0,
      "step": 1380
    },
    {
      "entropy": 6.537785530090332,
      "epoch": 0.10776113596576542,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999899317296288,
      "loss": 6.341,
      "mean_token_accuracy": 0.11776238903403283,
      "num_tokens": 2407815.0,
      "step": 1385
    },
    {
      "entropy": 6.536812734603882,
      "epoch": 0.10815016533748298,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999896678301121,
      "loss": 6.2117,
      "mean_token_accuracy": 0.12601763755083084,
      "num_tokens": 2417300.0,
      "step": 1390
    },
    {
      "entropy": 6.544566440582275,
      "epoch": 0.10853919470920055,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999894005167352,
      "loss": 6.3687,
      "mean_token_accuracy": 0.12166865617036819,
      "num_tokens": 2426257.0,
      "step": 1395
    },
    {
      "entropy": 6.673711729049683,
      "epoch": 0.10892822408091811,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499989129789502,
      "loss": 6.3822,
      "mean_token_accuracy": 0.12371926084160804,
      "num_tokens": 2434595.0,
      "step": 1400
    },
    {
      "entropy": 6.637689447402954,
      "epoch": 0.10931725345263567,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999888556484172,
      "loss": 6.3857,
      "mean_token_accuracy": 0.12067539989948273,
      "num_tokens": 2443945.0,
      "step": 1405
    },
    {
      "entropy": 6.449913835525512,
      "epoch": 0.10970628282435324,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999885780934845,
      "loss": 6.2578,
      "mean_token_accuracy": 0.12468339428305626,
      "num_tokens": 2453141.0,
      "step": 1410
    },
    {
      "entropy": 6.639901494979858,
      "epoch": 0.1100953121960708,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999882971247081,
      "loss": 6.3439,
      "mean_token_accuracy": 0.11852156817913055,
      "num_tokens": 2461873.0,
      "step": 1415
    },
    {
      "entropy": 6.518063354492187,
      "epoch": 0.11048434156778837,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999880127420926,
      "loss": 6.361,
      "mean_token_accuracy": 0.1180450290441513,
      "num_tokens": 2471238.0,
      "step": 1420
    },
    {
      "entropy": 6.570301914215088,
      "epoch": 0.11087337093950593,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999877249456421,
      "loss": 6.2637,
      "mean_token_accuracy": 0.12611638382077217,
      "num_tokens": 2479412.0,
      "step": 1425
    },
    {
      "entropy": 6.350358390808106,
      "epoch": 0.1112624003112235,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999874337353609,
      "loss": 6.1379,
      "mean_token_accuracy": 0.1284614808857441,
      "num_tokens": 2488150.0,
      "step": 1430
    },
    {
      "entropy": 6.525825834274292,
      "epoch": 0.11165142968294106,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999871391112535,
      "loss": 6.3105,
      "mean_token_accuracy": 0.12847553417086602,
      "num_tokens": 2498103.0,
      "step": 1435
    },
    {
      "entropy": 6.43725700378418,
      "epoch": 0.11204045905465862,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999868410733244,
      "loss": 6.1815,
      "mean_token_accuracy": 0.12891680896282195,
      "num_tokens": 2506211.0,
      "step": 1440
    },
    {
      "entropy": 6.403781414031982,
      "epoch": 0.11242948842637619,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499986539621578,
      "loss": 6.2016,
      "mean_token_accuracy": 0.12844212278723716,
      "num_tokens": 2514743.0,
      "step": 1445
    },
    {
      "entropy": 6.451426029205322,
      "epoch": 0.11281851779809375,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999862347560191,
      "loss": 6.2131,
      "mean_token_accuracy": 0.11876050010323524,
      "num_tokens": 2523824.0,
      "step": 1450
    },
    {
      "entropy": 6.63544545173645,
      "epoch": 0.11320754716981132,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999859264766521,
      "loss": 6.2911,
      "mean_token_accuracy": 0.11931056976318359,
      "num_tokens": 2532360.0,
      "step": 1455
    },
    {
      "entropy": 6.430943727493286,
      "epoch": 0.11359657654152888,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999856147834817,
      "loss": 6.2346,
      "mean_token_accuracy": 0.12711054608225822,
      "num_tokens": 2540547.0,
      "step": 1460
    },
    {
      "entropy": 6.486865091323852,
      "epoch": 0.11398560591324645,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999852996765129,
      "loss": 6.2473,
      "mean_token_accuracy": 0.12674545124173164,
      "num_tokens": 2549193.0,
      "step": 1465
    },
    {
      "entropy": 6.464196062088012,
      "epoch": 0.11437463528496401,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999849811557502,
      "loss": 6.2622,
      "mean_token_accuracy": 0.12100255340337754,
      "num_tokens": 2557553.0,
      "step": 1470
    },
    {
      "entropy": 6.629219913482666,
      "epoch": 0.11476366465668159,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999846592211984,
      "loss": 6.3607,
      "mean_token_accuracy": 0.12177548781037331,
      "num_tokens": 2565945.0,
      "step": 1475
    },
    {
      "entropy": 6.536821985244751,
      "epoch": 0.11515269402839914,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999843338728626,
      "loss": 6.2246,
      "mean_token_accuracy": 0.12030538022518159,
      "num_tokens": 2574364.0,
      "step": 1480
    },
    {
      "entropy": 6.438568162918091,
      "epoch": 0.1155417234001167,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999840051107476,
      "loss": 6.1639,
      "mean_token_accuracy": 0.12257452458143234,
      "num_tokens": 2582825.0,
      "step": 1485
    },
    {
      "entropy": 6.53958911895752,
      "epoch": 0.11593075277183428,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999836729348584,
      "loss": 6.2992,
      "mean_token_accuracy": 0.12617669850587845,
      "num_tokens": 2591628.0,
      "step": 1490
    },
    {
      "entropy": 6.568798446655274,
      "epoch": 0.11631978214355183,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999833373452,
      "loss": 6.3072,
      "mean_token_accuracy": 0.11595111265778542,
      "num_tokens": 2600436.0,
      "step": 1495
    },
    {
      "entropy": 6.432466316223144,
      "epoch": 0.1167088115152694,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999829983417778,
      "loss": 6.2444,
      "mean_token_accuracy": 0.1281425453722477,
      "num_tokens": 2609709.0,
      "step": 1500
    },
    {
      "entropy": 6.624283027648926,
      "epoch": 0.11709784088698696,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999826559245965,
      "loss": 6.2546,
      "mean_token_accuracy": 0.12523326128721238,
      "num_tokens": 2617992.0,
      "step": 1505
    },
    {
      "entropy": 6.3710126876831055,
      "epoch": 0.11748687025870454,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999823100936615,
      "loss": 6.1762,
      "mean_token_accuracy": 0.12328904047608376,
      "num_tokens": 2626706.0,
      "step": 1510
    },
    {
      "entropy": 6.600048542022705,
      "epoch": 0.1178758996304221,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999819608489781,
      "loss": 6.2453,
      "mean_token_accuracy": 0.12911369353532792,
      "num_tokens": 2634642.0,
      "step": 1515
    },
    {
      "entropy": 6.53199143409729,
      "epoch": 0.11826492900213967,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999816081905515,
      "loss": 6.2789,
      "mean_token_accuracy": 0.11790527626872063,
      "num_tokens": 2644160.0,
      "step": 1520
    },
    {
      "entropy": 6.3495841979980465,
      "epoch": 0.11865395837385723,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999812521183872,
      "loss": 6.2528,
      "mean_token_accuracy": 0.11885925233364106,
      "num_tokens": 2653390.0,
      "step": 1525
    },
    {
      "entropy": 6.5672304153442385,
      "epoch": 0.11904298774557479,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999808926324903,
      "loss": 6.2957,
      "mean_token_accuracy": 0.12170536145567894,
      "num_tokens": 2662561.0,
      "step": 1530
    },
    {
      "entropy": 6.494124794006348,
      "epoch": 0.11943201711729236,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999805297328665,
      "loss": 6.2467,
      "mean_token_accuracy": 0.12452413216233253,
      "num_tokens": 2670694.0,
      "step": 1535
    },
    {
      "entropy": 6.401147556304932,
      "epoch": 0.11982104648900992,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999801634195214,
      "loss": 6.1463,
      "mean_token_accuracy": 0.12829762995243071,
      "num_tokens": 2679583.0,
      "step": 1540
    },
    {
      "entropy": 6.418363094329834,
      "epoch": 0.12021007586072749,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999797936924603,
      "loss": 6.1377,
      "mean_token_accuracy": 0.13158272877335547,
      "num_tokens": 2688429.0,
      "step": 1545
    },
    {
      "entropy": 6.529831981658935,
      "epoch": 0.12059910523244505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499979420551689,
      "loss": 6.2132,
      "mean_token_accuracy": 0.12612780630588533,
      "num_tokens": 2697134.0,
      "step": 1550
    },
    {
      "entropy": 6.446161222457886,
      "epoch": 0.12098813460416262,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499979043997213,
      "loss": 6.3499,
      "mean_token_accuracy": 0.12430260628461838,
      "num_tokens": 2705641.0,
      "step": 1555
    },
    {
      "entropy": 6.519960689544678,
      "epoch": 0.12137716397588018,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999786640290381,
      "loss": 6.2822,
      "mean_token_accuracy": 0.13202693238854407,
      "num_tokens": 2714977.0,
      "step": 1560
    },
    {
      "entropy": 6.495924043655395,
      "epoch": 0.12176619334759774,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999782806471701,
      "loss": 6.1374,
      "mean_token_accuracy": 0.13872739523649216,
      "num_tokens": 2723120.0,
      "step": 1565
    },
    {
      "entropy": 6.295685195922852,
      "epoch": 0.12215522271931531,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999778938516147,
      "loss": 6.1402,
      "mean_token_accuracy": 0.13263760954141618,
      "num_tokens": 2731964.0,
      "step": 1570
    },
    {
      "entropy": 6.6035463333129885,
      "epoch": 0.12254425209103287,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499977503642378,
      "loss": 6.1997,
      "mean_token_accuracy": 0.12307174578309059,
      "num_tokens": 2741537.0,
      "step": 1575
    },
    {
      "entropy": 6.387080717086792,
      "epoch": 0.12293328146275044,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999771100194656,
      "loss": 6.313,
      "mean_token_accuracy": 0.12455416470766068,
      "num_tokens": 2751522.0,
      "step": 1580
    },
    {
      "entropy": 6.577677011489868,
      "epoch": 0.123322310834468,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999767129828837,
      "loss": 6.3166,
      "mean_token_accuracy": 0.11798776909708977,
      "num_tokens": 2761100.0,
      "step": 1585
    },
    {
      "entropy": 6.441084480285644,
      "epoch": 0.12371134020618557,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999763125326383,
      "loss": 6.2427,
      "mean_token_accuracy": 0.12579894289374352,
      "num_tokens": 2769171.0,
      "step": 1590
    },
    {
      "entropy": 6.440652084350586,
      "epoch": 0.12410036957790313,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999759086687355,
      "loss": 6.1895,
      "mean_token_accuracy": 0.12780183404684067,
      "num_tokens": 2778021.0,
      "step": 1595
    },
    {
      "entropy": 6.3568809032440186,
      "epoch": 0.1244893989496207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999755013911813,
      "loss": 6.1279,
      "mean_token_accuracy": 0.12737615406513214,
      "num_tokens": 2787206.0,
      "step": 1600
    },
    {
      "entropy": 6.398793506622314,
      "epoch": 0.12487842832133826,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499975090699982,
      "loss": 6.1907,
      "mean_token_accuracy": 0.12401381433010102,
      "num_tokens": 2796051.0,
      "step": 1605
    },
    {
      "entropy": 6.447634220123291,
      "epoch": 0.12526745769305583,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999746765951438,
      "loss": 6.1438,
      "mean_token_accuracy": 0.12876681834459305,
      "num_tokens": 2804961.0,
      "step": 1610
    },
    {
      "entropy": 6.419821071624756,
      "epoch": 0.1256564870647734,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999742590766729,
      "loss": 6.1004,
      "mean_token_accuracy": 0.1357290342450142,
      "num_tokens": 2813691.0,
      "step": 1615
    },
    {
      "entropy": 6.391876268386841,
      "epoch": 0.12604551643649095,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999738381445757,
      "loss": 6.2317,
      "mean_token_accuracy": 0.12731038108468057,
      "num_tokens": 2822248.0,
      "step": 1620
    },
    {
      "entropy": 6.477704286575317,
      "epoch": 0.1264345458082085,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999734137988586,
      "loss": 6.1942,
      "mean_token_accuracy": 0.12995147630572318,
      "num_tokens": 2830826.0,
      "step": 1625
    },
    {
      "entropy": 6.421813488006592,
      "epoch": 0.1268235751799261,
      "grad_norm": 1.125,
      "learning_rate": 0.000499972986039528,
      "loss": 6.1865,
      "mean_token_accuracy": 0.12956878170371056,
      "num_tokens": 2839766.0,
      "step": 1630
    },
    {
      "entropy": 6.311213779449463,
      "epoch": 0.12721260455164365,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999725548665904,
      "loss": 6.1189,
      "mean_token_accuracy": 0.1319350078701973,
      "num_tokens": 2848728.0,
      "step": 1635
    },
    {
      "entropy": 6.348865556716919,
      "epoch": 0.1276016339233612,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999721202800524,
      "loss": 6.1481,
      "mean_token_accuracy": 0.12775396332144737,
      "num_tokens": 2857221.0,
      "step": 1640
    },
    {
      "entropy": 6.470174407958984,
      "epoch": 0.12799066329507877,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999716822799206,
      "loss": 6.2085,
      "mean_token_accuracy": 0.13055417835712432,
      "num_tokens": 2865314.0,
      "step": 1645
    },
    {
      "entropy": 6.300682735443115,
      "epoch": 0.12837969266679633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999712408662016,
      "loss": 6.0805,
      "mean_token_accuracy": 0.13145830407738684,
      "num_tokens": 2873654.0,
      "step": 1650
    },
    {
      "entropy": 6.335351085662841,
      "epoch": 0.12876872203851392,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999707960389021,
      "loss": 5.9957,
      "mean_token_accuracy": 0.13766974061727524,
      "num_tokens": 2881604.0,
      "step": 1655
    },
    {
      "entropy": 6.3854491233825685,
      "epoch": 0.12915775141023147,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999703477980288,
      "loss": 6.1897,
      "mean_token_accuracy": 0.12948238775134085,
      "num_tokens": 2890427.0,
      "step": 1660
    },
    {
      "entropy": 6.4016913890838625,
      "epoch": 0.12954678078194903,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999698961435885,
      "loss": 6.1487,
      "mean_token_accuracy": 0.12871672213077545,
      "num_tokens": 2899206.0,
      "step": 1665
    },
    {
      "entropy": 6.379415559768677,
      "epoch": 0.1299358101536666,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999694410755882,
      "loss": 6.0845,
      "mean_token_accuracy": 0.13119084686040877,
      "num_tokens": 2907738.0,
      "step": 1670
    },
    {
      "entropy": 6.384844398498535,
      "epoch": 0.13032483952538418,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999689825940347,
      "loss": 6.2902,
      "mean_token_accuracy": 0.12188465967774391,
      "num_tokens": 2916881.0,
      "step": 1675
    },
    {
      "entropy": 6.461624526977539,
      "epoch": 0.13071386889710174,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999685206989349,
      "loss": 6.2395,
      "mean_token_accuracy": 0.1223972775042057,
      "num_tokens": 2926467.0,
      "step": 1680
    },
    {
      "entropy": 6.4319689750671385,
      "epoch": 0.1311028982688193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999680553902959,
      "loss": 6.1786,
      "mean_token_accuracy": 0.12664306312799453,
      "num_tokens": 2934671.0,
      "step": 1685
    },
    {
      "entropy": 6.371197986602783,
      "epoch": 0.13149192764053685,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999675866681247,
      "loss": 6.2,
      "mean_token_accuracy": 0.12353175431489945,
      "num_tokens": 2943925.0,
      "step": 1690
    },
    {
      "entropy": 6.441555547714233,
      "epoch": 0.1318809570122544,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999671145324286,
      "loss": 6.2338,
      "mean_token_accuracy": 0.12637216970324516,
      "num_tokens": 2952118.0,
      "step": 1695
    },
    {
      "entropy": 6.3267382144927975,
      "epoch": 0.132269986383972,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999666389832144,
      "loss": 6.2113,
      "mean_token_accuracy": 0.12672000229358674,
      "num_tokens": 2961110.0,
      "step": 1700
    },
    {
      "entropy": 6.460357189178467,
      "epoch": 0.13265901575568956,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999661600204897,
      "loss": 6.1454,
      "mean_token_accuracy": 0.13502151966094972,
      "num_tokens": 2968877.0,
      "step": 1705
    },
    {
      "entropy": 6.44069094657898,
      "epoch": 0.13304804512740712,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999656776442615,
      "loss": 6.1976,
      "mean_token_accuracy": 0.13004121631383897,
      "num_tokens": 2977346.0,
      "step": 1710
    },
    {
      "entropy": 6.449344873428345,
      "epoch": 0.13343707449912467,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999651918545372,
      "loss": 6.2464,
      "mean_token_accuracy": 0.12166893184185028,
      "num_tokens": 2987494.0,
      "step": 1715
    },
    {
      "entropy": 6.389680957794189,
      "epoch": 0.13382610387084226,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999647026513243,
      "loss": 6.1548,
      "mean_token_accuracy": 0.12699257507920264,
      "num_tokens": 2996728.0,
      "step": 1720
    },
    {
      "entropy": 6.319113826751709,
      "epoch": 0.13421513324255982,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049996421003463,
      "loss": 6.1861,
      "mean_token_accuracy": 0.12468351423740387,
      "num_tokens": 3005810.0,
      "step": 1725
    },
    {
      "entropy": 6.385270690917968,
      "epoch": 0.13460416261427738,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999637140044619,
      "loss": 6.2053,
      "mean_token_accuracy": 0.12655822560191154,
      "num_tokens": 3014603.0,
      "step": 1730
    },
    {
      "entropy": 6.544794750213623,
      "epoch": 0.13499319198599494,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999632145608275,
      "loss": 6.1965,
      "mean_token_accuracy": 0.13691407367587088,
      "num_tokens": 3023170.0,
      "step": 1735
    },
    {
      "entropy": 6.259229373931885,
      "epoch": 0.1353822213577125,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999627117037343,
      "loss": 5.9459,
      "mean_token_accuracy": 0.13729629442095756,
      "num_tokens": 3030565.0,
      "step": 1740
    },
    {
      "entropy": 6.357711887359619,
      "epoch": 0.13577125072943008,
      "grad_norm": 1.0,
      "learning_rate": 0.00049996220543319,
      "loss": 6.2181,
      "mean_token_accuracy": 0.1307998225092888,
      "num_tokens": 3040124.0,
      "step": 1745
    },
    {
      "entropy": 6.373897314071655,
      "epoch": 0.13616028010114764,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999616957492025,
      "loss": 6.0817,
      "mean_token_accuracy": 0.13463346734642984,
      "num_tokens": 3048458.0,
      "step": 1750
    },
    {
      "entropy": 6.304511833190918,
      "epoch": 0.1365493094728652,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999611826517793,
      "loss": 6.0558,
      "mean_token_accuracy": 0.12701260522007943,
      "num_tokens": 3057926.0,
      "step": 1755
    },
    {
      "entropy": 6.371023035049438,
      "epoch": 0.13693833884458276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999606661409281,
      "loss": 6.0875,
      "mean_token_accuracy": 0.12731098383665085,
      "num_tokens": 3066683.0,
      "step": 1760
    },
    {
      "entropy": 6.228750848770142,
      "epoch": 0.13732736821630034,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999601462166569,
      "loss": 6.1605,
      "mean_token_accuracy": 0.12849701195955276,
      "num_tokens": 3075117.0,
      "step": 1765
    },
    {
      "entropy": 6.453361701965332,
      "epoch": 0.1377163975880179,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999596228789736,
      "loss": 6.2112,
      "mean_token_accuracy": 0.12286571562290191,
      "num_tokens": 3084399.0,
      "step": 1770
    },
    {
      "entropy": 6.295282793045044,
      "epoch": 0.13810542695973546,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999590961278859,
      "loss": 6.0036,
      "mean_token_accuracy": 0.13310012146830558,
      "num_tokens": 3093030.0,
      "step": 1775
    },
    {
      "entropy": 6.348547077178955,
      "epoch": 0.13849445633145302,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999585659634021,
      "loss": 6.052,
      "mean_token_accuracy": 0.13083661049604417,
      "num_tokens": 3101670.0,
      "step": 1780
    },
    {
      "entropy": 6.234915637969971,
      "epoch": 0.13888348570317058,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999580323855301,
      "loss": 6.0479,
      "mean_token_accuracy": 0.13650041073560715,
      "num_tokens": 3109703.0,
      "step": 1785
    },
    {
      "entropy": 6.281410074234008,
      "epoch": 0.13927251507488816,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999574953942781,
      "loss": 6.0274,
      "mean_token_accuracy": 0.13567775562405587,
      "num_tokens": 3118388.0,
      "step": 1790
    },
    {
      "entropy": 6.353200960159302,
      "epoch": 0.13966154444660572,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999569549896541,
      "loss": 6.0302,
      "mean_token_accuracy": 0.13202294185757638,
      "num_tokens": 3126466.0,
      "step": 1795
    },
    {
      "entropy": 6.298340797424316,
      "epoch": 0.14005057381832328,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999564111716665,
      "loss": 6.0839,
      "mean_token_accuracy": 0.1356653556227684,
      "num_tokens": 3135322.0,
      "step": 1800
    },
    {
      "entropy": 6.337237930297851,
      "epoch": 0.14043960319004084,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999558639403232,
      "loss": 6.0588,
      "mean_token_accuracy": 0.1421361118555069,
      "num_tokens": 3144498.0,
      "step": 1805
    },
    {
      "entropy": 6.217201280593872,
      "epoch": 0.1408286325617584,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999553132956328,
      "loss": 6.0453,
      "mean_token_accuracy": 0.1293400026857853,
      "num_tokens": 3153573.0,
      "step": 1810
    },
    {
      "entropy": 6.2849945545196535,
      "epoch": 0.14121766193347599,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999547592376035,
      "loss": 5.9778,
      "mean_token_accuracy": 0.1387234814465046,
      "num_tokens": 3161705.0,
      "step": 1815
    },
    {
      "entropy": 6.3220963954925535,
      "epoch": 0.14160669130519354,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999542017662438,
      "loss": 6.1431,
      "mean_token_accuracy": 0.1342538885772228,
      "num_tokens": 3170068.0,
      "step": 1820
    },
    {
      "entropy": 6.324667310714721,
      "epoch": 0.1419957206769111,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999536408815623,
      "loss": 6.1425,
      "mean_token_accuracy": 0.12565823644399643,
      "num_tokens": 3178140.0,
      "step": 1825
    },
    {
      "entropy": 6.320392513275147,
      "epoch": 0.14238475004862866,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999530765835672,
      "loss": 6.1249,
      "mean_token_accuracy": 0.13419567197561263,
      "num_tokens": 3186518.0,
      "step": 1830
    },
    {
      "entropy": 6.2921764850616455,
      "epoch": 0.14277377942034625,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999525088722673,
      "loss": 6.1074,
      "mean_token_accuracy": 0.13023580014705657,
      "num_tokens": 3195246.0,
      "step": 1835
    },
    {
      "entropy": 6.419817972183227,
      "epoch": 0.1431628087920638,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999519377476711,
      "loss": 6.1211,
      "mean_token_accuracy": 0.12776365801692008,
      "num_tokens": 3203946.0,
      "step": 1840
    },
    {
      "entropy": 6.292534494400025,
      "epoch": 0.14355183816378136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999513632097872,
      "loss": 6.1428,
      "mean_token_accuracy": 0.13157911524176596,
      "num_tokens": 3212208.0,
      "step": 1845
    },
    {
      "entropy": 6.2916301727294925,
      "epoch": 0.14394086753549892,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999507852586245,
      "loss": 6.0611,
      "mean_token_accuracy": 0.1329232007265091,
      "num_tokens": 3221572.0,
      "step": 1850
    },
    {
      "entropy": 6.4152443408966064,
      "epoch": 0.14432989690721648,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999502038941917,
      "loss": 6.1794,
      "mean_token_accuracy": 0.12200985550880432,
      "num_tokens": 3230898.0,
      "step": 1855
    },
    {
      "entropy": 6.27156891822815,
      "epoch": 0.14471892627893407,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999496191164975,
      "loss": 6.1572,
      "mean_token_accuracy": 0.12799620404839515,
      "num_tokens": 3240524.0,
      "step": 1860
    },
    {
      "entropy": 6.280993509292602,
      "epoch": 0.14510795565065163,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499949030925551,
      "loss": 5.926,
      "mean_token_accuracy": 0.13632097840309143,
      "num_tokens": 3248864.0,
      "step": 1865
    },
    {
      "entropy": 6.335269403457642,
      "epoch": 0.14549698502236919,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999484393213609,
      "loss": 6.1133,
      "mean_token_accuracy": 0.1336534596979618,
      "num_tokens": 3256823.0,
      "step": 1870
    },
    {
      "entropy": 6.222145652770996,
      "epoch": 0.14588601439408674,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999478443039363,
      "loss": 6.0304,
      "mean_token_accuracy": 0.1363625131547451,
      "num_tokens": 3265621.0,
      "step": 1875
    },
    {
      "entropy": 6.254001569747925,
      "epoch": 0.14627504376580433,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999472458732862,
      "loss": 6.0469,
      "mean_token_accuracy": 0.13533957973122596,
      "num_tokens": 3273515.0,
      "step": 1880
    },
    {
      "entropy": 6.403596448898315,
      "epoch": 0.1466640731375219,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999466440294198,
      "loss": 6.127,
      "mean_token_accuracy": 0.13045344352722169,
      "num_tokens": 3282283.0,
      "step": 1885
    },
    {
      "entropy": 6.232098484039307,
      "epoch": 0.14705310250923945,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499946038772346,
      "loss": 6.1111,
      "mean_token_accuracy": 0.1315282925963402,
      "num_tokens": 3290792.0,
      "step": 1890
    },
    {
      "entropy": 6.209755945205688,
      "epoch": 0.147442131880957,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999454301020741,
      "loss": 5.9861,
      "mean_token_accuracy": 0.13717192485928537,
      "num_tokens": 3299430.0,
      "step": 1895
    },
    {
      "entropy": 6.237886095046997,
      "epoch": 0.14783116125267456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999448180186133,
      "loss": 6.0308,
      "mean_token_accuracy": 0.13776149302721025,
      "num_tokens": 3307949.0,
      "step": 1900
    },
    {
      "entropy": 6.336954259872437,
      "epoch": 0.14822019062439215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499944202521973,
      "loss": 6.0864,
      "mean_token_accuracy": 0.13430200293660163,
      "num_tokens": 3316900.0,
      "step": 1905
    },
    {
      "entropy": 6.256044054031372,
      "epoch": 0.1486092199961097,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999435836121624,
      "loss": 6.011,
      "mean_token_accuracy": 0.14276451393961906,
      "num_tokens": 3325656.0,
      "step": 1910
    },
    {
      "entropy": 6.280759811401367,
      "epoch": 0.14899824936782727,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499942961289191,
      "loss": 6.2337,
      "mean_token_accuracy": 0.12067875787615776,
      "num_tokens": 3334219.0,
      "step": 1915
    },
    {
      "entropy": 6.266615104675293,
      "epoch": 0.14938727873954483,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999423355530683,
      "loss": 5.9838,
      "mean_token_accuracy": 0.13729652389883995,
      "num_tokens": 3343637.0,
      "step": 1920
    },
    {
      "entropy": 6.2604649543762205,
      "epoch": 0.1497763081112624,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999417064038035,
      "loss": 6.1361,
      "mean_token_accuracy": 0.13217737153172493,
      "num_tokens": 3352624.0,
      "step": 1925
    },
    {
      "entropy": 6.27165207862854,
      "epoch": 0.15016533748297997,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999410738414065,
      "loss": 6.0498,
      "mean_token_accuracy": 0.13848454281687736,
      "num_tokens": 3361394.0,
      "step": 1930
    },
    {
      "entropy": 6.322489500045776,
      "epoch": 0.15055436685469753,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999404378658866,
      "loss": 6.0933,
      "mean_token_accuracy": 0.13730501234531403,
      "num_tokens": 3369406.0,
      "step": 1935
    },
    {
      "entropy": 6.297638177871704,
      "epoch": 0.1509433962264151,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999397984772536,
      "loss": 6.0326,
      "mean_token_accuracy": 0.1359100379049778,
      "num_tokens": 3377737.0,
      "step": 1940
    },
    {
      "entropy": 6.2445666790008545,
      "epoch": 0.15133242559813265,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999391556755172,
      "loss": 6.0976,
      "mean_token_accuracy": 0.1317235164344311,
      "num_tokens": 3385514.0,
      "step": 1945
    },
    {
      "entropy": 6.3746106147766115,
      "epoch": 0.15172145496985023,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999385094606872,
      "loss": 6.0018,
      "mean_token_accuracy": 0.13859265744686128,
      "num_tokens": 3394500.0,
      "step": 1950
    },
    {
      "entropy": 6.192457056045532,
      "epoch": 0.1521104843415678,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999378598327733,
      "loss": 5.9877,
      "mean_token_accuracy": 0.13467359468340873,
      "num_tokens": 3403289.0,
      "step": 1955
    },
    {
      "entropy": 6.171674203872681,
      "epoch": 0.15249951371328535,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999372067917854,
      "loss": 5.9218,
      "mean_token_accuracy": 0.1461538165807724,
      "num_tokens": 3410892.0,
      "step": 1960
    },
    {
      "entropy": 6.118824768066406,
      "epoch": 0.1528885430850029,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999365503377335,
      "loss": 6.0113,
      "mean_token_accuracy": 0.1375325731933117,
      "num_tokens": 3420329.0,
      "step": 1965
    },
    {
      "entropy": 6.259453964233399,
      "epoch": 0.1532775724567205,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999358904706273,
      "loss": 6.0045,
      "mean_token_accuracy": 0.13776946887373925,
      "num_tokens": 3429429.0,
      "step": 1970
    },
    {
      "entropy": 6.2952652931213375,
      "epoch": 0.15366660182843805,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999352271904771,
      "loss": 6.1282,
      "mean_token_accuracy": 0.1321935884654522,
      "num_tokens": 3438198.0,
      "step": 1975
    },
    {
      "entropy": 6.369889831542968,
      "epoch": 0.1540556312001556,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999345604972928,
      "loss": 6.262,
      "mean_token_accuracy": 0.13315195590257645,
      "num_tokens": 3447732.0,
      "step": 1980
    },
    {
      "entropy": 6.301447629928589,
      "epoch": 0.15444466057187317,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999338903910846,
      "loss": 6.027,
      "mean_token_accuracy": 0.13421724662184714,
      "num_tokens": 3456392.0,
      "step": 1985
    },
    {
      "entropy": 6.218537998199463,
      "epoch": 0.15483368994359073,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999332168718626,
      "loss": 6.0684,
      "mean_token_accuracy": 0.13393645808100701,
      "num_tokens": 3464473.0,
      "step": 1990
    },
    {
      "entropy": 6.275947332382202,
      "epoch": 0.15522271931530832,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999325399396371,
      "loss": 6.0328,
      "mean_token_accuracy": 0.13263694792985917,
      "num_tokens": 3473740.0,
      "step": 1995
    },
    {
      "entropy": 6.1024096488952635,
      "epoch": 0.15561174868702587,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999318595944183,
      "loss": 5.9452,
      "mean_token_accuracy": 0.13779015839099884,
      "num_tokens": 3482249.0,
      "step": 2000
    },
    {
      "entropy": 6.3668300151824955,
      "epoch": 0.15600077805874343,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999311758362166,
      "loss": 6.1654,
      "mean_token_accuracy": 0.13282385990023612,
      "num_tokens": 3490753.0,
      "step": 2005
    },
    {
      "entropy": 6.333145952224731,
      "epoch": 0.156389807430461,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999304886650422,
      "loss": 6.1167,
      "mean_token_accuracy": 0.13160909786820413,
      "num_tokens": 3499373.0,
      "step": 2010
    },
    {
      "entropy": 6.249400472640991,
      "epoch": 0.15677883680217858,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999297980809058,
      "loss": 6.0791,
      "mean_token_accuracy": 0.12948477640748024,
      "num_tokens": 3508344.0,
      "step": 2015
    },
    {
      "entropy": 6.3681032180786135,
      "epoch": 0.15716786617389614,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999291040838176,
      "loss": 6.0576,
      "mean_token_accuracy": 0.13269878849387168,
      "num_tokens": 3517009.0,
      "step": 2020
    },
    {
      "entropy": 6.21099362373352,
      "epoch": 0.1575568955456137,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999284066737884,
      "loss": 6.0727,
      "mean_token_accuracy": 0.1345763050019741,
      "num_tokens": 3525989.0,
      "step": 2025
    },
    {
      "entropy": 6.309366989135742,
      "epoch": 0.15794592491733125,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999277058508285,
      "loss": 5.96,
      "mean_token_accuracy": 0.1353427805006504,
      "num_tokens": 3533947.0,
      "step": 2030
    },
    {
      "entropy": 6.186928176879883,
      "epoch": 0.1583349542890488,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999270016149486,
      "loss": 6.0351,
      "mean_token_accuracy": 0.1369633160531521,
      "num_tokens": 3541974.0,
      "step": 2035
    },
    {
      "entropy": 6.187212657928467,
      "epoch": 0.1587239836607664,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999262939661596,
      "loss": 5.9727,
      "mean_token_accuracy": 0.14454701095819472,
      "num_tokens": 3550436.0,
      "step": 2040
    },
    {
      "entropy": 6.289414739608764,
      "epoch": 0.15911301303248396,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999255829044721,
      "loss": 5.9739,
      "mean_token_accuracy": 0.13857920691370965,
      "num_tokens": 3558283.0,
      "step": 2045
    },
    {
      "entropy": 6.116622686386108,
      "epoch": 0.15950204240420152,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999248684298968,
      "loss": 5.8858,
      "mean_token_accuracy": 0.144998886436224,
      "num_tokens": 3566968.0,
      "step": 2050
    },
    {
      "entropy": 6.24771900177002,
      "epoch": 0.15989107177591907,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999241505424447,
      "loss": 6.0221,
      "mean_token_accuracy": 0.13150276690721513,
      "num_tokens": 3575580.0,
      "step": 2055
    },
    {
      "entropy": 6.248853874206543,
      "epoch": 0.16028010114763663,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999234292421265,
      "loss": 6.101,
      "mean_token_accuracy": 0.1321192666888237,
      "num_tokens": 3584389.0,
      "step": 2060
    },
    {
      "entropy": 6.242433500289917,
      "epoch": 0.16066913051935422,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999227045289535,
      "loss": 5.9212,
      "mean_token_accuracy": 0.13805124387145043,
      "num_tokens": 3593009.0,
      "step": 2065
    },
    {
      "entropy": 6.113874626159668,
      "epoch": 0.16105815989107178,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999219764029363,
      "loss": 5.9619,
      "mean_token_accuracy": 0.14041583985090256,
      "num_tokens": 3601914.0,
      "step": 2070
    },
    {
      "entropy": 6.152489185333252,
      "epoch": 0.16144718926278934,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999212448640861,
      "loss": 5.9363,
      "mean_token_accuracy": 0.13822590857744216,
      "num_tokens": 3609883.0,
      "step": 2075
    },
    {
      "entropy": 6.18636155128479,
      "epoch": 0.1618362186345069,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999205099124141,
      "loss": 5.8719,
      "mean_token_accuracy": 0.14322010651230813,
      "num_tokens": 3618103.0,
      "step": 2080
    },
    {
      "entropy": 6.312455940246582,
      "epoch": 0.16222524800622448,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999197715479313,
      "loss": 6.144,
      "mean_token_accuracy": 0.13272519782185555,
      "num_tokens": 3625861.0,
      "step": 2085
    },
    {
      "entropy": 6.127534246444702,
      "epoch": 0.16261427737794204,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999190297706489,
      "loss": 5.9282,
      "mean_token_accuracy": 0.13653620779514314,
      "num_tokens": 3634498.0,
      "step": 2090
    },
    {
      "entropy": 6.104422616958618,
      "epoch": 0.1630033067496596,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999182845805783,
      "loss": 5.8667,
      "mean_token_accuracy": 0.136298568546772,
      "num_tokens": 3644186.0,
      "step": 2095
    },
    {
      "entropy": 6.148884201049805,
      "epoch": 0.16339233612137716,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999175359777306,
      "loss": 6.0077,
      "mean_token_accuracy": 0.13732302114367484,
      "num_tokens": 3652831.0,
      "step": 2100
    },
    {
      "entropy": 6.29127516746521,
      "epoch": 0.16378136549309472,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999167839621174,
      "loss": 5.8715,
      "mean_token_accuracy": 0.14166171625256538,
      "num_tokens": 3661230.0,
      "step": 2105
    },
    {
      "entropy": 6.123455286026001,
      "epoch": 0.1641703948648123,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999160285337501,
      "loss": 6.0161,
      "mean_token_accuracy": 0.13657576814293862,
      "num_tokens": 3669250.0,
      "step": 2110
    },
    {
      "entropy": 6.189156150817871,
      "epoch": 0.16455942423652986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999152696926399,
      "loss": 5.9946,
      "mean_token_accuracy": 0.13579504638910295,
      "num_tokens": 3677568.0,
      "step": 2115
    },
    {
      "entropy": 6.241203927993775,
      "epoch": 0.16494845360824742,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999145074387985,
      "loss": 5.9924,
      "mean_token_accuracy": 0.1418316200375557,
      "num_tokens": 3685488.0,
      "step": 2120
    },
    {
      "entropy": 6.1650886058807375,
      "epoch": 0.16533748297996498,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999137417722374,
      "loss": 5.9612,
      "mean_token_accuracy": 0.13932890743017196,
      "num_tokens": 3694357.0,
      "step": 2125
    },
    {
      "entropy": 6.243571758270264,
      "epoch": 0.16572651235168256,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999129726929684,
      "loss": 6.025,
      "mean_token_accuracy": 0.13471606448292733,
      "num_tokens": 3703173.0,
      "step": 2130
    },
    {
      "entropy": 6.174264240264892,
      "epoch": 0.16611554172340012,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999122002010029,
      "loss": 6.0072,
      "mean_token_accuracy": 0.13919395282864572,
      "num_tokens": 3712070.0,
      "step": 2135
    },
    {
      "entropy": 6.181047868728638,
      "epoch": 0.16650457109511768,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999114242963527,
      "loss": 5.9059,
      "mean_token_accuracy": 0.14334050118923186,
      "num_tokens": 3720245.0,
      "step": 2140
    },
    {
      "entropy": 6.130413055419922,
      "epoch": 0.16689360046683524,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999106449790298,
      "loss": 5.9891,
      "mean_token_accuracy": 0.13500823676586152,
      "num_tokens": 3730064.0,
      "step": 2145
    },
    {
      "entropy": 6.13914041519165,
      "epoch": 0.1672826298385528,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999098622490458,
      "loss": 5.9473,
      "mean_token_accuracy": 0.1373259760439396,
      "num_tokens": 3738741.0,
      "step": 2150
    },
    {
      "entropy": 6.230259609222412,
      "epoch": 0.16767165921027039,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999090761064125,
      "loss": 6.0108,
      "mean_token_accuracy": 0.13627254143357276,
      "num_tokens": 3747472.0,
      "step": 2155
    },
    {
      "entropy": 6.231689691543579,
      "epoch": 0.16806068858198794,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499908286551142,
      "loss": 6.0384,
      "mean_token_accuracy": 0.1343052066862583,
      "num_tokens": 3756643.0,
      "step": 2160
    },
    {
      "entropy": 6.230075120925903,
      "epoch": 0.1684497179537055,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999074935832463,
      "loss": 6.1358,
      "mean_token_accuracy": 0.12708577439188956,
      "num_tokens": 3765490.0,
      "step": 2165
    },
    {
      "entropy": 6.219008159637451,
      "epoch": 0.16883874732542306,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999066972027373,
      "loss": 5.8962,
      "mean_token_accuracy": 0.13918948471546172,
      "num_tokens": 3773063.0,
      "step": 2170
    },
    {
      "entropy": 6.2020885944366455,
      "epoch": 0.16922777669714065,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999058974096271,
      "loss": 6.0591,
      "mean_token_accuracy": 0.13676124885678292,
      "num_tokens": 3781763.0,
      "step": 2175
    },
    {
      "entropy": 6.239950656890869,
      "epoch": 0.1696168060688582,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499905094203928,
      "loss": 6.0534,
      "mean_token_accuracy": 0.13679237961769103,
      "num_tokens": 3790582.0,
      "step": 2180
    },
    {
      "entropy": 6.150587844848633,
      "epoch": 0.17000583544057576,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499904287585652,
      "loss": 5.867,
      "mean_token_accuracy": 0.1422402709722519,
      "num_tokens": 3798745.0,
      "step": 2185
    },
    {
      "entropy": 6.110662794113159,
      "epoch": 0.17039486481229332,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999034775548113,
      "loss": 5.9639,
      "mean_token_accuracy": 0.1360377162694931,
      "num_tokens": 3806901.0,
      "step": 2190
    },
    {
      "entropy": 6.1845409870147705,
      "epoch": 0.17078389418401088,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004999026641114185,
      "loss": 5.9155,
      "mean_token_accuracy": 0.14578790441155434,
      "num_tokens": 3815742.0,
      "step": 2195
    },
    {
      "entropy": 6.143110513687134,
      "epoch": 0.17117292355572847,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999018472554857,
      "loss": 5.9944,
      "mean_token_accuracy": 0.13587727323174476,
      "num_tokens": 3824468.0,
      "step": 2200
    },
    {
      "entropy": 6.135712718963623,
      "epoch": 0.17156195292744603,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999010269870253,
      "loss": 5.9584,
      "mean_token_accuracy": 0.143868188560009,
      "num_tokens": 3832834.0,
      "step": 2205
    },
    {
      "entropy": 6.166628980636597,
      "epoch": 0.17195098229916359,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999002033060498,
      "loss": 5.9977,
      "mean_token_accuracy": 0.14069228172302245,
      "num_tokens": 3841780.0,
      "step": 2210
    },
    {
      "entropy": 6.188239908218383,
      "epoch": 0.17234001167088114,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998993762125716,
      "loss": 6.1029,
      "mean_token_accuracy": 0.13046709224581718,
      "num_tokens": 3849840.0,
      "step": 2215
    },
    {
      "entropy": 6.256720972061157,
      "epoch": 0.17272904104259873,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998985457066034,
      "loss": 6.0247,
      "mean_token_accuracy": 0.13210274502635003,
      "num_tokens": 3858204.0,
      "step": 2220
    },
    {
      "entropy": 6.179963779449463,
      "epoch": 0.1731180704143163,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998977117881576,
      "loss": 5.9783,
      "mean_token_accuracy": 0.1354764722287655,
      "num_tokens": 3866615.0,
      "step": 2225
    },
    {
      "entropy": 6.147153520584107,
      "epoch": 0.17350709978603385,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998968744572472,
      "loss": 6.0107,
      "mean_token_accuracy": 0.14103505611419678,
      "num_tokens": 3876003.0,
      "step": 2230
    },
    {
      "entropy": 6.217945766448975,
      "epoch": 0.1738961291577514,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998960337138845,
      "loss": 5.9371,
      "mean_token_accuracy": 0.14079653322696686,
      "num_tokens": 3884362.0,
      "step": 2235
    },
    {
      "entropy": 6.073576307296753,
      "epoch": 0.17428515852946896,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998951895580826,
      "loss": 5.897,
      "mean_token_accuracy": 0.14070344492793083,
      "num_tokens": 3893177.0,
      "step": 2240
    },
    {
      "entropy": 6.157261037826538,
      "epoch": 0.17467418790118655,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998943419898542,
      "loss": 5.9474,
      "mean_token_accuracy": 0.1386074975132942,
      "num_tokens": 3903006.0,
      "step": 2245
    },
    {
      "entropy": 6.155741262435913,
      "epoch": 0.1750632172729041,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998934910092119,
      "loss": 5.8105,
      "mean_token_accuracy": 0.14036383107304573,
      "num_tokens": 3912256.0,
      "step": 2250
    },
    {
      "entropy": 6.107159423828125,
      "epoch": 0.17545224664462167,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998926366161689,
      "loss": 5.9881,
      "mean_token_accuracy": 0.13784814327955247,
      "num_tokens": 3921121.0,
      "step": 2255
    },
    {
      "entropy": 6.162649297714234,
      "epoch": 0.17584127601633923,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998917788107383,
      "loss": 5.936,
      "mean_token_accuracy": 0.14180586934089662,
      "num_tokens": 3929348.0,
      "step": 2260
    },
    {
      "entropy": 6.186119127273559,
      "epoch": 0.17623030538805678,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998909175929326,
      "loss": 6.1057,
      "mean_token_accuracy": 0.13549725115299224,
      "num_tokens": 3938073.0,
      "step": 2265
    },
    {
      "entropy": 6.232157850265503,
      "epoch": 0.17661933475977437,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998900529627653,
      "loss": 6.0756,
      "mean_token_accuracy": 0.13264880701899529,
      "num_tokens": 3946641.0,
      "step": 2270
    },
    {
      "entropy": 6.23204550743103,
      "epoch": 0.17700836413149193,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998891849202495,
      "loss": 5.9243,
      "mean_token_accuracy": 0.14050190299749374,
      "num_tokens": 3954586.0,
      "step": 2275
    },
    {
      "entropy": 6.094369125366211,
      "epoch": 0.1773973935032095,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499888313465398,
      "loss": 5.9946,
      "mean_token_accuracy": 0.13950968906283379,
      "num_tokens": 3963745.0,
      "step": 2280
    },
    {
      "entropy": 6.103000974655151,
      "epoch": 0.17778642287492705,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998874385982247,
      "loss": 5.8128,
      "mean_token_accuracy": 0.14982413798570632,
      "num_tokens": 3972452.0,
      "step": 2285
    },
    {
      "entropy": 6.083374404907227,
      "epoch": 0.17817545224664463,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998865603187421,
      "loss": 5.9248,
      "mean_token_accuracy": 0.14283788874745368,
      "num_tokens": 3980770.0,
      "step": 2290
    },
    {
      "entropy": 6.199732637405395,
      "epoch": 0.1785644816183622,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499885678626964,
      "loss": 5.9604,
      "mean_token_accuracy": 0.13827717006206514,
      "num_tokens": 3989545.0,
      "step": 2295
    },
    {
      "entropy": 6.1625518798828125,
      "epoch": 0.17895351099007975,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998847935229038,
      "loss": 5.9269,
      "mean_token_accuracy": 0.13749036863446235,
      "num_tokens": 3998034.0,
      "step": 2300
    },
    {
      "entropy": 6.0795563697814945,
      "epoch": 0.1793425403617973,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998839050065746,
      "loss": 5.9428,
      "mean_token_accuracy": 0.1421938106417656,
      "num_tokens": 4007110.0,
      "step": 2305
    },
    {
      "entropy": 6.205491638183593,
      "epoch": 0.17973156973351487,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998830130779902,
      "loss": 6.0211,
      "mean_token_accuracy": 0.1329338550567627,
      "num_tokens": 4015930.0,
      "step": 2310
    },
    {
      "entropy": 6.017514371871949,
      "epoch": 0.18012059910523245,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499882117737164,
      "loss": 5.8263,
      "mean_token_accuracy": 0.1466153845191002,
      "num_tokens": 4024510.0,
      "step": 2315
    },
    {
      "entropy": 6.066116762161255,
      "epoch": 0.18050962847695,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998812189841096,
      "loss": 5.9342,
      "mean_token_accuracy": 0.13980292826890944,
      "num_tokens": 4033714.0,
      "step": 2320
    },
    {
      "entropy": 6.282118463516236,
      "epoch": 0.18089865784866757,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998803168188407,
      "loss": 6.1212,
      "mean_token_accuracy": 0.1319077990949154,
      "num_tokens": 4042376.0,
      "step": 2325
    },
    {
      "entropy": 6.0946657180786135,
      "epoch": 0.18128768722038513,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998794112413708,
      "loss": 5.8755,
      "mean_token_accuracy": 0.1430434338748455,
      "num_tokens": 4050521.0,
      "step": 2330
    },
    {
      "entropy": 6.216298198699951,
      "epoch": 0.18167671659210272,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998785022517137,
      "loss": 6.0448,
      "mean_token_accuracy": 0.13777603581547737,
      "num_tokens": 4059384.0,
      "step": 2335
    },
    {
      "entropy": 6.148222923278809,
      "epoch": 0.18206574596382027,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998775898498835,
      "loss": 5.9296,
      "mean_token_accuracy": 0.13958509340882302,
      "num_tokens": 4067655.0,
      "step": 2340
    },
    {
      "entropy": 6.157079124450684,
      "epoch": 0.18245477533553783,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998766740358936,
      "loss": 5.9141,
      "mean_token_accuracy": 0.1455913409590721,
      "num_tokens": 4075828.0,
      "step": 2345
    },
    {
      "entropy": 6.141853141784668,
      "epoch": 0.1828438047072554,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998757548097582,
      "loss": 5.8682,
      "mean_token_accuracy": 0.1426569327712059,
      "num_tokens": 4083796.0,
      "step": 2350
    },
    {
      "entropy": 6.067271041870117,
      "epoch": 0.18323283407897295,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998748321714911,
      "loss": 6.0011,
      "mean_token_accuracy": 0.14092514365911485,
      "num_tokens": 4092981.0,
      "step": 2355
    },
    {
      "entropy": 6.112659883499146,
      "epoch": 0.18362186345069054,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998739061211065,
      "loss": 5.827,
      "mean_token_accuracy": 0.14185563325881959,
      "num_tokens": 4101363.0,
      "step": 2360
    },
    {
      "entropy": 6.115339231491089,
      "epoch": 0.1840108928224081,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998729766586181,
      "loss": 5.9131,
      "mean_token_accuracy": 0.14075915217399598,
      "num_tokens": 4110164.0,
      "step": 2365
    },
    {
      "entropy": 6.212285280227661,
      "epoch": 0.18439992219412565,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998720437840403,
      "loss": 6.0276,
      "mean_token_accuracy": 0.13580502569675446,
      "num_tokens": 4119098.0,
      "step": 2370
    },
    {
      "entropy": 6.153889179229736,
      "epoch": 0.1847889515658432,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998711074973871,
      "loss": 5.9253,
      "mean_token_accuracy": 0.14312097132205964,
      "num_tokens": 4127522.0,
      "step": 2375
    },
    {
      "entropy": 6.095731449127197,
      "epoch": 0.1851779809375608,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998701677986728,
      "loss": 5.9557,
      "mean_token_accuracy": 0.1446719117462635,
      "num_tokens": 4136311.0,
      "step": 2380
    },
    {
      "entropy": 6.094022607803344,
      "epoch": 0.18556701030927836,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998692246879116,
      "loss": 5.8754,
      "mean_token_accuracy": 0.14761602729558945,
      "num_tokens": 4145160.0,
      "step": 2385
    },
    {
      "entropy": 6.067728042602539,
      "epoch": 0.18595603968099592,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998682781651178,
      "loss": 5.8564,
      "mean_token_accuracy": 0.14451601579785348,
      "num_tokens": 4153921.0,
      "step": 2390
    },
    {
      "entropy": 6.078082847595215,
      "epoch": 0.18634506905271347,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998673282303059,
      "loss": 5.9202,
      "mean_token_accuracy": 0.13949069231748581,
      "num_tokens": 4162580.0,
      "step": 2395
    },
    {
      "entropy": 6.101584386825562,
      "epoch": 0.18673409842443103,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998663748834901,
      "loss": 5.9162,
      "mean_token_accuracy": 0.1455289751291275,
      "num_tokens": 4170623.0,
      "step": 2400
    },
    {
      "entropy": 6.083553886413574,
      "epoch": 0.18712312779614862,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998654181246851,
      "loss": 5.9049,
      "mean_token_accuracy": 0.1480133354663849,
      "num_tokens": 4178724.0,
      "step": 2405
    },
    {
      "entropy": 6.097364902496338,
      "epoch": 0.18751215716786618,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998644579539052,
      "loss": 5.9147,
      "mean_token_accuracy": 0.1468910664319992,
      "num_tokens": 4187773.0,
      "step": 2410
    },
    {
      "entropy": 6.116354370117188,
      "epoch": 0.18790118653958374,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499863494371165,
      "loss": 5.9139,
      "mean_token_accuracy": 0.13767781779170035,
      "num_tokens": 4196628.0,
      "step": 2415
    },
    {
      "entropy": 6.088314533233643,
      "epoch": 0.1882902159113013,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998625273764793,
      "loss": 5.9477,
      "mean_token_accuracy": 0.13856262788176538,
      "num_tokens": 4205090.0,
      "step": 2420
    },
    {
      "entropy": 6.103470802307129,
      "epoch": 0.18867924528301888,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998615569698626,
      "loss": 5.8589,
      "mean_token_accuracy": 0.143842813372612,
      "num_tokens": 4212757.0,
      "step": 2425
    },
    {
      "entropy": 6.075373220443725,
      "epoch": 0.18906827465473644,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998605831513296,
      "loss": 5.9298,
      "mean_token_accuracy": 0.14338203594088555,
      "num_tokens": 4221556.0,
      "step": 2430
    },
    {
      "entropy": 6.053972434997559,
      "epoch": 0.189457304026454,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998596059208953,
      "loss": 5.9098,
      "mean_token_accuracy": 0.14371097087860107,
      "num_tokens": 4231524.0,
      "step": 2435
    },
    {
      "entropy": 6.094544792175293,
      "epoch": 0.18984633339817156,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998586252785743,
      "loss": 5.9522,
      "mean_token_accuracy": 0.14406583532691003,
      "num_tokens": 4240229.0,
      "step": 2440
    },
    {
      "entropy": 6.133808565139771,
      "epoch": 0.19023536276988912,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998576412243816,
      "loss": 5.8211,
      "mean_token_accuracy": 0.14667747020721436,
      "num_tokens": 4248053.0,
      "step": 2445
    },
    {
      "entropy": 6.111063623428345,
      "epoch": 0.1906243921416067,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998566537583321,
      "loss": 5.8754,
      "mean_token_accuracy": 0.14372672364115716,
      "num_tokens": 4256019.0,
      "step": 2450
    },
    {
      "entropy": 6.058719348907471,
      "epoch": 0.19101342151332426,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998556628804408,
      "loss": 5.8515,
      "mean_token_accuracy": 0.14850348010659217,
      "num_tokens": 4263805.0,
      "step": 2455
    },
    {
      "entropy": 6.0601448059082035,
      "epoch": 0.19140245088504182,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998546685907225,
      "loss": 5.9217,
      "mean_token_accuracy": 0.14053093194961547,
      "num_tokens": 4272712.0,
      "step": 2460
    },
    {
      "entropy": 6.184451723098755,
      "epoch": 0.19179148025675938,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998536708891927,
      "loss": 5.8494,
      "mean_token_accuracy": 0.14523410499095918,
      "num_tokens": 4281222.0,
      "step": 2465
    },
    {
      "entropy": 6.039083003997803,
      "epoch": 0.19218050962847694,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998526697758663,
      "loss": 5.9574,
      "mean_token_accuracy": 0.1356569655239582,
      "num_tokens": 4290558.0,
      "step": 2470
    },
    {
      "entropy": 6.214207363128662,
      "epoch": 0.19256953900019452,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998516652507585,
      "loss": 5.992,
      "mean_token_accuracy": 0.1386689119040966,
      "num_tokens": 4299684.0,
      "step": 2475
    },
    {
      "entropy": 6.056800842285156,
      "epoch": 0.19295856837191208,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998506573138846,
      "loss": 5.9337,
      "mean_token_accuracy": 0.14160904437303543,
      "num_tokens": 4308938.0,
      "step": 2480
    },
    {
      "entropy": 6.002795457839966,
      "epoch": 0.19334759774362964,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998496459652598,
      "loss": 5.7948,
      "mean_token_accuracy": 0.1451106436550617,
      "num_tokens": 4317646.0,
      "step": 2485
    },
    {
      "entropy": 6.065030527114868,
      "epoch": 0.1937366271153472,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998486312048995,
      "loss": 5.9336,
      "mean_token_accuracy": 0.13834921047091484,
      "num_tokens": 4326876.0,
      "step": 2490
    },
    {
      "entropy": 6.060744047164917,
      "epoch": 0.19412565648706478,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998476130328193,
      "loss": 5.8946,
      "mean_token_accuracy": 0.1431366868317127,
      "num_tokens": 4335391.0,
      "step": 2495
    },
    {
      "entropy": 6.179794597625732,
      "epoch": 0.19451468585878234,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998465914490343,
      "loss": 5.876,
      "mean_token_accuracy": 0.1446529857814312,
      "num_tokens": 4343253.0,
      "step": 2500
    },
    {
      "entropy": 5.931164121627807,
      "epoch": 0.1949037152304999,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00049984556645356,
      "loss": 5.8497,
      "mean_token_accuracy": 0.14115574061870576,
      "num_tokens": 4352161.0,
      "step": 2505
    },
    {
      "entropy": 6.055073976516724,
      "epoch": 0.19529274460221746,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998445380464124,
      "loss": 5.8066,
      "mean_token_accuracy": 0.15022924542427063,
      "num_tokens": 4360414.0,
      "step": 2510
    },
    {
      "entropy": 6.062896585464477,
      "epoch": 0.19568177397393502,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998435062276066,
      "loss": 5.7918,
      "mean_token_accuracy": 0.1486874118447304,
      "num_tokens": 4369375.0,
      "step": 2515
    },
    {
      "entropy": 5.977962446212769,
      "epoch": 0.1960708033456526,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998424709971586,
      "loss": 5.8276,
      "mean_token_accuracy": 0.14558200165629387,
      "num_tokens": 4378638.0,
      "step": 2520
    },
    {
      "entropy": 6.117749452590942,
      "epoch": 0.19645983271737016,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998414323550839,
      "loss": 5.8265,
      "mean_token_accuracy": 0.1476297803223133,
      "num_tokens": 4388262.0,
      "step": 2525
    },
    {
      "entropy": 6.057101821899414,
      "epoch": 0.19684886208908772,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998403903013984,
      "loss": 5.8196,
      "mean_token_accuracy": 0.14260421246290206,
      "num_tokens": 4396894.0,
      "step": 2530
    },
    {
      "entropy": 6.098726224899292,
      "epoch": 0.19723789146080528,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998393448361179,
      "loss": 5.9515,
      "mean_token_accuracy": 0.14099942818284034,
      "num_tokens": 4405455.0,
      "step": 2535
    },
    {
      "entropy": 6.090051698684692,
      "epoch": 0.19762692083252287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499838295959258,
      "loss": 5.8408,
      "mean_token_accuracy": 0.14691002145409585,
      "num_tokens": 4414355.0,
      "step": 2540
    },
    {
      "entropy": 5.944388818740845,
      "epoch": 0.19801595020424043,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499837243670835,
      "loss": 5.7418,
      "mean_token_accuracy": 0.15249036699533464,
      "num_tokens": 4422305.0,
      "step": 2545
    },
    {
      "entropy": 6.083198404312133,
      "epoch": 0.19840497957595798,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998361879708646,
      "loss": 5.8824,
      "mean_token_accuracy": 0.14111227691173553,
      "num_tokens": 4430813.0,
      "step": 2550
    },
    {
      "entropy": 6.062122344970703,
      "epoch": 0.19879400894767554,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499835128859363,
      "loss": 5.845,
      "mean_token_accuracy": 0.1449693761765957,
      "num_tokens": 4440229.0,
      "step": 2555
    },
    {
      "entropy": 5.969413232803345,
      "epoch": 0.1991830383193931,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998340663363461,
      "loss": 5.8636,
      "mean_token_accuracy": 0.14286035373806955,
      "num_tokens": 4449417.0,
      "step": 2560
    },
    {
      "entropy": 6.115420150756836,
      "epoch": 0.1995720676911107,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998330004018301,
      "loss": 5.9049,
      "mean_token_accuracy": 0.14032256752252578,
      "num_tokens": 4457377.0,
      "step": 2565
    },
    {
      "entropy": 6.076362800598145,
      "epoch": 0.19996109706282825,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998319310558312,
      "loss": 5.8205,
      "mean_token_accuracy": 0.14231560602784157,
      "num_tokens": 4466839.0,
      "step": 2570
    },
    {
      "entropy": 6.02218279838562,
      "epoch": 0.2003501264345458,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998308582983657,
      "loss": 5.8572,
      "mean_token_accuracy": 0.14494595527648926,
      "num_tokens": 4476369.0,
      "step": 2575
    },
    {
      "entropy": 6.032584857940674,
      "epoch": 0.20073915580626336,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998297821294497,
      "loss": 5.921,
      "mean_token_accuracy": 0.14298095181584358,
      "num_tokens": 4485609.0,
      "step": 2580
    },
    {
      "entropy": 6.070321846008301,
      "epoch": 0.20112818517798095,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998287025490995,
      "loss": 5.873,
      "mean_token_accuracy": 0.14022249430418016,
      "num_tokens": 4494535.0,
      "step": 2585
    },
    {
      "entropy": 6.069472455978394,
      "epoch": 0.2015172145496985,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998276195573317,
      "loss": 5.8392,
      "mean_token_accuracy": 0.1400721177458763,
      "num_tokens": 4503133.0,
      "step": 2590
    },
    {
      "entropy": 5.981370306015014,
      "epoch": 0.20190624392141607,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998265331541627,
      "loss": 5.7476,
      "mean_token_accuracy": 0.15337812453508376,
      "num_tokens": 4511465.0,
      "step": 2595
    },
    {
      "entropy": 6.02926983833313,
      "epoch": 0.20229527329313363,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998254433396088,
      "loss": 5.8029,
      "mean_token_accuracy": 0.14853528290987014,
      "num_tokens": 4520177.0,
      "step": 2600
    },
    {
      "entropy": 6.066452360153198,
      "epoch": 0.20268430266485118,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998243501136867,
      "loss": 5.9622,
      "mean_token_accuracy": 0.14383991807699203,
      "num_tokens": 4528370.0,
      "step": 2605
    },
    {
      "entropy": 6.163117504119873,
      "epoch": 0.20307333203656877,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998232534764129,
      "loss": 5.99,
      "mean_token_accuracy": 0.13582537844777107,
      "num_tokens": 4538258.0,
      "step": 2610
    },
    {
      "entropy": 6.0580810546875,
      "epoch": 0.20346236140828633,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998221534278041,
      "loss": 5.8843,
      "mean_token_accuracy": 0.14223846048116684,
      "num_tokens": 4547300.0,
      "step": 2615
    },
    {
      "entropy": 6.087620735168457,
      "epoch": 0.2038513907800039,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998210499678769,
      "loss": 5.9576,
      "mean_token_accuracy": 0.13359709531068803,
      "num_tokens": 4556877.0,
      "step": 2620
    },
    {
      "entropy": 6.031925868988037,
      "epoch": 0.20424042015172145,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998199430966481,
      "loss": 5.8074,
      "mean_token_accuracy": 0.1435726575553417,
      "num_tokens": 4565999.0,
      "step": 2625
    },
    {
      "entropy": 6.016840887069702,
      "epoch": 0.20462944952343903,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998188328141346,
      "loss": 5.8493,
      "mean_token_accuracy": 0.13968632593750954,
      "num_tokens": 4574027.0,
      "step": 2630
    },
    {
      "entropy": 5.995946645736694,
      "epoch": 0.2050184788951566,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998177191203531,
      "loss": 5.8536,
      "mean_token_accuracy": 0.13539135009050368,
      "num_tokens": 4582402.0,
      "step": 2635
    },
    {
      "entropy": 6.051613092422485,
      "epoch": 0.20540750826687415,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998166020153204,
      "loss": 5.8055,
      "mean_token_accuracy": 0.14385574758052827,
      "num_tokens": 4590381.0,
      "step": 2640
    },
    {
      "entropy": 6.0391960620880125,
      "epoch": 0.2057965376385917,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998154814990538,
      "loss": 5.8637,
      "mean_token_accuracy": 0.14247775822877884,
      "num_tokens": 4598910.0,
      "step": 2645
    },
    {
      "entropy": 6.170034503936767,
      "epoch": 0.20618556701030927,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049981435757157,
      "loss": 5.9733,
      "mean_token_accuracy": 0.13756560906767845,
      "num_tokens": 4607400.0,
      "step": 2650
    },
    {
      "entropy": 6.0044433116912845,
      "epoch": 0.20657459638202685,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998132302328862,
      "loss": 5.9158,
      "mean_token_accuracy": 0.14620504081249236,
      "num_tokens": 4616510.0,
      "step": 2655
    },
    {
      "entropy": 6.089318561553955,
      "epoch": 0.2069636257537444,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998120994830195,
      "loss": 5.8585,
      "mean_token_accuracy": 0.15085278004407882,
      "num_tokens": 4624529.0,
      "step": 2660
    },
    {
      "entropy": 6.052582168579102,
      "epoch": 0.20735265512546197,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499810965321987,
      "loss": 5.7948,
      "mean_token_accuracy": 0.1443331226706505,
      "num_tokens": 4632141.0,
      "step": 2665
    },
    {
      "entropy": 6.077098178863525,
      "epoch": 0.20774168449717953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998098277498059,
      "loss": 6.0011,
      "mean_token_accuracy": 0.14341832026839257,
      "num_tokens": 4640378.0,
      "step": 2670
    },
    {
      "entropy": 6.038132619857788,
      "epoch": 0.2081307138688971,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998086867664934,
      "loss": 5.7666,
      "mean_token_accuracy": 0.14619476199150086,
      "num_tokens": 4649313.0,
      "step": 2675
    },
    {
      "entropy": 6.0627112865448,
      "epoch": 0.20851974324061467,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499807542372067,
      "loss": 5.8704,
      "mean_token_accuracy": 0.14677273184061052,
      "num_tokens": 4658272.0,
      "step": 2680
    },
    {
      "entropy": 5.999119997024536,
      "epoch": 0.20890877261233223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998063945665439,
      "loss": 5.7272,
      "mean_token_accuracy": 0.1503640040755272,
      "num_tokens": 4665970.0,
      "step": 2685
    },
    {
      "entropy": 6.0891471862792965,
      "epoch": 0.2092978019840498,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998052433499416,
      "loss": 5.9918,
      "mean_token_accuracy": 0.1369311533868313,
      "num_tokens": 4675308.0,
      "step": 2690
    },
    {
      "entropy": 6.027482986450195,
      "epoch": 0.20968683135576735,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998040887222776,
      "loss": 5.8264,
      "mean_token_accuracy": 0.14104743599891661,
      "num_tokens": 4684316.0,
      "step": 2695
    },
    {
      "entropy": 6.0261436939239506,
      "epoch": 0.21007586072748494,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998029306835693,
      "loss": 5.7484,
      "mean_token_accuracy": 0.1449325144290924,
      "num_tokens": 4693029.0,
      "step": 2700
    },
    {
      "entropy": 6.008954334259033,
      "epoch": 0.2104648900992025,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998017692338344,
      "loss": 5.9501,
      "mean_token_accuracy": 0.1345457024872303,
      "num_tokens": 4702404.0,
      "step": 2705
    },
    {
      "entropy": 6.1038126945495605,
      "epoch": 0.21085391947092005,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998006043730905,
      "loss": 5.9081,
      "mean_token_accuracy": 0.1360231891274452,
      "num_tokens": 4711126.0,
      "step": 2710
    },
    {
      "entropy": 6.061972713470459,
      "epoch": 0.2112429488426376,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997994361013551,
      "loss": 5.9103,
      "mean_token_accuracy": 0.1473036840558052,
      "num_tokens": 4719816.0,
      "step": 2715
    },
    {
      "entropy": 6.112641429901123,
      "epoch": 0.21163197821435517,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997982644186461,
      "loss": 5.8544,
      "mean_token_accuracy": 0.14111861065030099,
      "num_tokens": 4728294.0,
      "step": 2720
    },
    {
      "entropy": 6.039841270446777,
      "epoch": 0.21202100758607276,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997970893249813,
      "loss": 5.781,
      "mean_token_accuracy": 0.14560880810022353,
      "num_tokens": 4736708.0,
      "step": 2725
    },
    {
      "entropy": 6.011950349807739,
      "epoch": 0.21241003695779032,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997959108203785,
      "loss": 5.9664,
      "mean_token_accuracy": 0.13347356244921685,
      "num_tokens": 4744893.0,
      "step": 2730
    },
    {
      "entropy": 6.164001941680908,
      "epoch": 0.21279906632950787,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997947289048554,
      "loss": 5.909,
      "mean_token_accuracy": 0.14289160668849946,
      "num_tokens": 4753246.0,
      "step": 2735
    },
    {
      "entropy": 6.004565238952637,
      "epoch": 0.21318809570122543,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997935435784302,
      "loss": 5.8313,
      "mean_token_accuracy": 0.13941832184791564,
      "num_tokens": 4762229.0,
      "step": 2740
    },
    {
      "entropy": 5.904656267166137,
      "epoch": 0.21357712507294302,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997923548411208,
      "loss": 5.788,
      "mean_token_accuracy": 0.1491306185722351,
      "num_tokens": 4771315.0,
      "step": 2745
    },
    {
      "entropy": 5.89280514717102,
      "epoch": 0.21396615444466058,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499791162692945,
      "loss": 5.7262,
      "mean_token_accuracy": 0.15562662929296495,
      "num_tokens": 4779848.0,
      "step": 2750
    },
    {
      "entropy": 6.105467939376831,
      "epoch": 0.21435518381637814,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997899671339214,
      "loss": 5.953,
      "mean_token_accuracy": 0.13940655589103698,
      "num_tokens": 4788939.0,
      "step": 2755
    },
    {
      "entropy": 6.0590917587280275,
      "epoch": 0.2147442131880957,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997887681640675,
      "loss": 5.8526,
      "mean_token_accuracy": 0.14215253964066504,
      "num_tokens": 4797719.0,
      "step": 2760
    },
    {
      "entropy": 6.048555850982666,
      "epoch": 0.21513324255981325,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997875657834021,
      "loss": 5.9096,
      "mean_token_accuracy": 0.14236008748412132,
      "num_tokens": 4805601.0,
      "step": 2765
    },
    {
      "entropy": 6.133038139343261,
      "epoch": 0.21552227193153084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997863599919432,
      "loss": 5.8669,
      "mean_token_accuracy": 0.14362844973802566,
      "num_tokens": 4814993.0,
      "step": 2770
    },
    {
      "entropy": 6.037166595458984,
      "epoch": 0.2159113013032484,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997851507897089,
      "loss": 5.8201,
      "mean_token_accuracy": 0.1418660469353199,
      "num_tokens": 4824156.0,
      "step": 2775
    },
    {
      "entropy": 5.9804931640625,
      "epoch": 0.21630033067496596,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997839381767178,
      "loss": 5.914,
      "mean_token_accuracy": 0.13883057311177255,
      "num_tokens": 4833827.0,
      "step": 2780
    },
    {
      "entropy": 6.000218725204467,
      "epoch": 0.21668936004668352,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997827221529882,
      "loss": 5.8457,
      "mean_token_accuracy": 0.14448366090655326,
      "num_tokens": 4842801.0,
      "step": 2785
    },
    {
      "entropy": 5.977024841308594,
      "epoch": 0.2170783894184011,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997815027185386,
      "loss": 5.7228,
      "mean_token_accuracy": 0.1479553274810314,
      "num_tokens": 4850928.0,
      "step": 2790
    },
    {
      "entropy": 6.019153690338134,
      "epoch": 0.21746741879011866,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997802798733874,
      "loss": 5.8354,
      "mean_token_accuracy": 0.14865680709481238,
      "num_tokens": 4859510.0,
      "step": 2795
    },
    {
      "entropy": 6.055089998245239,
      "epoch": 0.21785644816183622,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997790536175534,
      "loss": 5.8612,
      "mean_token_accuracy": 0.14595925584435462,
      "num_tokens": 4868201.0,
      "step": 2800
    },
    {
      "entropy": 5.951241731643677,
      "epoch": 0.21824547753355378,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997778239510548,
      "loss": 5.7896,
      "mean_token_accuracy": 0.1492677852511406,
      "num_tokens": 4876896.0,
      "step": 2805
    },
    {
      "entropy": 5.933494901657104,
      "epoch": 0.21863450690527134,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997765908739106,
      "loss": 5.7462,
      "mean_token_accuracy": 0.1509405829012394,
      "num_tokens": 4885136.0,
      "step": 2810
    },
    {
      "entropy": 5.998682165145874,
      "epoch": 0.21902353627698892,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997753543861395,
      "loss": 5.7656,
      "mean_token_accuracy": 0.14779986441135406,
      "num_tokens": 4893277.0,
      "step": 2815
    },
    {
      "entropy": 5.959256601333618,
      "epoch": 0.21941256564870648,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049977411448776,
      "loss": 5.9097,
      "mean_token_accuracy": 0.14146498516201972,
      "num_tokens": 4901977.0,
      "step": 2820
    },
    {
      "entropy": 6.075650453567505,
      "epoch": 0.21980159502042404,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997728711787912,
      "loss": 5.9125,
      "mean_token_accuracy": 0.13340190574526786,
      "num_tokens": 4910888.0,
      "step": 2825
    },
    {
      "entropy": 6.026413440704346,
      "epoch": 0.2201906243921416,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997716244592518,
      "loss": 5.9052,
      "mean_token_accuracy": 0.13878937885165216,
      "num_tokens": 4920369.0,
      "step": 2830
    },
    {
      "entropy": 6.093588161468506,
      "epoch": 0.22057965376385918,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997703743291607,
      "loss": 5.8955,
      "mean_token_accuracy": 0.14121510535478593,
      "num_tokens": 4928799.0,
      "step": 2835
    },
    {
      "entropy": 6.091677141189575,
      "epoch": 0.22096868313557674,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499769120788537,
      "loss": 5.7852,
      "mean_token_accuracy": 0.15266166031360626,
      "num_tokens": 4937194.0,
      "step": 2840
    },
    {
      "entropy": 5.987790822982788,
      "epoch": 0.2213577125072943,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997678638373995,
      "loss": 5.8642,
      "mean_token_accuracy": 0.14156924486160277,
      "num_tokens": 4946174.0,
      "step": 2845
    },
    {
      "entropy": 6.006899881362915,
      "epoch": 0.22174674187901186,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997666034757676,
      "loss": 5.8244,
      "mean_token_accuracy": 0.13754797130823135,
      "num_tokens": 4955376.0,
      "step": 2850
    },
    {
      "entropy": 5.980024337768555,
      "epoch": 0.22213577125072942,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997653397036603,
      "loss": 5.8092,
      "mean_token_accuracy": 0.15121594890952111,
      "num_tokens": 4964138.0,
      "step": 2855
    },
    {
      "entropy": 6.042385482788086,
      "epoch": 0.222524800622447,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997640725210965,
      "loss": 5.9065,
      "mean_token_accuracy": 0.14113098457455636,
      "num_tokens": 4973868.0,
      "step": 2860
    },
    {
      "entropy": 6.009365510940552,
      "epoch": 0.22291382999416456,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997628019280958,
      "loss": 5.7881,
      "mean_token_accuracy": 0.1450905978679657,
      "num_tokens": 4982687.0,
      "step": 2865
    },
    {
      "entropy": 6.019914531707764,
      "epoch": 0.22330285936588212,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997615279246773,
      "loss": 5.8441,
      "mean_token_accuracy": 0.14207333847880363,
      "num_tokens": 4992072.0,
      "step": 2870
    },
    {
      "entropy": 5.99617919921875,
      "epoch": 0.22369188873759968,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997602505108603,
      "loss": 5.7744,
      "mean_token_accuracy": 0.14771850109100343,
      "num_tokens": 5000551.0,
      "step": 2875
    },
    {
      "entropy": 6.006196022033691,
      "epoch": 0.22408091810931724,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997589696866644,
      "loss": 5.9401,
      "mean_token_accuracy": 0.14075398147106172,
      "num_tokens": 5009292.0,
      "step": 2880
    },
    {
      "entropy": 6.030385875701905,
      "epoch": 0.22446994748103483,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997576854521088,
      "loss": 5.7943,
      "mean_token_accuracy": 0.14536833316087722,
      "num_tokens": 5018407.0,
      "step": 2885
    },
    {
      "entropy": 5.986768054962158,
      "epoch": 0.22485897685275238,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499756397807213,
      "loss": 5.8587,
      "mean_token_accuracy": 0.14210979640483856,
      "num_tokens": 5027353.0,
      "step": 2890
    },
    {
      "entropy": 6.054828786849976,
      "epoch": 0.22524800622446994,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997551067519966,
      "loss": 5.8523,
      "mean_token_accuracy": 0.14193106740713118,
      "num_tokens": 5036476.0,
      "step": 2895
    },
    {
      "entropy": 6.0023867130279545,
      "epoch": 0.2256370355961875,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997538122864792,
      "loss": 5.808,
      "mean_token_accuracy": 0.14258752316236495,
      "num_tokens": 5045132.0,
      "step": 2900
    },
    {
      "entropy": 6.03665246963501,
      "epoch": 0.2260260649679051,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997525144106804,
      "loss": 5.8281,
      "mean_token_accuracy": 0.14626746699213983,
      "num_tokens": 5053665.0,
      "step": 2905
    },
    {
      "entropy": 6.007569980621338,
      "epoch": 0.22641509433962265,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049975121312462,
      "loss": 5.7764,
      "mean_token_accuracy": 0.15300558432936667,
      "num_tokens": 5062516.0,
      "step": 2910
    },
    {
      "entropy": 5.989815378189087,
      "epoch": 0.2268041237113402,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997499084283177,
      "loss": 5.8238,
      "mean_token_accuracy": 0.14331282824277877,
      "num_tokens": 5071260.0,
      "step": 2915
    },
    {
      "entropy": 5.912539625167847,
      "epoch": 0.22719315308305776,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997486003217932,
      "loss": 5.7846,
      "mean_token_accuracy": 0.15056708306074143,
      "num_tokens": 5079215.0,
      "step": 2920
    },
    {
      "entropy": 5.9982171058654785,
      "epoch": 0.22758218245477532,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997472888050664,
      "loss": 5.8364,
      "mean_token_accuracy": 0.14199614971876146,
      "num_tokens": 5088404.0,
      "step": 2925
    },
    {
      "entropy": 5.948773670196533,
      "epoch": 0.2279712118264929,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997459738781573,
      "loss": 5.7779,
      "mean_token_accuracy": 0.14737778007984162,
      "num_tokens": 5096756.0,
      "step": 2930
    },
    {
      "entropy": 6.046881580352784,
      "epoch": 0.22836024119821047,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997446555410857,
      "loss": 5.9382,
      "mean_token_accuracy": 0.14340072497725487,
      "num_tokens": 5106160.0,
      "step": 2935
    },
    {
      "entropy": 5.882016944885254,
      "epoch": 0.22874927056992803,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997433337938716,
      "loss": 5.8607,
      "mean_token_accuracy": 0.14600483030080796,
      "num_tokens": 5114915.0,
      "step": 2940
    },
    {
      "entropy": 6.0260986328125,
      "epoch": 0.22913829994164558,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997420086365351,
      "loss": 5.7735,
      "mean_token_accuracy": 0.1435887724161148,
      "num_tokens": 5123249.0,
      "step": 2945
    },
    {
      "entropy": 5.994471597671509,
      "epoch": 0.22952732931336317,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997406800690966,
      "loss": 5.7702,
      "mean_token_accuracy": 0.14430145174264908,
      "num_tokens": 5131906.0,
      "step": 2950
    },
    {
      "entropy": 5.992271280288696,
      "epoch": 0.22991635868508073,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997393480915758,
      "loss": 5.7451,
      "mean_token_accuracy": 0.15099966078996657,
      "num_tokens": 5140740.0,
      "step": 2955
    },
    {
      "entropy": 6.006669330596924,
      "epoch": 0.2303053880567983,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997380127039931,
      "loss": 5.9405,
      "mean_token_accuracy": 0.14362909346818925,
      "num_tokens": 5149748.0,
      "step": 2960
    },
    {
      "entropy": 5.904706525802612,
      "epoch": 0.23069441742851585,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997366739063688,
      "loss": 5.6624,
      "mean_token_accuracy": 0.15620378851890565,
      "num_tokens": 5158475.0,
      "step": 2965
    },
    {
      "entropy": 5.965284395217895,
      "epoch": 0.2310834468002334,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997353316987231,
      "loss": 5.8742,
      "mean_token_accuracy": 0.13686980605125426,
      "num_tokens": 5166593.0,
      "step": 2970
    },
    {
      "entropy": 6.072088766098022,
      "epoch": 0.231472476171951,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997339860810765,
      "loss": 5.8417,
      "mean_token_accuracy": 0.14505124390125274,
      "num_tokens": 5175557.0,
      "step": 2975
    },
    {
      "entropy": 5.97886700630188,
      "epoch": 0.23186150554366855,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997326370534495,
      "loss": 5.7418,
      "mean_token_accuracy": 0.14757711663842202,
      "num_tokens": 5184140.0,
      "step": 2980
    },
    {
      "entropy": 5.984816789627075,
      "epoch": 0.2322505349153861,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997312846158622,
      "loss": 5.8323,
      "mean_token_accuracy": 0.14205295890569686,
      "num_tokens": 5192347.0,
      "step": 2985
    },
    {
      "entropy": 6.038908290863037,
      "epoch": 0.23263956428710367,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997299287683355,
      "loss": 5.7938,
      "mean_token_accuracy": 0.15015682727098464,
      "num_tokens": 5201327.0,
      "step": 2990
    },
    {
      "entropy": 5.9845648288726805,
      "epoch": 0.23302859365882125,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997285695108898,
      "loss": 5.8845,
      "mean_token_accuracy": 0.1420726865530014,
      "num_tokens": 5209873.0,
      "step": 2995
    },
    {
      "entropy": 6.02182092666626,
      "epoch": 0.2334176230305388,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997272068435458,
      "loss": 5.7891,
      "mean_token_accuracy": 0.14845749586820603,
      "num_tokens": 5219176.0,
      "step": 3000
    },
    {
      "epoch": 0.2334176230305388,
      "eval_entropy": 5.766755393746957,
      "eval_loss": 5.836002349853516,
      "eval_mean_token_accuracy": 0.15392624230282712,
      "eval_num_tokens": 5219176.0,
      "eval_runtime": 21.8281,
      "eval_samples_per_second": 1903.876,
      "eval_steps_per_second": 237.996,
      "step": 3000
    },
    {
      "entropy": 5.914532804489136,
      "epoch": 0.23380665240225637,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499725840766324,
      "loss": 5.7209,
      "mean_token_accuracy": 0.15771688520908356,
      "num_tokens": 5227310.0,
      "step": 3005
    },
    {
      "entropy": 6.004350805282593,
      "epoch": 0.23419568177397393,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997244712792453,
      "loss": 5.8185,
      "mean_token_accuracy": 0.1469304844737053,
      "num_tokens": 5235882.0,
      "step": 3010
    },
    {
      "entropy": 5.97558012008667,
      "epoch": 0.2345847111456915,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997230983823305,
      "loss": 5.733,
      "mean_token_accuracy": 0.1464428722858429,
      "num_tokens": 5244149.0,
      "step": 3015
    },
    {
      "entropy": 5.811860370635986,
      "epoch": 0.23497374051740907,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997217220756003,
      "loss": 5.5949,
      "mean_token_accuracy": 0.1532672643661499,
      "num_tokens": 5252473.0,
      "step": 3020
    },
    {
      "entropy": 5.928388214111328,
      "epoch": 0.23536276988912663,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997203423590757,
      "loss": 5.7517,
      "mean_token_accuracy": 0.15568342953920364,
      "num_tokens": 5260972.0,
      "step": 3025
    },
    {
      "entropy": 5.972634315490723,
      "epoch": 0.2357517992608442,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997189592327775,
      "loss": 5.7873,
      "mean_token_accuracy": 0.13899434357881546,
      "num_tokens": 5270513.0,
      "step": 3030
    },
    {
      "entropy": 5.970373249053955,
      "epoch": 0.23614082863256175,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997175726967268,
      "loss": 5.7495,
      "mean_token_accuracy": 0.1445789471268654,
      "num_tokens": 5278433.0,
      "step": 3035
    },
    {
      "entropy": 5.905675411224365,
      "epoch": 0.23652985800427934,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997161827509447,
      "loss": 5.7585,
      "mean_token_accuracy": 0.14649613872170447,
      "num_tokens": 5287389.0,
      "step": 3040
    },
    {
      "entropy": 5.956258916854859,
      "epoch": 0.2369188873759969,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997147893954521,
      "loss": 5.8565,
      "mean_token_accuracy": 0.14266041815280914,
      "num_tokens": 5296220.0,
      "step": 3045
    },
    {
      "entropy": 5.96304931640625,
      "epoch": 0.23730791674771445,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997133926302702,
      "loss": 5.7852,
      "mean_token_accuracy": 0.14856525361537934,
      "num_tokens": 5304152.0,
      "step": 3050
    },
    {
      "entropy": 6.020824432373047,
      "epoch": 0.237696946119432,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997119924554204,
      "loss": 5.8112,
      "mean_token_accuracy": 0.14321792423725127,
      "num_tokens": 5312374.0,
      "step": 3055
    },
    {
      "entropy": 5.898687171936035,
      "epoch": 0.23808597549114957,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997105888709236,
      "loss": 5.7338,
      "mean_token_accuracy": 0.1488700330257416,
      "num_tokens": 5320696.0,
      "step": 3060
    },
    {
      "entropy": 5.996364736557007,
      "epoch": 0.23847500486286716,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997091818768015,
      "loss": 5.8581,
      "mean_token_accuracy": 0.14631845131516458,
      "num_tokens": 5328754.0,
      "step": 3065
    },
    {
      "entropy": 6.014861583709717,
      "epoch": 0.23886403423458472,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499707771473075,
      "loss": 5.8258,
      "mean_token_accuracy": 0.14964541494846345,
      "num_tokens": 5337553.0,
      "step": 3070
    },
    {
      "entropy": 5.994286060333252,
      "epoch": 0.23925306360630227,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997063576597659,
      "loss": 5.7521,
      "mean_token_accuracy": 0.15279997289180755,
      "num_tokens": 5346063.0,
      "step": 3075
    },
    {
      "entropy": 5.946002388000489,
      "epoch": 0.23964209297801983,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997049404368954,
      "loss": 5.8209,
      "mean_token_accuracy": 0.14650205597281457,
      "num_tokens": 5354187.0,
      "step": 3080
    },
    {
      "entropy": 5.965845775604248,
      "epoch": 0.2400311223497374,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499703519804485,
      "loss": 5.8657,
      "mean_token_accuracy": 0.1399562805891037,
      "num_tokens": 5362819.0,
      "step": 3085
    },
    {
      "entropy": 5.967871189117432,
      "epoch": 0.24042015172145498,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997020957625564,
      "loss": 5.7274,
      "mean_token_accuracy": 0.15995107144117354,
      "num_tokens": 5371249.0,
      "step": 3090
    },
    {
      "entropy": 5.9214027404785154,
      "epoch": 0.24080918109317254,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997006683111312,
      "loss": 5.7441,
      "mean_token_accuracy": 0.14917703121900558,
      "num_tokens": 5379284.0,
      "step": 3095
    },
    {
      "entropy": 5.997532606124878,
      "epoch": 0.2411982104648901,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499699237450231,
      "loss": 5.8315,
      "mean_token_accuracy": 0.1462565302848816,
      "num_tokens": 5388541.0,
      "step": 3100
    },
    {
      "entropy": 6.00367169380188,
      "epoch": 0.24158723983660765,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996978031798774,
      "loss": 5.9068,
      "mean_token_accuracy": 0.14637081027030946,
      "num_tokens": 5397226.0,
      "step": 3105
    },
    {
      "entropy": 5.973370695114136,
      "epoch": 0.24197626920832524,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996963655000924,
      "loss": 5.8089,
      "mean_token_accuracy": 0.14054676070809363,
      "num_tokens": 5405579.0,
      "step": 3110
    },
    {
      "entropy": 6.009214591979981,
      "epoch": 0.2423652985800428,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996949244108977,
      "loss": 5.9245,
      "mean_token_accuracy": 0.13549011573195457,
      "num_tokens": 5414529.0,
      "step": 3115
    },
    {
      "entropy": 6.035836887359619,
      "epoch": 0.24275432795176036,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499693479912315,
      "loss": 5.8796,
      "mean_token_accuracy": 0.14944745674729348,
      "num_tokens": 5423641.0,
      "step": 3120
    },
    {
      "entropy": 5.9449841499328615,
      "epoch": 0.24314335732347792,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996920320043666,
      "loss": 5.7772,
      "mean_token_accuracy": 0.14886331856250762,
      "num_tokens": 5432558.0,
      "step": 3125
    },
    {
      "entropy": 5.950368738174438,
      "epoch": 0.24353238669519547,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996905806870742,
      "loss": 5.7315,
      "mean_token_accuracy": 0.14748364686965942,
      "num_tokens": 5441449.0,
      "step": 3130
    },
    {
      "entropy": 5.87958459854126,
      "epoch": 0.24392141606691306,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996891259604598,
      "loss": 5.7568,
      "mean_token_accuracy": 0.1489865630865097,
      "num_tokens": 5451142.0,
      "step": 3135
    },
    {
      "entropy": 5.970316457748413,
      "epoch": 0.24431044543863062,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996876678245455,
      "loss": 5.6999,
      "mean_token_accuracy": 0.14703621417284013,
      "num_tokens": 5459290.0,
      "step": 3140
    },
    {
      "entropy": 5.939565229415893,
      "epoch": 0.24469947481034818,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996862062793536,
      "loss": 5.7295,
      "mean_token_accuracy": 0.1500839561223984,
      "num_tokens": 5467776.0,
      "step": 3145
    },
    {
      "entropy": 5.856816625595092,
      "epoch": 0.24508850418206574,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996847413249061,
      "loss": 5.6092,
      "mean_token_accuracy": 0.15638387948274612,
      "num_tokens": 5476005.0,
      "step": 3150
    },
    {
      "entropy": 5.990639543533325,
      "epoch": 0.24547753355378332,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996832729612252,
      "loss": 5.7943,
      "mean_token_accuracy": 0.1473146602511406,
      "num_tokens": 5484225.0,
      "step": 3155
    },
    {
      "entropy": 5.839318799972534,
      "epoch": 0.24586656292550088,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996818011883333,
      "loss": 5.7023,
      "mean_token_accuracy": 0.15613459944725036,
      "num_tokens": 5493111.0,
      "step": 3160
    },
    {
      "entropy": 5.878914022445679,
      "epoch": 0.24625559229721844,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996803260062527,
      "loss": 5.7295,
      "mean_token_accuracy": 0.15031494274735452,
      "num_tokens": 5501574.0,
      "step": 3165
    },
    {
      "entropy": 5.983736276626587,
      "epoch": 0.246644621668936,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996788474150057,
      "loss": 5.7602,
      "mean_token_accuracy": 0.14838263839483262,
      "num_tokens": 5511136.0,
      "step": 3170
    },
    {
      "entropy": 5.882587671279907,
      "epoch": 0.24703365104065356,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996773654146149,
      "loss": 5.7409,
      "mean_token_accuracy": 0.15212823152542115,
      "num_tokens": 5520324.0,
      "step": 3175
    },
    {
      "entropy": 5.967652320861816,
      "epoch": 0.24742268041237114,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996758800051026,
      "loss": 5.7024,
      "mean_token_accuracy": 0.15401688665151597,
      "num_tokens": 5528960.0,
      "step": 3180
    },
    {
      "entropy": 5.9393322467803955,
      "epoch": 0.2478117097840887,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996743911864916,
      "loss": 5.8557,
      "mean_token_accuracy": 0.1460486263036728,
      "num_tokens": 5537391.0,
      "step": 3185
    },
    {
      "entropy": 5.9707262992858885,
      "epoch": 0.24820073915580626,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996728989588041,
      "loss": 5.7953,
      "mean_token_accuracy": 0.14829789772629737,
      "num_tokens": 5546577.0,
      "step": 3190
    },
    {
      "entropy": 6.049851703643799,
      "epoch": 0.24858976852752382,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996714033220632,
      "loss": 5.742,
      "mean_token_accuracy": 0.14722291827201844,
      "num_tokens": 5555247.0,
      "step": 3195
    },
    {
      "entropy": 5.890546751022339,
      "epoch": 0.2489787978992414,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996699042762911,
      "loss": 5.6942,
      "mean_token_accuracy": 0.15134429037570954,
      "num_tokens": 5563598.0,
      "step": 3200
    },
    {
      "entropy": 5.940383005142212,
      "epoch": 0.24936782727095896,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996684018215111,
      "loss": 5.725,
      "mean_token_accuracy": 0.14869000166654586,
      "num_tokens": 5572734.0,
      "step": 3205
    },
    {
      "entropy": 5.9394947528839115,
      "epoch": 0.24975685664267652,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996668959577453,
      "loss": 5.8488,
      "mean_token_accuracy": 0.1437419608235359,
      "num_tokens": 5581224.0,
      "step": 3210
    },
    {
      "entropy": 6.008606481552124,
      "epoch": 0.2501458860143941,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996653866850173,
      "loss": 5.8409,
      "mean_token_accuracy": 0.14231252372264863,
      "num_tokens": 5590138.0,
      "step": 3215
    },
    {
      "entropy": 5.949661636352539,
      "epoch": 0.25053491538611167,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996638740033495,
      "loss": 5.7713,
      "mean_token_accuracy": 0.15526752471923827,
      "num_tokens": 5598654.0,
      "step": 3220
    },
    {
      "entropy": 5.846512508392334,
      "epoch": 0.2509239447578292,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996623579127651,
      "loss": 5.7136,
      "mean_token_accuracy": 0.1543463483452797,
      "num_tokens": 5607404.0,
      "step": 3225
    },
    {
      "entropy": 5.943040132522583,
      "epoch": 0.2513129741295468,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996608384132868,
      "loss": 5.7429,
      "mean_token_accuracy": 0.14950166493654252,
      "num_tokens": 5617133.0,
      "step": 3230
    },
    {
      "entropy": 5.90880651473999,
      "epoch": 0.25170200350126437,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499659315504938,
      "loss": 5.7707,
      "mean_token_accuracy": 0.1484447628259659,
      "num_tokens": 5625389.0,
      "step": 3235
    },
    {
      "entropy": 5.857119131088257,
      "epoch": 0.2520910328729819,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996577891877417,
      "loss": 5.6411,
      "mean_token_accuracy": 0.16271414309740068,
      "num_tokens": 5633985.0,
      "step": 3240
    },
    {
      "entropy": 5.973174619674682,
      "epoch": 0.2524800622446995,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996562594617209,
      "loss": 5.7649,
      "mean_token_accuracy": 0.14711527973413469,
      "num_tokens": 5642524.0,
      "step": 3245
    },
    {
      "entropy": 5.936709356307984,
      "epoch": 0.252869091616417,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996547263268991,
      "loss": 5.7178,
      "mean_token_accuracy": 0.14921012967824937,
      "num_tokens": 5651788.0,
      "step": 3250
    },
    {
      "entropy": 5.888123035430908,
      "epoch": 0.2532581209881346,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996531897832993,
      "loss": 5.7602,
      "mean_token_accuracy": 0.14676389396190642,
      "num_tokens": 5660757.0,
      "step": 3255
    },
    {
      "entropy": 6.043964958190918,
      "epoch": 0.2536471503598522,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499651649830945,
      "loss": 5.8528,
      "mean_token_accuracy": 0.1506285548210144,
      "num_tokens": 5669760.0,
      "step": 3260
    },
    {
      "entropy": 5.922834873199463,
      "epoch": 0.2540361797315697,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996501064698593,
      "loss": 5.7811,
      "mean_token_accuracy": 0.15074339210987092,
      "num_tokens": 5678313.0,
      "step": 3265
    },
    {
      "entropy": 5.939870929718017,
      "epoch": 0.2544252091032873,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499648559700066,
      "loss": 5.6956,
      "mean_token_accuracy": 0.15352511703968047,
      "num_tokens": 5686942.0,
      "step": 3270
    },
    {
      "entropy": 5.876242780685425,
      "epoch": 0.25481423847500484,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996470095215884,
      "loss": 5.6997,
      "mean_token_accuracy": 0.15511421784758567,
      "num_tokens": 5695612.0,
      "step": 3275
    },
    {
      "entropy": 5.961434316635132,
      "epoch": 0.2552032678467224,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996454559344498,
      "loss": 5.8798,
      "mean_token_accuracy": 0.1368470698595047,
      "num_tokens": 5705067.0,
      "step": 3280
    },
    {
      "entropy": 5.967849016189575,
      "epoch": 0.25559229721844,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996438989386741,
      "loss": 5.7754,
      "mean_token_accuracy": 0.15305520966649055,
      "num_tokens": 5714092.0,
      "step": 3285
    },
    {
      "entropy": 5.969217824935913,
      "epoch": 0.25598132659015754,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996423385342847,
      "loss": 5.7351,
      "mean_token_accuracy": 0.15343412458896638,
      "num_tokens": 5722638.0,
      "step": 3290
    },
    {
      "entropy": 5.878257513046265,
      "epoch": 0.25637035596187513,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996407747213055,
      "loss": 5.7449,
      "mean_token_accuracy": 0.14493217319250107,
      "num_tokens": 5731511.0,
      "step": 3295
    },
    {
      "entropy": 5.914233160018921,
      "epoch": 0.25675938533359266,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00049963920749976,
      "loss": 5.7517,
      "mean_token_accuracy": 0.15331401228904723,
      "num_tokens": 5740950.0,
      "step": 3300
    },
    {
      "entropy": 5.908319330215454,
      "epoch": 0.25714841470531025,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996376368696721,
      "loss": 5.705,
      "mean_token_accuracy": 0.14798064678907394,
      "num_tokens": 5749163.0,
      "step": 3305
    },
    {
      "entropy": 5.932552862167358,
      "epoch": 0.25753744407702783,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996360628310656,
      "loss": 5.7536,
      "mean_token_accuracy": 0.14659023731946946,
      "num_tokens": 5758321.0,
      "step": 3310
    },
    {
      "entropy": 5.938609218597412,
      "epoch": 0.25792647344874536,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996344853839644,
      "loss": 5.8042,
      "mean_token_accuracy": 0.14989116787910461,
      "num_tokens": 5767277.0,
      "step": 3315
    },
    {
      "entropy": 5.8936632633209225,
      "epoch": 0.25831550282046295,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996329045283924,
      "loss": 5.7001,
      "mean_token_accuracy": 0.15724189579486847,
      "num_tokens": 5775694.0,
      "step": 3320
    },
    {
      "entropy": 5.810319709777832,
      "epoch": 0.25870453219218054,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996313202643737,
      "loss": 5.6225,
      "mean_token_accuracy": 0.15532325357198715,
      "num_tokens": 5783932.0,
      "step": 3325
    },
    {
      "entropy": 5.897975730895996,
      "epoch": 0.25909356156389807,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996297325919321,
      "loss": 5.7529,
      "mean_token_accuracy": 0.15493739545345306,
      "num_tokens": 5792100.0,
      "step": 3330
    },
    {
      "entropy": 5.90851583480835,
      "epoch": 0.25948259093561565,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996281415110919,
      "loss": 5.7307,
      "mean_token_accuracy": 0.14977378994226456,
      "num_tokens": 5801481.0,
      "step": 3335
    },
    {
      "entropy": 5.9347193241119385,
      "epoch": 0.2598716203073332,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996265470218773,
      "loss": 5.8223,
      "mean_token_accuracy": 0.14538975059986115,
      "num_tokens": 5810053.0,
      "step": 3340
    },
    {
      "entropy": 5.880348157882691,
      "epoch": 0.26026064967905077,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996249491243122,
      "loss": 5.6137,
      "mean_token_accuracy": 0.16179749518632888,
      "num_tokens": 5818454.0,
      "step": 3345
    },
    {
      "entropy": 5.8937677383422855,
      "epoch": 0.26064967905076836,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996233478184211,
      "loss": 5.7778,
      "mean_token_accuracy": 0.14190399944782256,
      "num_tokens": 5827013.0,
      "step": 3350
    },
    {
      "entropy": 5.825893449783325,
      "epoch": 0.2610387084224859,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996217431042282,
      "loss": 5.7808,
      "mean_token_accuracy": 0.1439354784786701,
      "num_tokens": 5835862.0,
      "step": 3355
    },
    {
      "entropy": 5.8302641868591305,
      "epoch": 0.2614277377942035,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499620134981758,
      "loss": 5.5539,
      "mean_token_accuracy": 0.15147414058446884,
      "num_tokens": 5843827.0,
      "step": 3360
    },
    {
      "entropy": 5.872374200820923,
      "epoch": 0.261816767165921,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996185234510346,
      "loss": 5.7502,
      "mean_token_accuracy": 0.14820392578840255,
      "num_tokens": 5852677.0,
      "step": 3365
    },
    {
      "entropy": 5.92858190536499,
      "epoch": 0.2622057965376386,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996169085120828,
      "loss": 5.7093,
      "mean_token_accuracy": 0.15133587270975113,
      "num_tokens": 5861686.0,
      "step": 3370
    },
    {
      "entropy": 5.819103622436524,
      "epoch": 0.2625948259093562,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996152901649268,
      "loss": 5.669,
      "mean_token_accuracy": 0.14954098016023637,
      "num_tokens": 5870690.0,
      "step": 3375
    },
    {
      "entropy": 5.902088308334351,
      "epoch": 0.2629838552810737,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996136684095913,
      "loss": 5.7409,
      "mean_token_accuracy": 0.14959589689970015,
      "num_tokens": 5879804.0,
      "step": 3380
    },
    {
      "entropy": 5.966397905349732,
      "epoch": 0.2633728846527913,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996120432461009,
      "loss": 5.8384,
      "mean_token_accuracy": 0.14443578869104384,
      "num_tokens": 5888781.0,
      "step": 3385
    },
    {
      "entropy": 5.987284231185913,
      "epoch": 0.2637619140245088,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996104146744804,
      "loss": 5.7414,
      "mean_token_accuracy": 0.14745590537786485,
      "num_tokens": 5897439.0,
      "step": 3390
    },
    {
      "entropy": 5.88730821609497,
      "epoch": 0.2641509433962264,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996087826947541,
      "loss": 5.7026,
      "mean_token_accuracy": 0.161055551469326,
      "num_tokens": 5905471.0,
      "step": 3395
    },
    {
      "entropy": 5.804925203323364,
      "epoch": 0.264539972767944,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996071473069471,
      "loss": 5.7367,
      "mean_token_accuracy": 0.15316548198461533,
      "num_tokens": 5913795.0,
      "step": 3400
    },
    {
      "entropy": 6.04078631401062,
      "epoch": 0.26492900213966153,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996055085110842,
      "loss": 5.7997,
      "mean_token_accuracy": 0.15046787038445472,
      "num_tokens": 5922119.0,
      "step": 3405
    },
    {
      "entropy": 5.822887897491455,
      "epoch": 0.2653180315113791,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996038663071902,
      "loss": 5.6847,
      "mean_token_accuracy": 0.1509345754981041,
      "num_tokens": 5930759.0,
      "step": 3410
    },
    {
      "entropy": 5.903107595443726,
      "epoch": 0.26570706088309665,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996022206952901,
      "loss": 5.7198,
      "mean_token_accuracy": 0.15288078486919404,
      "num_tokens": 5939830.0,
      "step": 3415
    },
    {
      "entropy": 5.929271078109741,
      "epoch": 0.26609609025481423,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996005716754086,
      "loss": 5.7181,
      "mean_token_accuracy": 0.15389925241470337,
      "num_tokens": 5947737.0,
      "step": 3420
    },
    {
      "entropy": 5.778887224197388,
      "epoch": 0.2664851196265318,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499598919247571,
      "loss": 5.6762,
      "mean_token_accuracy": 0.15809207260608674,
      "num_tokens": 5955905.0,
      "step": 3425
    },
    {
      "entropy": 5.849305629730225,
      "epoch": 0.26687414899824935,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995972634118023,
      "loss": 5.7386,
      "mean_token_accuracy": 0.15352326929569243,
      "num_tokens": 5965223.0,
      "step": 3430
    },
    {
      "entropy": 5.93175745010376,
      "epoch": 0.26726317836996694,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995956041681277,
      "loss": 5.7526,
      "mean_token_accuracy": 0.15678024291992188,
      "num_tokens": 5973615.0,
      "step": 3435
    },
    {
      "entropy": 5.861295080184936,
      "epoch": 0.2676522077416845,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995939415165722,
      "loss": 5.669,
      "mean_token_accuracy": 0.15896339267492293,
      "num_tokens": 5981600.0,
      "step": 3440
    },
    {
      "entropy": 5.877530956268311,
      "epoch": 0.26804123711340205,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995922754571611,
      "loss": 5.683,
      "mean_token_accuracy": 0.15407007783651352,
      "num_tokens": 5989928.0,
      "step": 3445
    },
    {
      "entropy": 5.829905796051025,
      "epoch": 0.26843026648511964,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995906059899197,
      "loss": 5.7397,
      "mean_token_accuracy": 0.15136908739805222,
      "num_tokens": 5998704.0,
      "step": 3450
    },
    {
      "entropy": 5.886248350143433,
      "epoch": 0.26881929585683717,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995889331148733,
      "loss": 5.7208,
      "mean_token_accuracy": 0.1513502597808838,
      "num_tokens": 6007476.0,
      "step": 3455
    },
    {
      "entropy": 5.973094367980957,
      "epoch": 0.26920832522855476,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995872568320474,
      "loss": 5.7813,
      "mean_token_accuracy": 0.14543018639087676,
      "num_tokens": 6016324.0,
      "step": 3460
    },
    {
      "entropy": 5.911590433120727,
      "epoch": 0.26959735460027234,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995855771414673,
      "loss": 5.7337,
      "mean_token_accuracy": 0.15289803743362426,
      "num_tokens": 6024709.0,
      "step": 3465
    },
    {
      "entropy": 5.858439874649048,
      "epoch": 0.2699863839719899,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995838940431585,
      "loss": 5.5745,
      "mean_token_accuracy": 0.15924426466226577,
      "num_tokens": 6033131.0,
      "step": 3470
    },
    {
      "entropy": 5.817553472518921,
      "epoch": 0.27037541334370746,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995822075371466,
      "loss": 5.7783,
      "mean_token_accuracy": 0.14763862043619155,
      "num_tokens": 6042310.0,
      "step": 3475
    },
    {
      "entropy": 5.891201305389404,
      "epoch": 0.270764442715425,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995805176234571,
      "loss": 5.5947,
      "mean_token_accuracy": 0.15696365535259246,
      "num_tokens": 6050617.0,
      "step": 3480
    },
    {
      "entropy": 5.897118473052979,
      "epoch": 0.2711534720871426,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995788243021158,
      "loss": 5.7699,
      "mean_token_accuracy": 0.14828943312168122,
      "num_tokens": 6060090.0,
      "step": 3485
    },
    {
      "entropy": 5.909312772750854,
      "epoch": 0.27154250145886016,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995771275731483,
      "loss": 5.7084,
      "mean_token_accuracy": 0.15499398112297058,
      "num_tokens": 6068646.0,
      "step": 3490
    },
    {
      "entropy": 5.904059219360351,
      "epoch": 0.2719315308305777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995754274365802,
      "loss": 5.769,
      "mean_token_accuracy": 0.15764224529266357,
      "num_tokens": 6077896.0,
      "step": 3495
    },
    {
      "entropy": 5.8887194156646725,
      "epoch": 0.2723205602022953,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995737238924376,
      "loss": 5.691,
      "mean_token_accuracy": 0.15045809149742126,
      "num_tokens": 6087189.0,
      "step": 3500
    },
    {
      "entropy": 5.8543816089630125,
      "epoch": 0.2727095895740128,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995720169407461,
      "loss": 5.6928,
      "mean_token_accuracy": 0.15313887149095534,
      "num_tokens": 6095394.0,
      "step": 3505
    },
    {
      "entropy": 5.90059175491333,
      "epoch": 0.2730986189457304,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995703065815317,
      "loss": 5.6558,
      "mean_token_accuracy": 0.15778090357780455,
      "num_tokens": 6104135.0,
      "step": 3510
    },
    {
      "entropy": 5.874152612686157,
      "epoch": 0.273487648317448,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995685928148203,
      "loss": 5.774,
      "mean_token_accuracy": 0.14561692923307418,
      "num_tokens": 6112160.0,
      "step": 3515
    },
    {
      "entropy": 5.9434703350067135,
      "epoch": 0.2738766776891655,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499566875640638,
      "loss": 5.775,
      "mean_token_accuracy": 0.15645800679922103,
      "num_tokens": 6120575.0,
      "step": 3520
    },
    {
      "entropy": 5.929290294647217,
      "epoch": 0.2742657070608831,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995651550590108,
      "loss": 5.7745,
      "mean_token_accuracy": 0.15056986957788468,
      "num_tokens": 6129367.0,
      "step": 3525
    },
    {
      "entropy": 5.876437711715698,
      "epoch": 0.2746547364326007,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995634310699647,
      "loss": 5.744,
      "mean_token_accuracy": 0.15265202671289443,
      "num_tokens": 6137440.0,
      "step": 3530
    },
    {
      "entropy": 5.872723627090454,
      "epoch": 0.2750437658043182,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995617036735261,
      "loss": 5.673,
      "mean_token_accuracy": 0.154018035531044,
      "num_tokens": 6146770.0,
      "step": 3535
    },
    {
      "entropy": 6.0044420719146725,
      "epoch": 0.2754327951760358,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995599728697211,
      "loss": 5.7703,
      "mean_token_accuracy": 0.1499793127179146,
      "num_tokens": 6155994.0,
      "step": 3540
    },
    {
      "entropy": 5.789418649673462,
      "epoch": 0.27582182454775334,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995582386585759,
      "loss": 5.6619,
      "mean_token_accuracy": 0.14758216962218285,
      "num_tokens": 6163941.0,
      "step": 3545
    },
    {
      "entropy": 5.8281224250793455,
      "epoch": 0.2762108539194709,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995565010401167,
      "loss": 5.7346,
      "mean_token_accuracy": 0.14870691299438477,
      "num_tokens": 6172681.0,
      "step": 3550
    },
    {
      "entropy": 5.9554907321929935,
      "epoch": 0.2765998832911885,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995547600143702,
      "loss": 5.7072,
      "mean_token_accuracy": 0.14850514829158784,
      "num_tokens": 6181449.0,
      "step": 3555
    },
    {
      "entropy": 5.873755407333374,
      "epoch": 0.27698891266290604,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995530155813625,
      "loss": 5.6887,
      "mean_token_accuracy": 0.15598696768283843,
      "num_tokens": 6190275.0,
      "step": 3560
    },
    {
      "entropy": 5.884641075134278,
      "epoch": 0.2773779420346236,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995512677411203,
      "loss": 5.7351,
      "mean_token_accuracy": 0.1550119176506996,
      "num_tokens": 6198521.0,
      "step": 3565
    },
    {
      "entropy": 5.901494312286377,
      "epoch": 0.27776697140634116,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995495164936698,
      "loss": 5.7628,
      "mean_token_accuracy": 0.1469577133655548,
      "num_tokens": 6207318.0,
      "step": 3570
    },
    {
      "entropy": 5.902689838409424,
      "epoch": 0.27815600077805874,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995477618390381,
      "loss": 5.651,
      "mean_token_accuracy": 0.15427243560552598,
      "num_tokens": 6216693.0,
      "step": 3575
    },
    {
      "entropy": 5.85602765083313,
      "epoch": 0.27854503014977633,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995460037772513,
      "loss": 5.7077,
      "mean_token_accuracy": 0.14970823526382446,
      "num_tokens": 6225207.0,
      "step": 3580
    },
    {
      "entropy": 5.885422992706299,
      "epoch": 0.27893405952149386,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995442423083365,
      "loss": 5.6551,
      "mean_token_accuracy": 0.16121675074100494,
      "num_tokens": 6232653.0,
      "step": 3585
    },
    {
      "entropy": 5.828827238082885,
      "epoch": 0.27932308889321145,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995424774323201,
      "loss": 5.6419,
      "mean_token_accuracy": 0.15972325801849366,
      "num_tokens": 6241093.0,
      "step": 3590
    },
    {
      "entropy": 5.789525175094605,
      "epoch": 0.279712118264929,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499540709149229,
      "loss": 5.564,
      "mean_token_accuracy": 0.16275173425674438,
      "num_tokens": 6249144.0,
      "step": 3595
    },
    {
      "entropy": 5.865627717971802,
      "epoch": 0.28010114763664656,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995389374590901,
      "loss": 5.6389,
      "mean_token_accuracy": 0.15749034732580186,
      "num_tokens": 6257622.0,
      "step": 3600
    },
    {
      "entropy": 5.882948589324951,
      "epoch": 0.28049017700836415,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995371623619301,
      "loss": 5.6636,
      "mean_token_accuracy": 0.15338655188679695,
      "num_tokens": 6266435.0,
      "step": 3605
    },
    {
      "entropy": 5.8372331142425535,
      "epoch": 0.2808792063800817,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499535383857776,
      "loss": 5.7877,
      "mean_token_accuracy": 0.149023699760437,
      "num_tokens": 6275053.0,
      "step": 3610
    },
    {
      "entropy": 5.8476598262786865,
      "epoch": 0.28126823575179927,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499533601946655,
      "loss": 5.6726,
      "mean_token_accuracy": 0.1459781378507614,
      "num_tokens": 6283240.0,
      "step": 3615
    },
    {
      "entropy": 5.800600624084472,
      "epoch": 0.2816572651235168,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995318166285938,
      "loss": 5.6236,
      "mean_token_accuracy": 0.1582282230257988,
      "num_tokens": 6292133.0,
      "step": 3620
    },
    {
      "entropy": 5.931145143508911,
      "epoch": 0.2820462944952344,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995300279036199,
      "loss": 5.7512,
      "mean_token_accuracy": 0.15329633504152299,
      "num_tokens": 6300954.0,
      "step": 3625
    },
    {
      "entropy": 5.803219652175903,
      "epoch": 0.28243532386695197,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049952823577176,
      "loss": 5.623,
      "mean_token_accuracy": 0.16018344163894654,
      "num_tokens": 6309637.0,
      "step": 3630
    },
    {
      "entropy": 5.870447778701783,
      "epoch": 0.2828243532386695,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995264402330416,
      "loss": 5.6889,
      "mean_token_accuracy": 0.153514064848423,
      "num_tokens": 6318539.0,
      "step": 3635
    },
    {
      "entropy": 5.877357006072998,
      "epoch": 0.2832133826103871,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995246412874919,
      "loss": 5.6063,
      "mean_token_accuracy": 0.15626251548528672,
      "num_tokens": 6327109.0,
      "step": 3640
    },
    {
      "entropy": 5.796565914154053,
      "epoch": 0.2836024119821047,
      "grad_norm": 1.0,
      "learning_rate": 0.000499522838935138,
      "loss": 5.6188,
      "mean_token_accuracy": 0.1544191963970661,
      "num_tokens": 6335555.0,
      "step": 3645
    },
    {
      "entropy": 5.89104151725769,
      "epoch": 0.2839914413538222,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995210331760074,
      "loss": 5.7761,
      "mean_token_accuracy": 0.15136173665523528,
      "num_tokens": 6344716.0,
      "step": 3650
    },
    {
      "entropy": 5.848493671417236,
      "epoch": 0.2843804707255398,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995192240101275,
      "loss": 5.6321,
      "mean_token_accuracy": 0.16200316846370696,
      "num_tokens": 6353618.0,
      "step": 3655
    },
    {
      "entropy": 5.9154908657073975,
      "epoch": 0.2847695000972573,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995174114375258,
      "loss": 5.7482,
      "mean_token_accuracy": 0.14789825528860093,
      "num_tokens": 6362039.0,
      "step": 3660
    },
    {
      "entropy": 5.913218307495117,
      "epoch": 0.2851585294689749,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995155954582297,
      "loss": 5.6811,
      "mean_token_accuracy": 0.15068026036024093,
      "num_tokens": 6370331.0,
      "step": 3665
    },
    {
      "entropy": 5.845167112350464,
      "epoch": 0.2855475588406925,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995137760722668,
      "loss": 5.6551,
      "mean_token_accuracy": 0.15810975283384324,
      "num_tokens": 6378943.0,
      "step": 3670
    },
    {
      "entropy": 5.879886245727539,
      "epoch": 0.28593658821241,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995119532796646,
      "loss": 5.7192,
      "mean_token_accuracy": 0.1432995818555355,
      "num_tokens": 6387745.0,
      "step": 3675
    },
    {
      "entropy": 5.7256762981414795,
      "epoch": 0.2863256175841276,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499510127080451,
      "loss": 5.5589,
      "mean_token_accuracy": 0.1562187597155571,
      "num_tokens": 6396915.0,
      "step": 3680
    },
    {
      "entropy": 5.798629856109619,
      "epoch": 0.28671464695584514,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995082974746535,
      "loss": 5.6084,
      "mean_token_accuracy": 0.15959389954805375,
      "num_tokens": 6405453.0,
      "step": 3685
    },
    {
      "entropy": 5.9083412170410154,
      "epoch": 0.28710367632756273,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995064644622999,
      "loss": 5.6961,
      "mean_token_accuracy": 0.15132231414318084,
      "num_tokens": 6413693.0,
      "step": 3690
    },
    {
      "entropy": 5.764428520202637,
      "epoch": 0.2874927056992803,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995046280434181,
      "loss": 5.682,
      "mean_token_accuracy": 0.14993201121687888,
      "num_tokens": 6422738.0,
      "step": 3695
    },
    {
      "entropy": 5.837015295028687,
      "epoch": 0.28788173507099785,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995027882180358,
      "loss": 5.6273,
      "mean_token_accuracy": 0.15370226055383682,
      "num_tokens": 6431743.0,
      "step": 3700
    },
    {
      "entropy": 5.881600713729858,
      "epoch": 0.28827076444271543,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995009449861812,
      "loss": 5.6616,
      "mean_token_accuracy": 0.15563700199127198,
      "num_tokens": 6440586.0,
      "step": 3705
    },
    {
      "entropy": 5.804992055892944,
      "epoch": 0.28865979381443296,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499499098347882,
      "loss": 5.7168,
      "mean_token_accuracy": 0.1542726181447506,
      "num_tokens": 6449644.0,
      "step": 3710
    },
    {
      "entropy": 5.833136034011841,
      "epoch": 0.28904882318615055,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994972483031662,
      "loss": 5.6053,
      "mean_token_accuracy": 0.16261792182922363,
      "num_tokens": 6458752.0,
      "step": 3715
    },
    {
      "entropy": 5.797244644165039,
      "epoch": 0.28943785255786814,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499495394852062,
      "loss": 5.6716,
      "mean_token_accuracy": 0.15677656084299088,
      "num_tokens": 6468333.0,
      "step": 3720
    },
    {
      "entropy": 5.707669687271118,
      "epoch": 0.28982688192958567,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994935379945977,
      "loss": 5.5339,
      "mean_token_accuracy": 0.15437540858983995,
      "num_tokens": 6477406.0,
      "step": 3725
    },
    {
      "entropy": 5.859027671813965,
      "epoch": 0.29021591130130325,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994916777308012,
      "loss": 5.6851,
      "mean_token_accuracy": 0.1511788085103035,
      "num_tokens": 6486399.0,
      "step": 3730
    },
    {
      "entropy": 5.791624116897583,
      "epoch": 0.29060494067302084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994898140607007,
      "loss": 5.5922,
      "mean_token_accuracy": 0.15610771924257277,
      "num_tokens": 6494853.0,
      "step": 3735
    },
    {
      "entropy": 5.872024250030518,
      "epoch": 0.29099397004473837,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994879469843247,
      "loss": 5.7565,
      "mean_token_accuracy": 0.1508682370185852,
      "num_tokens": 6503301.0,
      "step": 3740
    },
    {
      "entropy": 5.838651132583618,
      "epoch": 0.29138299941645596,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004994860765017013,
      "loss": 5.6465,
      "mean_token_accuracy": 0.15341054350137712,
      "num_tokens": 6512685.0,
      "step": 3745
    },
    {
      "entropy": 5.834665298461914,
      "epoch": 0.2917720287881735,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499484202612859,
      "loss": 5.7428,
      "mean_token_accuracy": 0.1516564279794693,
      "num_tokens": 6520673.0,
      "step": 3750
    },
    {
      "entropy": 5.879530334472657,
      "epoch": 0.2921610581598911,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994823253178264,
      "loss": 5.657,
      "mean_token_accuracy": 0.15068983137607575,
      "num_tokens": 6529039.0,
      "step": 3755
    },
    {
      "entropy": 5.832844400405884,
      "epoch": 0.29255008753160866,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994804446166317,
      "loss": 5.7447,
      "mean_token_accuracy": 0.1464340180158615,
      "num_tokens": 6537932.0,
      "step": 3760
    },
    {
      "entropy": 5.897266864776611,
      "epoch": 0.2929391169033262,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994785605093035,
      "loss": 5.6243,
      "mean_token_accuracy": 0.15435930714011192,
      "num_tokens": 6546278.0,
      "step": 3765
    },
    {
      "entropy": 5.840794086456299,
      "epoch": 0.2933281462750438,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994766729958705,
      "loss": 5.6752,
      "mean_token_accuracy": 0.16036221235990525,
      "num_tokens": 6554449.0,
      "step": 3770
    },
    {
      "entropy": 5.830932760238648,
      "epoch": 0.2937171756467613,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994747820763612,
      "loss": 5.6631,
      "mean_token_accuracy": 0.15727959871292113,
      "num_tokens": 6563272.0,
      "step": 3775
    },
    {
      "entropy": 5.917678117752075,
      "epoch": 0.2941062050184789,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994728877508046,
      "loss": 5.6737,
      "mean_token_accuracy": 0.15500990450382232,
      "num_tokens": 6571864.0,
      "step": 3780
    },
    {
      "entropy": 5.834184646606445,
      "epoch": 0.2944952343901965,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994709900192289,
      "loss": 5.6644,
      "mean_token_accuracy": 0.15712282955646514,
      "num_tokens": 6580435.0,
      "step": 3785
    },
    {
      "entropy": 5.909469509124756,
      "epoch": 0.294884263761914,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994690888816635,
      "loss": 5.6642,
      "mean_token_accuracy": 0.15690299719572068,
      "num_tokens": 6589261.0,
      "step": 3790
    },
    {
      "entropy": 5.864827394485474,
      "epoch": 0.2952732931336316,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994671843381367,
      "loss": 5.7462,
      "mean_token_accuracy": 0.15970321893692016,
      "num_tokens": 6598057.0,
      "step": 3795
    },
    {
      "entropy": 5.95170750617981,
      "epoch": 0.29566232250534913,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994652763886778,
      "loss": 5.7119,
      "mean_token_accuracy": 0.14755399972200395,
      "num_tokens": 6606439.0,
      "step": 3800
    },
    {
      "entropy": 5.879061555862426,
      "epoch": 0.2960513518770667,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994633650333155,
      "loss": 5.714,
      "mean_token_accuracy": 0.15110620558261872,
      "num_tokens": 6615350.0,
      "step": 3805
    },
    {
      "entropy": 5.87994384765625,
      "epoch": 0.2964403812487843,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994614502720792,
      "loss": 5.769,
      "mean_token_accuracy": 0.15212479382753372,
      "num_tokens": 6624258.0,
      "step": 3810
    },
    {
      "entropy": 5.970225715637207,
      "epoch": 0.29682941062050183,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994595321049974,
      "loss": 5.748,
      "mean_token_accuracy": 0.1487318679690361,
      "num_tokens": 6632382.0,
      "step": 3815
    },
    {
      "entropy": 5.829820251464843,
      "epoch": 0.2972184399922194,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994576105320994,
      "loss": 5.6629,
      "mean_token_accuracy": 0.15289796888828278,
      "num_tokens": 6640482.0,
      "step": 3820
    },
    {
      "entropy": 5.817120790481567,
      "epoch": 0.29760746936393695,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994556855534146,
      "loss": 5.6476,
      "mean_token_accuracy": 0.15640270859003066,
      "num_tokens": 6648912.0,
      "step": 3825
    },
    {
      "entropy": 5.801205730438232,
      "epoch": 0.29799649873565454,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994537571689719,
      "loss": 5.6288,
      "mean_token_accuracy": 0.15806550532579422,
      "num_tokens": 6656912.0,
      "step": 3830
    },
    {
      "entropy": 5.728374814987182,
      "epoch": 0.2983855281073721,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994518253788007,
      "loss": 5.6519,
      "mean_token_accuracy": 0.15597154051065446,
      "num_tokens": 6665900.0,
      "step": 3835
    },
    {
      "entropy": 5.88460373878479,
      "epoch": 0.29877455747908965,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994498901829304,
      "loss": 5.6988,
      "mean_token_accuracy": 0.15435834527015685,
      "num_tokens": 6674426.0,
      "step": 3840
    },
    {
      "entropy": 5.884772253036499,
      "epoch": 0.29916358685080724,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994479515813903,
      "loss": 5.6541,
      "mean_token_accuracy": 0.15605763345956802,
      "num_tokens": 6683566.0,
      "step": 3845
    },
    {
      "entropy": 5.798302364349365,
      "epoch": 0.2995526162225248,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994460095742096,
      "loss": 5.6867,
      "mean_token_accuracy": 0.1584527686238289,
      "num_tokens": 6693341.0,
      "step": 3850
    },
    {
      "entropy": 5.827801609039307,
      "epoch": 0.29994164559424236,
      "grad_norm": 1.0,
      "learning_rate": 0.000499444064161418,
      "loss": 5.5855,
      "mean_token_accuracy": 0.15363955348730088,
      "num_tokens": 6701734.0,
      "step": 3855
    },
    {
      "entropy": 5.810207843780518,
      "epoch": 0.30033067496595994,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499442115343045,
      "loss": 5.6245,
      "mean_token_accuracy": 0.1571832537651062,
      "num_tokens": 6710572.0,
      "step": 3860
    },
    {
      "entropy": 5.891124248504639,
      "epoch": 0.3007197043376775,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994401631191202,
      "loss": 5.7396,
      "mean_token_accuracy": 0.14851216599345207,
      "num_tokens": 6719355.0,
      "step": 3865
    },
    {
      "entropy": 5.982782506942749,
      "epoch": 0.30110873370939506,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994382074896731,
      "loss": 5.7702,
      "mean_token_accuracy": 0.14574634730815889,
      "num_tokens": 6729279.0,
      "step": 3870
    },
    {
      "entropy": 5.794736433029175,
      "epoch": 0.30149776308111265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994362484547335,
      "loss": 5.5907,
      "mean_token_accuracy": 0.15171013325452803,
      "num_tokens": 6737318.0,
      "step": 3875
    },
    {
      "entropy": 5.8127483367919925,
      "epoch": 0.3018867924528302,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994342860143309,
      "loss": 5.6749,
      "mean_token_accuracy": 0.1597735196352005,
      "num_tokens": 6746520.0,
      "step": 3880
    },
    {
      "entropy": 5.835038423538208,
      "epoch": 0.30227582182454776,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994323201684953,
      "loss": 5.5972,
      "mean_token_accuracy": 0.1554494395852089,
      "num_tokens": 6755685.0,
      "step": 3885
    },
    {
      "entropy": 5.8171398639678955,
      "epoch": 0.3026648511962653,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994303509172566,
      "loss": 5.7417,
      "mean_token_accuracy": 0.15470685362815856,
      "num_tokens": 6764347.0,
      "step": 3890
    },
    {
      "entropy": 5.926692819595337,
      "epoch": 0.3030538805679829,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994283782606444,
      "loss": 5.7076,
      "mean_token_accuracy": 0.15835860669612883,
      "num_tokens": 6772887.0,
      "step": 3895
    },
    {
      "entropy": 5.7954470157623295,
      "epoch": 0.30344290993970047,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994264021986888,
      "loss": 5.6726,
      "mean_token_accuracy": 0.14857605174183847,
      "num_tokens": 6780860.0,
      "step": 3900
    },
    {
      "entropy": 5.790612697601318,
      "epoch": 0.303831939311418,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994244227314197,
      "loss": 5.6931,
      "mean_token_accuracy": 0.15410419106483458,
      "num_tokens": 6789491.0,
      "step": 3905
    },
    {
      "entropy": 5.917981243133545,
      "epoch": 0.3042209686831356,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994224398588672,
      "loss": 5.6175,
      "mean_token_accuracy": 0.15719443410634995,
      "num_tokens": 6797128.0,
      "step": 3910
    },
    {
      "entropy": 5.842257022857666,
      "epoch": 0.3046099980548531,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994204535810615,
      "loss": 5.6526,
      "mean_token_accuracy": 0.15690061897039415,
      "num_tokens": 6805235.0,
      "step": 3915
    },
    {
      "entropy": 5.784561395645142,
      "epoch": 0.3049990274265707,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994184638980326,
      "loss": 5.577,
      "mean_token_accuracy": 0.1572084903717041,
      "num_tokens": 6813863.0,
      "step": 3920
    },
    {
      "entropy": 5.830573415756225,
      "epoch": 0.3053880567982883,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994164708098107,
      "loss": 5.6372,
      "mean_token_accuracy": 0.15513723939657212,
      "num_tokens": 6822015.0,
      "step": 3925
    },
    {
      "entropy": 5.889108514785766,
      "epoch": 0.3057770861700058,
      "grad_norm": 1.0,
      "learning_rate": 0.000499414474316426,
      "loss": 5.6945,
      "mean_token_accuracy": 0.15233523547649383,
      "num_tokens": 6830598.0,
      "step": 3930
    },
    {
      "entropy": 5.857270097732544,
      "epoch": 0.3061661155417234,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994124744179088,
      "loss": 5.6859,
      "mean_token_accuracy": 0.16359390765428544,
      "num_tokens": 6839129.0,
      "step": 3935
    },
    {
      "entropy": 5.8585127830505375,
      "epoch": 0.306555144913441,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994104711142896,
      "loss": 5.6445,
      "mean_token_accuracy": 0.15050655901432036,
      "num_tokens": 6848199.0,
      "step": 3940
    },
    {
      "entropy": 5.837422466278076,
      "epoch": 0.3069441742851585,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994084644055986,
      "loss": 5.6615,
      "mean_token_accuracy": 0.15743141025304794,
      "num_tokens": 6857340.0,
      "step": 3945
    },
    {
      "entropy": 5.805282306671143,
      "epoch": 0.3073332036568761,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994064542918664,
      "loss": 5.6911,
      "mean_token_accuracy": 0.1498067356646061,
      "num_tokens": 6865830.0,
      "step": 3950
    },
    {
      "entropy": 5.843528699874878,
      "epoch": 0.30772223302859364,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994044407731235,
      "loss": 5.7512,
      "mean_token_accuracy": 0.1488284096121788,
      "num_tokens": 6874318.0,
      "step": 3955
    },
    {
      "entropy": 6.015487432479858,
      "epoch": 0.3081112624003112,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994024238494002,
      "loss": 5.7414,
      "mean_token_accuracy": 0.15100244730710982,
      "num_tokens": 6882008.0,
      "step": 3960
    },
    {
      "entropy": 5.84633150100708,
      "epoch": 0.3085002917720288,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994004035207274,
      "loss": 5.595,
      "mean_token_accuracy": 0.1584687501192093,
      "num_tokens": 6890411.0,
      "step": 3965
    },
    {
      "entropy": 5.800809621810913,
      "epoch": 0.30888932114374634,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993983797871356,
      "loss": 5.6356,
      "mean_token_accuracy": 0.152000392973423,
      "num_tokens": 6899201.0,
      "step": 3970
    },
    {
      "entropy": 5.80447564125061,
      "epoch": 0.30927835051546393,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993963526486555,
      "loss": 5.6534,
      "mean_token_accuracy": 0.15221117436885834,
      "num_tokens": 6907470.0,
      "step": 3975
    },
    {
      "entropy": 5.788397216796875,
      "epoch": 0.30966737988718146,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499394322105318,
      "loss": 5.6828,
      "mean_token_accuracy": 0.15586572289466857,
      "num_tokens": 6916542.0,
      "step": 3980
    },
    {
      "entropy": 5.851479148864746,
      "epoch": 0.31005640925889905,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993922881571538,
      "loss": 5.6814,
      "mean_token_accuracy": 0.15342268645763396,
      "num_tokens": 6925974.0,
      "step": 3985
    },
    {
      "entropy": 5.911538314819336,
      "epoch": 0.31044543863061663,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993902508041939,
      "loss": 5.703,
      "mean_token_accuracy": 0.14747922271490096,
      "num_tokens": 6934987.0,
      "step": 3990
    },
    {
      "entropy": 5.766879081726074,
      "epoch": 0.31083446800233416,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499388210046469,
      "loss": 5.5871,
      "mean_token_accuracy": 0.15272050797939302,
      "num_tokens": 6944497.0,
      "step": 3995
    },
    {
      "entropy": 5.847130870819091,
      "epoch": 0.31122349737405175,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993861658840102,
      "loss": 5.7243,
      "mean_token_accuracy": 0.14793727397918702,
      "num_tokens": 6953394.0,
      "step": 4000
    },
    {
      "entropy": 5.84558367729187,
      "epoch": 0.3116125267457693,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993841183168484,
      "loss": 5.6365,
      "mean_token_accuracy": 0.15578553080558777,
      "num_tokens": 6961715.0,
      "step": 4005
    },
    {
      "entropy": 5.784832239151001,
      "epoch": 0.31200155611748687,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993820673450148,
      "loss": 5.6423,
      "mean_token_accuracy": 0.15401849076151847,
      "num_tokens": 6970041.0,
      "step": 4010
    },
    {
      "entropy": 5.887779331207275,
      "epoch": 0.31239058548920445,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993800129685404,
      "loss": 5.7804,
      "mean_token_accuracy": 0.14733504951000215,
      "num_tokens": 6979522.0,
      "step": 4015
    },
    {
      "entropy": 5.884895706176758,
      "epoch": 0.312779614860922,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993779551874565,
      "loss": 5.7401,
      "mean_token_accuracy": 0.14512501955032348,
      "num_tokens": 6988470.0,
      "step": 4020
    },
    {
      "entropy": 5.913160419464111,
      "epoch": 0.31316864423263957,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993758940017943,
      "loss": 5.745,
      "mean_token_accuracy": 0.15878498256206514,
      "num_tokens": 6996696.0,
      "step": 4025
    },
    {
      "entropy": 5.890307474136352,
      "epoch": 0.31355767360435716,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499373829411585,
      "loss": 5.6215,
      "mean_token_accuracy": 0.15232147574424743,
      "num_tokens": 7005297.0,
      "step": 4030
    },
    {
      "entropy": 5.814219665527344,
      "epoch": 0.3139467029760747,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049937176141686,
      "loss": 5.6209,
      "mean_token_accuracy": 0.15302640944719315,
      "num_tokens": 7013569.0,
      "step": 4035
    },
    {
      "entropy": 5.770711660385132,
      "epoch": 0.3143357323477923,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993696900176506,
      "loss": 5.5739,
      "mean_token_accuracy": 0.15835325047373772,
      "num_tokens": 7021449.0,
      "step": 4040
    },
    {
      "entropy": 5.825836849212647,
      "epoch": 0.3147247617195098,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993676152139883,
      "loss": 5.6855,
      "mean_token_accuracy": 0.1571549579501152,
      "num_tokens": 7029663.0,
      "step": 4045
    },
    {
      "entropy": 5.803576135635376,
      "epoch": 0.3151137910912274,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993655370059046,
      "loss": 5.6877,
      "mean_token_accuracy": 0.154311640560627,
      "num_tokens": 7039038.0,
      "step": 4050
    },
    {
      "entropy": 5.823104667663574,
      "epoch": 0.315502820462945,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004993634553934309,
      "loss": 5.645,
      "mean_token_accuracy": 0.15580660626292228,
      "num_tokens": 7047595.0,
      "step": 4055
    },
    {
      "entropy": 5.835638093948364,
      "epoch": 0.3158918498346625,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993613703765989,
      "loss": 5.6582,
      "mean_token_accuracy": 0.15214158296585084,
      "num_tokens": 7055711.0,
      "step": 4060
    },
    {
      "entropy": 5.870501375198364,
      "epoch": 0.3162808792063801,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993592819554402,
      "loss": 5.6272,
      "mean_token_accuracy": 0.15837595313787461,
      "num_tokens": 7064100.0,
      "step": 4065
    },
    {
      "entropy": 5.80830135345459,
      "epoch": 0.3166699085780976,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993571901299866,
      "loss": 5.5916,
      "mean_token_accuracy": 0.15826652497053145,
      "num_tokens": 7073040.0,
      "step": 4070
    },
    {
      "entropy": 5.811197662353516,
      "epoch": 0.3170589379498152,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993550949002697,
      "loss": 5.6968,
      "mean_token_accuracy": 0.1517757922410965,
      "num_tokens": 7081513.0,
      "step": 4075
    },
    {
      "entropy": 5.772473526000977,
      "epoch": 0.3174479673215328,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993529962663213,
      "loss": 5.595,
      "mean_token_accuracy": 0.1613408699631691,
      "num_tokens": 7090704.0,
      "step": 4080
    },
    {
      "entropy": 5.818785381317139,
      "epoch": 0.31783699669325033,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004993508942281732,
      "loss": 5.7533,
      "mean_token_accuracy": 0.14853261411190033,
      "num_tokens": 7099692.0,
      "step": 4085
    },
    {
      "entropy": 5.7818577766418455,
      "epoch": 0.3182260260649679,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993487887858575,
      "loss": 5.5793,
      "mean_token_accuracy": 0.15307215079665185,
      "num_tokens": 7108021.0,
      "step": 4090
    },
    {
      "entropy": 5.80325174331665,
      "epoch": 0.31861505543668545,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499346679939406,
      "loss": 5.6545,
      "mean_token_accuracy": 0.1577788397669792,
      "num_tokens": 7116689.0,
      "step": 4095
    },
    {
      "entropy": 5.832878112792969,
      "epoch": 0.31900408480840303,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993445676888507,
      "loss": 5.7519,
      "mean_token_accuracy": 0.1521778553724289,
      "num_tokens": 7125453.0,
      "step": 4100
    },
    {
      "entropy": 5.942759323120117,
      "epoch": 0.3193931141801206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993424520342236,
      "loss": 5.7843,
      "mean_token_accuracy": 0.15119208842515947,
      "num_tokens": 7133692.0,
      "step": 4105
    },
    {
      "entropy": 5.860222673416137,
      "epoch": 0.31978214355183815,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004993403329755569,
      "loss": 5.5968,
      "mean_token_accuracy": 0.15934908539056777,
      "num_tokens": 7142362.0,
      "step": 4110
    },
    {
      "entropy": 5.746327972412109,
      "epoch": 0.32017117292355574,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993382105128828,
      "loss": 5.6226,
      "mean_token_accuracy": 0.15567729622125626,
      "num_tokens": 7151020.0,
      "step": 4115
    },
    {
      "entropy": 5.768520021438599,
      "epoch": 0.32056020229527327,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993360846462333,
      "loss": 5.5285,
      "mean_token_accuracy": 0.16255829334259034,
      "num_tokens": 7159453.0,
      "step": 4120
    },
    {
      "entropy": 5.889703607559204,
      "epoch": 0.32094923166699085,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004993339553756408,
      "loss": 5.7937,
      "mean_token_accuracy": 0.1460296593606472,
      "num_tokens": 7168758.0,
      "step": 4125
    },
    {
      "entropy": 5.824968242645264,
      "epoch": 0.32133826103870844,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004993318227011378,
      "loss": 5.6026,
      "mean_token_accuracy": 0.15972633063793182,
      "num_tokens": 7177835.0,
      "step": 4130
    },
    {
      "entropy": 5.817380714416504,
      "epoch": 0.32172729041042597,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993296866227562,
      "loss": 5.6396,
      "mean_token_accuracy": 0.16363780200481415,
      "num_tokens": 7186424.0,
      "step": 4135
    },
    {
      "entropy": 5.817440557479858,
      "epoch": 0.32211631978214356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993275471405288,
      "loss": 5.6284,
      "mean_token_accuracy": 0.16064855307340623,
      "num_tokens": 7194800.0,
      "step": 4140
    },
    {
      "entropy": 5.759157609939575,
      "epoch": 0.32250534915386114,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993254042544879,
      "loss": 5.5543,
      "mean_token_accuracy": 0.16035097762942313,
      "num_tokens": 7203820.0,
      "step": 4145
    },
    {
      "entropy": 5.781700611114502,
      "epoch": 0.3228943785255787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499323257964666,
      "loss": 5.5454,
      "mean_token_accuracy": 0.1592766359448433,
      "num_tokens": 7211793.0,
      "step": 4150
    },
    {
      "entropy": 5.802039670944214,
      "epoch": 0.32328340789729626,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004993211082710957,
      "loss": 5.6027,
      "mean_token_accuracy": 0.1580141991376877,
      "num_tokens": 7220471.0,
      "step": 4155
    },
    {
      "entropy": 5.803948020935058,
      "epoch": 0.3236724372690138,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993189551738097,
      "loss": 5.6609,
      "mean_token_accuracy": 0.1602739378809929,
      "num_tokens": 7228788.0,
      "step": 4160
    },
    {
      "entropy": 5.875597524642944,
      "epoch": 0.3240614666407314,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993167986728405,
      "loss": 5.8232,
      "mean_token_accuracy": 0.15174989998340607,
      "num_tokens": 7237707.0,
      "step": 4165
    },
    {
      "entropy": 5.858319091796875,
      "epoch": 0.32445049601244896,
      "grad_norm": 1.0,
      "learning_rate": 0.000499314638768221,
      "loss": 5.5789,
      "mean_token_accuracy": 0.1572333574295044,
      "num_tokens": 7246296.0,
      "step": 4170
    },
    {
      "entropy": 5.758087635040283,
      "epoch": 0.3248395253841665,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993124754599839,
      "loss": 5.5103,
      "mean_token_accuracy": 0.16214703321456908,
      "num_tokens": 7255076.0,
      "step": 4175
    },
    {
      "entropy": 5.6478697776794435,
      "epoch": 0.3252285547558841,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993103087481619,
      "loss": 5.599,
      "mean_token_accuracy": 0.15897218883037567,
      "num_tokens": 7263169.0,
      "step": 4180
    },
    {
      "entropy": 5.875939607620239,
      "epoch": 0.3256175841276016,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993081386327881,
      "loss": 5.6549,
      "mean_token_accuracy": 0.1627318650484085,
      "num_tokens": 7271748.0,
      "step": 4185
    },
    {
      "entropy": 5.7767092227935795,
      "epoch": 0.3260066134993192,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004993059651138953,
      "loss": 5.4638,
      "mean_token_accuracy": 0.1650639668107033,
      "num_tokens": 7280291.0,
      "step": 4190
    },
    {
      "entropy": 5.714006567001343,
      "epoch": 0.3263956428710368,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993037881915165,
      "loss": 5.6706,
      "mean_token_accuracy": 0.15370635092258453,
      "num_tokens": 7290307.0,
      "step": 4195
    },
    {
      "entropy": 5.853940105438232,
      "epoch": 0.3267846722427543,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993016078656847,
      "loss": 5.754,
      "mean_token_accuracy": 0.1487473540008068,
      "num_tokens": 7297942.0,
      "step": 4200
    },
    {
      "entropy": 5.7838770866394045,
      "epoch": 0.3271737016144719,
      "grad_norm": 1.0,
      "learning_rate": 0.000499299424136433,
      "loss": 5.6306,
      "mean_token_accuracy": 0.15384245216846465,
      "num_tokens": 7307378.0,
      "step": 4205
    },
    {
      "entropy": 5.778017473220825,
      "epoch": 0.32756273098618943,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992972370037946,
      "loss": 5.5811,
      "mean_token_accuracy": 0.1592706561088562,
      "num_tokens": 7315985.0,
      "step": 4210
    },
    {
      "entropy": 5.843814849853516,
      "epoch": 0.327951760357907,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992950464678026,
      "loss": 5.691,
      "mean_token_accuracy": 0.1587142139673233,
      "num_tokens": 7324142.0,
      "step": 4215
    },
    {
      "entropy": 5.740630483627319,
      "epoch": 0.3283407897296246,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992928525284903,
      "loss": 5.5362,
      "mean_token_accuracy": 0.16127856969833373,
      "num_tokens": 7333012.0,
      "step": 4220
    },
    {
      "entropy": 5.765237712860108,
      "epoch": 0.32872981910134214,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499290655185891,
      "loss": 5.5906,
      "mean_token_accuracy": 0.160532608628273,
      "num_tokens": 7341838.0,
      "step": 4225
    },
    {
      "entropy": 5.7777024269104,
      "epoch": 0.3291188484730597,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499288454440038,
      "loss": 5.5255,
      "mean_token_accuracy": 0.17211012840270995,
      "num_tokens": 7349986.0,
      "step": 4230
    },
    {
      "entropy": 5.856859874725342,
      "epoch": 0.3295078778447773,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992862502909645,
      "loss": 5.7791,
      "mean_token_accuracy": 0.14971803575754167,
      "num_tokens": 7359428.0,
      "step": 4235
    },
    {
      "entropy": 5.868747282028198,
      "epoch": 0.32989690721649484,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992840427387043,
      "loss": 5.6902,
      "mean_token_accuracy": 0.15117943584918975,
      "num_tokens": 7368524.0,
      "step": 4240
    },
    {
      "entropy": 5.838213348388672,
      "epoch": 0.3302859365882124,
      "grad_norm": 0.875,
      "learning_rate": 0.0004992818317832908,
      "loss": 5.5728,
      "mean_token_accuracy": 0.1599339485168457,
      "num_tokens": 7377979.0,
      "step": 4245
    },
    {
      "entropy": 5.684649276733398,
      "epoch": 0.33067496595992996,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992796174247574,
      "loss": 5.4922,
      "mean_token_accuracy": 0.15754138231277465,
      "num_tokens": 7386283.0,
      "step": 4250
    },
    {
      "entropy": 5.7293774604797365,
      "epoch": 0.33106399533164754,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004992773996631378,
      "loss": 5.6276,
      "mean_token_accuracy": 0.15059794411063193,
      "num_tokens": 7395409.0,
      "step": 4255
    },
    {
      "entropy": 5.841209554672242,
      "epoch": 0.33145302470336513,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992751784984656,
      "loss": 5.705,
      "mean_token_accuracy": 0.1574711725115776,
      "num_tokens": 7403389.0,
      "step": 4260
    },
    {
      "entropy": 5.814565706253052,
      "epoch": 0.33184205407508266,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992729539307745,
      "loss": 5.6165,
      "mean_token_accuracy": 0.15815107226371766,
      "num_tokens": 7411506.0,
      "step": 4265
    },
    {
      "entropy": 5.849872016906739,
      "epoch": 0.33223108344680025,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992707259600984,
      "loss": 5.7025,
      "mean_token_accuracy": 0.15787684619426728,
      "num_tokens": 7420219.0,
      "step": 4270
    },
    {
      "entropy": 5.829010963439941,
      "epoch": 0.3326201128185178,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992684945864709,
      "loss": 5.6124,
      "mean_token_accuracy": 0.16116924285888673,
      "num_tokens": 7428442.0,
      "step": 4275
    },
    {
      "entropy": 5.769632053375244,
      "epoch": 0.33300914219023536,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992662598099261,
      "loss": 5.5713,
      "mean_token_accuracy": 0.1633827120065689,
      "num_tokens": 7436364.0,
      "step": 4280
    },
    {
      "entropy": 5.811918067932129,
      "epoch": 0.33339817156195295,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992640216304975,
      "loss": 5.6446,
      "mean_token_accuracy": 0.14872778430581093,
      "num_tokens": 7444122.0,
      "step": 4285
    },
    {
      "entropy": 5.867029237747192,
      "epoch": 0.3337872009336705,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992617800482196,
      "loss": 5.6601,
      "mean_token_accuracy": 0.15238194465637206,
      "num_tokens": 7454122.0,
      "step": 4290
    },
    {
      "entropy": 5.827371215820312,
      "epoch": 0.33417623030538807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499259535063126,
      "loss": 5.6652,
      "mean_token_accuracy": 0.16036169081926346,
      "num_tokens": 7463057.0,
      "step": 4295
    },
    {
      "entropy": 5.80002613067627,
      "epoch": 0.3345652596771056,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499257286675251,
      "loss": 5.6295,
      "mean_token_accuracy": 0.1520880788564682,
      "num_tokens": 7471675.0,
      "step": 4300
    },
    {
      "entropy": 5.708404207229615,
      "epoch": 0.3349542890488232,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004992550348846285,
      "loss": 5.5731,
      "mean_token_accuracy": 0.15868975967168808,
      "num_tokens": 7479870.0,
      "step": 4305
    },
    {
      "entropy": 5.807326030731201,
      "epoch": 0.33534331842054077,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992527796912928,
      "loss": 5.6442,
      "mean_token_accuracy": 0.1565382771193981,
      "num_tokens": 7489830.0,
      "step": 4310
    },
    {
      "entropy": 5.791453504562378,
      "epoch": 0.3357323477922583,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992505210952782,
      "loss": 5.5757,
      "mean_token_accuracy": 0.1656323567032814,
      "num_tokens": 7497852.0,
      "step": 4315
    },
    {
      "entropy": 5.785038661956787,
      "epoch": 0.3361213771639759,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992482590966188,
      "loss": 5.7069,
      "mean_token_accuracy": 0.1559569127857685,
      "num_tokens": 7506419.0,
      "step": 4320
    },
    {
      "entropy": 5.837784051895142,
      "epoch": 0.3365104065356934,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004992459936953489,
      "loss": 5.5972,
      "mean_token_accuracy": 0.16566000878810883,
      "num_tokens": 7514757.0,
      "step": 4325
    },
    {
      "entropy": 5.790379810333252,
      "epoch": 0.336899435907411,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992437248915032,
      "loss": 5.664,
      "mean_token_accuracy": 0.15615198463201524,
      "num_tokens": 7523912.0,
      "step": 4330
    },
    {
      "entropy": 5.751581430435181,
      "epoch": 0.3372884652791286,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992414526851157,
      "loss": 5.5474,
      "mean_token_accuracy": 0.16585139334201812,
      "num_tokens": 7532401.0,
      "step": 4335
    },
    {
      "entropy": 5.669851446151734,
      "epoch": 0.3376774946508461,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992391770762212,
      "loss": 5.5144,
      "mean_token_accuracy": 0.16344813257455826,
      "num_tokens": 7541257.0,
      "step": 4340
    },
    {
      "entropy": 5.706036615371704,
      "epoch": 0.3380665240225637,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499236898064854,
      "loss": 5.4562,
      "mean_token_accuracy": 0.15973441749811174,
      "num_tokens": 7549812.0,
      "step": 4345
    },
    {
      "entropy": 5.743966579437256,
      "epoch": 0.3384555533942813,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992346156510488,
      "loss": 5.5513,
      "mean_token_accuracy": 0.15965283364057542,
      "num_tokens": 7559319.0,
      "step": 4350
    },
    {
      "entropy": 5.813631963729859,
      "epoch": 0.3388445827659988,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992323298348403,
      "loss": 5.6947,
      "mean_token_accuracy": 0.1522468313574791,
      "num_tokens": 7567900.0,
      "step": 4355
    },
    {
      "entropy": 5.833976316452026,
      "epoch": 0.3392336121377164,
      "grad_norm": 1.0,
      "learning_rate": 0.000499230040616263,
      "loss": 5.6194,
      "mean_token_accuracy": 0.1545071080327034,
      "num_tokens": 7576163.0,
      "step": 4360
    },
    {
      "entropy": 5.769789743423462,
      "epoch": 0.33962264150943394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992277479953518,
      "loss": 5.5671,
      "mean_token_accuracy": 0.15694677829742432,
      "num_tokens": 7584221.0,
      "step": 4365
    },
    {
      "entropy": 5.700899267196656,
      "epoch": 0.34001167088115153,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004992254519721414,
      "loss": 5.5473,
      "mean_token_accuracy": 0.15996042490005494,
      "num_tokens": 7593569.0,
      "step": 4370
    },
    {
      "entropy": 5.7665114402771,
      "epoch": 0.3404007002528691,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992231525466666,
      "loss": 5.6089,
      "mean_token_accuracy": 0.16032602339982988,
      "num_tokens": 7602666.0,
      "step": 4375
    },
    {
      "entropy": 5.74814510345459,
      "epoch": 0.34078972962458665,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992208497189624,
      "loss": 5.6791,
      "mean_token_accuracy": 0.15110585689544678,
      "num_tokens": 7611874.0,
      "step": 4380
    },
    {
      "entropy": 5.841089820861816,
      "epoch": 0.34117875899630423,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992185434890637,
      "loss": 5.6336,
      "mean_token_accuracy": 0.15836991518735885,
      "num_tokens": 7620385.0,
      "step": 4385
    },
    {
      "entropy": 5.894673633575439,
      "epoch": 0.34156778836802176,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992162338570055,
      "loss": 5.6854,
      "mean_token_accuracy": 0.148434978723526,
      "num_tokens": 7628786.0,
      "step": 4390
    },
    {
      "entropy": 5.769052648544312,
      "epoch": 0.34195681773973935,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004992139208228227,
      "loss": 5.5032,
      "mean_token_accuracy": 0.16336381584405898,
      "num_tokens": 7637389.0,
      "step": 4395
    },
    {
      "entropy": 5.750074100494385,
      "epoch": 0.34234584711145694,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992116043865506,
      "loss": 5.6689,
      "mean_token_accuracy": 0.15916045010089874,
      "num_tokens": 7645793.0,
      "step": 4400
    },
    {
      "entropy": 5.878291654586792,
      "epoch": 0.34273487648317447,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992092845482244,
      "loss": 5.6357,
      "mean_token_accuracy": 0.16118928492069245,
      "num_tokens": 7653951.0,
      "step": 4405
    },
    {
      "entropy": 5.724064445495605,
      "epoch": 0.34312390585489205,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992069613078791,
      "loss": 5.4944,
      "mean_token_accuracy": 0.16808485239744186,
      "num_tokens": 7662753.0,
      "step": 4410
    },
    {
      "entropy": 5.686445617675782,
      "epoch": 0.3435129352266096,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992046346655499,
      "loss": 5.4727,
      "mean_token_accuracy": 0.16301707774400712,
      "num_tokens": 7671832.0,
      "step": 4415
    },
    {
      "entropy": 5.801256561279297,
      "epoch": 0.34390196459832717,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992023046212723,
      "loss": 5.6282,
      "mean_token_accuracy": 0.1588641732931137,
      "num_tokens": 7680969.0,
      "step": 4420
    },
    {
      "entropy": 5.8781603336334225,
      "epoch": 0.34429099397004476,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004991999711750816,
      "loss": 5.6316,
      "mean_token_accuracy": 0.15277016907930374,
      "num_tokens": 7689277.0,
      "step": 4425
    },
    {
      "entropy": 5.728748893737793,
      "epoch": 0.3446800233417623,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991976343270133,
      "loss": 5.5592,
      "mean_token_accuracy": 0.1625969737768173,
      "num_tokens": 7697373.0,
      "step": 4430
    },
    {
      "entropy": 5.857252168655395,
      "epoch": 0.3450690527134799,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991952940771026,
      "loss": 5.6249,
      "mean_token_accuracy": 0.15589040964841844,
      "num_tokens": 7705767.0,
      "step": 4435
    },
    {
      "entropy": 5.715005779266358,
      "epoch": 0.34545808208519746,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991929504253852,
      "loss": 5.5207,
      "mean_token_accuracy": 0.1657451167702675,
      "num_tokens": 7713746.0,
      "step": 4440
    },
    {
      "entropy": 5.747983598709107,
      "epoch": 0.345847111456915,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991906033718966,
      "loss": 5.66,
      "mean_token_accuracy": 0.15185474902391433,
      "num_tokens": 7721829.0,
      "step": 4445
    },
    {
      "entropy": 5.7072593688964846,
      "epoch": 0.3462361408286326,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004991882529166724,
      "loss": 5.5254,
      "mean_token_accuracy": 0.1630813479423523,
      "num_tokens": 7730479.0,
      "step": 4450
    },
    {
      "entropy": 5.768003606796265,
      "epoch": 0.3466251702003501,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991858990597483,
      "loss": 5.4953,
      "mean_token_accuracy": 0.16315876841545104,
      "num_tokens": 7738659.0,
      "step": 4455
    },
    {
      "entropy": 5.835393476486206,
      "epoch": 0.3470141995720677,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991835418011601,
      "loss": 5.7726,
      "mean_token_accuracy": 0.15530015379190446,
      "num_tokens": 7747597.0,
      "step": 4460
    },
    {
      "entropy": 5.778534841537476,
      "epoch": 0.3474032289437853,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004991811811409434,
      "loss": 5.6145,
      "mean_token_accuracy": 0.1572774849832058,
      "num_tokens": 7756666.0,
      "step": 4465
    },
    {
      "entropy": 5.8093178272247314,
      "epoch": 0.3477922583155028,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991788170791341,
      "loss": 5.6412,
      "mean_token_accuracy": 0.15569517463445665,
      "num_tokens": 7765621.0,
      "step": 4470
    },
    {
      "entropy": 5.762197351455688,
      "epoch": 0.3481812876872204,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991764496157682,
      "loss": 5.5449,
      "mean_token_accuracy": 0.15648710131645202,
      "num_tokens": 7774137.0,
      "step": 4475
    },
    {
      "entropy": 5.828894186019897,
      "epoch": 0.34857031705893793,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004991740787508814,
      "loss": 5.6492,
      "mean_token_accuracy": 0.15211960375308992,
      "num_tokens": 7783144.0,
      "step": 4480
    },
    {
      "entropy": 5.78465690612793,
      "epoch": 0.3489593464306555,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991717044845097,
      "loss": 5.6866,
      "mean_token_accuracy": 0.14937788471579552,
      "num_tokens": 7791530.0,
      "step": 4485
    },
    {
      "entropy": 5.761826896667481,
      "epoch": 0.3493483758023731,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991693268166892,
      "loss": 5.5461,
      "mean_token_accuracy": 0.1595842033624649,
      "num_tokens": 7800869.0,
      "step": 4490
    },
    {
      "entropy": 5.781609058380127,
      "epoch": 0.34973740517409063,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991669457474559,
      "loss": 5.6224,
      "mean_token_accuracy": 0.15640588253736495,
      "num_tokens": 7810201.0,
      "step": 4495
    },
    {
      "entropy": 5.749261569976807,
      "epoch": 0.3501264345458082,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991645612768461,
      "loss": 5.5152,
      "mean_token_accuracy": 0.1649569660425186,
      "num_tokens": 7818529.0,
      "step": 4500
    },
    {
      "entropy": 5.7531813621521,
      "epoch": 0.35051546391752575,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991621734048958,
      "loss": 5.5601,
      "mean_token_accuracy": 0.16200138330459596,
      "num_tokens": 7827254.0,
      "step": 4505
    },
    {
      "entropy": 5.77031102180481,
      "epoch": 0.35090449328924334,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991597821316415,
      "loss": 5.5726,
      "mean_token_accuracy": 0.16260626912117004,
      "num_tokens": 7836437.0,
      "step": 4510
    },
    {
      "entropy": 5.717511510848999,
      "epoch": 0.3512935226609609,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991573874571192,
      "loss": 5.5881,
      "mean_token_accuracy": 0.16121748238801956,
      "num_tokens": 7844939.0,
      "step": 4515
    },
    {
      "entropy": 5.755042934417725,
      "epoch": 0.35168255203267845,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991549893813653,
      "loss": 5.5594,
      "mean_token_accuracy": 0.1579587161540985,
      "num_tokens": 7852889.0,
      "step": 4520
    },
    {
      "entropy": 5.809964561462403,
      "epoch": 0.35207158140439604,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991525879044162,
      "loss": 5.7009,
      "mean_token_accuracy": 0.1521313928067684,
      "num_tokens": 7861867.0,
      "step": 4525
    },
    {
      "entropy": 5.759050130844116,
      "epoch": 0.35246061077611357,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991501830263084,
      "loss": 5.4925,
      "mean_token_accuracy": 0.16492490023374556,
      "num_tokens": 7869343.0,
      "step": 4530
    },
    {
      "entropy": 5.743049049377442,
      "epoch": 0.35284964014783116,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991477747470783,
      "loss": 5.5496,
      "mean_token_accuracy": 0.16336526572704316,
      "num_tokens": 7877142.0,
      "step": 4535
    },
    {
      "entropy": 5.715654516220093,
      "epoch": 0.35323866951954874,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991453630667625,
      "loss": 5.5718,
      "mean_token_accuracy": 0.1644924059510231,
      "num_tokens": 7885183.0,
      "step": 4540
    },
    {
      "entropy": 5.660160732269287,
      "epoch": 0.3536276988912663,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991429479853976,
      "loss": 5.4073,
      "mean_token_accuracy": 0.17003071159124375,
      "num_tokens": 7893848.0,
      "step": 4545
    },
    {
      "entropy": 5.752790403366089,
      "epoch": 0.35401672826298386,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991405295030202,
      "loss": 5.5774,
      "mean_token_accuracy": 0.15930684059858322,
      "num_tokens": 7902438.0,
      "step": 4550
    },
    {
      "entropy": 5.7751688957214355,
      "epoch": 0.35440575763470145,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991381076196671,
      "loss": 5.543,
      "mean_token_accuracy": 0.15595637112855912,
      "num_tokens": 7910951.0,
      "step": 4555
    },
    {
      "entropy": 5.748727226257325,
      "epoch": 0.354794787006419,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991356823353748,
      "loss": 5.6526,
      "mean_token_accuracy": 0.1568855568766594,
      "num_tokens": 7920375.0,
      "step": 4560
    },
    {
      "entropy": 5.702395009994507,
      "epoch": 0.35518381637813656,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991332536501804,
      "loss": 5.5248,
      "mean_token_accuracy": 0.1637159138917923,
      "num_tokens": 7928656.0,
      "step": 4565
    },
    {
      "entropy": 5.88414044380188,
      "epoch": 0.3555728457498541,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991308215641205,
      "loss": 5.7693,
      "mean_token_accuracy": 0.15237502455711366,
      "num_tokens": 7936929.0,
      "step": 4570
    },
    {
      "entropy": 5.835222911834717,
      "epoch": 0.3559618751215717,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991283860772323,
      "loss": 5.5781,
      "mean_token_accuracy": 0.16577111184597015,
      "num_tokens": 7945592.0,
      "step": 4575
    },
    {
      "entropy": 5.725835466384888,
      "epoch": 0.35635090449328927,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991259471895525,
      "loss": 5.5138,
      "mean_token_accuracy": 0.16628557145595552,
      "num_tokens": 7953656.0,
      "step": 4580
    },
    {
      "entropy": 5.777119445800781,
      "epoch": 0.3567399338650068,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991235049011182,
      "loss": 5.5921,
      "mean_token_accuracy": 0.16313516050577165,
      "num_tokens": 7962038.0,
      "step": 4585
    },
    {
      "entropy": 5.779297161102295,
      "epoch": 0.3571289632367244,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991210592119663,
      "loss": 5.673,
      "mean_token_accuracy": 0.1561742752790451,
      "num_tokens": 7970338.0,
      "step": 4590
    },
    {
      "entropy": 5.824539995193481,
      "epoch": 0.3575179926084419,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991186101221342,
      "loss": 5.5993,
      "mean_token_accuracy": 0.16285818666219712,
      "num_tokens": 7979173.0,
      "step": 4595
    },
    {
      "entropy": 5.753220415115356,
      "epoch": 0.3579070219801595,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499116157631659,
      "loss": 5.4754,
      "mean_token_accuracy": 0.16968862116336822,
      "num_tokens": 7987245.0,
      "step": 4600
    },
    {
      "entropy": 5.782096433639526,
      "epoch": 0.3582960513518771,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991137017405777,
      "loss": 5.6827,
      "mean_token_accuracy": 0.15123804807662963,
      "num_tokens": 7995551.0,
      "step": 4605
    },
    {
      "entropy": 5.699934291839599,
      "epoch": 0.3586850807235946,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004991112424489277,
      "loss": 5.4975,
      "mean_token_accuracy": 0.16038129031658171,
      "num_tokens": 8004069.0,
      "step": 4610
    },
    {
      "entropy": 5.725274276733399,
      "epoch": 0.3590741100953122,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991087797567462,
      "loss": 5.5675,
      "mean_token_accuracy": 0.16466078162193298,
      "num_tokens": 8012537.0,
      "step": 4615
    },
    {
      "entropy": 5.838127136230469,
      "epoch": 0.35946313946702974,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991063136640709,
      "loss": 5.5434,
      "mean_token_accuracy": 0.15995005071163176,
      "num_tokens": 8021387.0,
      "step": 4620
    },
    {
      "entropy": 5.732734203338623,
      "epoch": 0.3598521688387473,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991038441709388,
      "loss": 5.499,
      "mean_token_accuracy": 0.1588419958949089,
      "num_tokens": 8030261.0,
      "step": 4625
    },
    {
      "entropy": 5.771962738037109,
      "epoch": 0.3602411982104649,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991013712773877,
      "loss": 5.581,
      "mean_token_accuracy": 0.15491561591625214,
      "num_tokens": 8038161.0,
      "step": 4630
    },
    {
      "entropy": 5.7120036602020265,
      "epoch": 0.36063022758218244,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990988949834549,
      "loss": 5.6502,
      "mean_token_accuracy": 0.16104790568351746,
      "num_tokens": 8046340.0,
      "step": 4635
    },
    {
      "entropy": 5.80100302696228,
      "epoch": 0.3610192569539,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499096415289178,
      "loss": 5.5868,
      "mean_token_accuracy": 0.1563143476843834,
      "num_tokens": 8054997.0,
      "step": 4640
    },
    {
      "entropy": 5.75185375213623,
      "epoch": 0.3614082863256176,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004990939321945948,
      "loss": 5.5207,
      "mean_token_accuracy": 0.16830721646547317,
      "num_tokens": 8063686.0,
      "step": 4645
    },
    {
      "entropy": 5.738094854354858,
      "epoch": 0.36179731569733514,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990914456997427,
      "loss": 5.6455,
      "mean_token_accuracy": 0.15778837502002716,
      "num_tokens": 8072999.0,
      "step": 4650
    },
    {
      "entropy": 5.734006452560425,
      "epoch": 0.36218634506905273,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990889558046598,
      "loss": 5.4954,
      "mean_token_accuracy": 0.1635401204228401,
      "num_tokens": 8081596.0,
      "step": 4655
    },
    {
      "entropy": 5.792643308639526,
      "epoch": 0.36257537444077026,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990864625093836,
      "loss": 5.7211,
      "mean_token_accuracy": 0.1504440724849701,
      "num_tokens": 8090948.0,
      "step": 4660
    },
    {
      "entropy": 5.738381576538086,
      "epoch": 0.36296440381248785,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990839658139519,
      "loss": 5.7118,
      "mean_token_accuracy": 0.15039844661951066,
      "num_tokens": 8099445.0,
      "step": 4665
    },
    {
      "entropy": 5.817491865158081,
      "epoch": 0.36335343318420543,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004990814657184028,
      "loss": 5.5696,
      "mean_token_accuracy": 0.15574323982000352,
      "num_tokens": 8108119.0,
      "step": 4670
    },
    {
      "entropy": 5.782070636749268,
      "epoch": 0.36374246255592296,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004990789622227741,
      "loss": 5.5735,
      "mean_token_accuracy": 0.15753818452358245,
      "num_tokens": 8117060.0,
      "step": 4675
    },
    {
      "entropy": 5.714137172698974,
      "epoch": 0.36413149192764055,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990764553271038,
      "loss": 5.5652,
      "mean_token_accuracy": 0.15798794180154802,
      "num_tokens": 8126229.0,
      "step": 4680
    },
    {
      "entropy": 5.788286256790161,
      "epoch": 0.3645205212993581,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990739450314299,
      "loss": 5.5223,
      "mean_token_accuracy": 0.16824372112751007,
      "num_tokens": 8134694.0,
      "step": 4685
    },
    {
      "entropy": 5.772018814086914,
      "epoch": 0.36490955067107567,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004990714313357906,
      "loss": 5.5969,
      "mean_token_accuracy": 0.1634421780705452,
      "num_tokens": 8143329.0,
      "step": 4690
    },
    {
      "entropy": 5.750579977035523,
      "epoch": 0.36529858004279325,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499068914240224,
      "loss": 5.5701,
      "mean_token_accuracy": 0.16435180157423018,
      "num_tokens": 8151432.0,
      "step": 4695
    },
    {
      "entropy": 5.76611328125,
      "epoch": 0.3656876094145108,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990663937447682,
      "loss": 5.5442,
      "mean_token_accuracy": 0.16045622676610946,
      "num_tokens": 8159677.0,
      "step": 4700
    },
    {
      "entropy": 5.727065992355347,
      "epoch": 0.36607663878622837,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990638698494615,
      "loss": 5.6065,
      "mean_token_accuracy": 0.15686380863189697,
      "num_tokens": 8168710.0,
      "step": 4705
    },
    {
      "entropy": 5.742712450027466,
      "epoch": 0.3664656681579459,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990613425543423,
      "loss": 5.5955,
      "mean_token_accuracy": 0.1621291607618332,
      "num_tokens": 8177501.0,
      "step": 4710
    },
    {
      "entropy": 5.843996238708496,
      "epoch": 0.3668546975296635,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990588118594487,
      "loss": 5.6682,
      "mean_token_accuracy": 0.15319716781377793,
      "num_tokens": 8185791.0,
      "step": 4715
    },
    {
      "entropy": 5.831458473205567,
      "epoch": 0.3672437269013811,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990562777648194,
      "loss": 5.6938,
      "mean_token_accuracy": 0.15462879985570907,
      "num_tokens": 8195020.0,
      "step": 4720
    },
    {
      "entropy": 5.756261014938355,
      "epoch": 0.3676327562730986,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990537402704928,
      "loss": 5.5623,
      "mean_token_accuracy": 0.15908532589673996,
      "num_tokens": 8203540.0,
      "step": 4725
    },
    {
      "entropy": 5.799097156524658,
      "epoch": 0.3680217856448162,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004990511993765071,
      "loss": 5.5506,
      "mean_token_accuracy": 0.15895125716924668,
      "num_tokens": 8211985.0,
      "step": 4730
    },
    {
      "entropy": 5.733615446090698,
      "epoch": 0.3684108150165337,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990486550829011,
      "loss": 5.6262,
      "mean_token_accuracy": 0.1645650178194046,
      "num_tokens": 8220289.0,
      "step": 4735
    },
    {
      "entropy": 5.8451371669769285,
      "epoch": 0.3687998443882513,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990461073897134,
      "loss": 5.6569,
      "mean_token_accuracy": 0.14948247075080873,
      "num_tokens": 8229652.0,
      "step": 4740
    },
    {
      "entropy": 5.806291913986206,
      "epoch": 0.3691888737599689,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990435562969827,
      "loss": 5.6478,
      "mean_token_accuracy": 0.15665459036827087,
      "num_tokens": 8237728.0,
      "step": 4745
    },
    {
      "entropy": 5.7270866394042965,
      "epoch": 0.3695779031316864,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990410018047475,
      "loss": 5.5425,
      "mean_token_accuracy": 0.16309150755405427,
      "num_tokens": 8246393.0,
      "step": 4750
    },
    {
      "entropy": 5.695544958114624,
      "epoch": 0.369966932503404,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004990384439130467,
      "loss": 5.4984,
      "mean_token_accuracy": 0.16200296729803085,
      "num_tokens": 8256736.0,
      "step": 4755
    },
    {
      "entropy": 5.77980260848999,
      "epoch": 0.3703559618751216,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004990358826219192,
      "loss": 5.5975,
      "mean_token_accuracy": 0.16518477648496627,
      "num_tokens": 8265126.0,
      "step": 4760
    },
    {
      "entropy": 5.778947877883911,
      "epoch": 0.37074499124683913,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990333179314038,
      "loss": 5.4984,
      "mean_token_accuracy": 0.16346690207719802,
      "num_tokens": 8273136.0,
      "step": 4765
    },
    {
      "entropy": 5.620182704925537,
      "epoch": 0.3711340206185567,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990307498415393,
      "loss": 5.4276,
      "mean_token_accuracy": 0.16780980080366134,
      "num_tokens": 8281697.0,
      "step": 4770
    },
    {
      "entropy": 5.665127325057983,
      "epoch": 0.37152304999027425,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990281783523648,
      "loss": 5.5628,
      "mean_token_accuracy": 0.16574217826128007,
      "num_tokens": 8290765.0,
      "step": 4775
    },
    {
      "entropy": 5.823178434371949,
      "epoch": 0.37191207936199183,
      "grad_norm": 1.0,
      "learning_rate": 0.0004990256034639192,
      "loss": 5.6786,
      "mean_token_accuracy": 0.15433327853679657,
      "num_tokens": 8299875.0,
      "step": 4780
    },
    {
      "entropy": 5.750207090377808,
      "epoch": 0.3723011087337094,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990230251762418,
      "loss": 5.5241,
      "mean_token_accuracy": 0.16539545208215714,
      "num_tokens": 8308515.0,
      "step": 4785
    },
    {
      "entropy": 5.682763338088989,
      "epoch": 0.37269013810542695,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990204434893713,
      "loss": 5.4468,
      "mean_token_accuracy": 0.16895510554313659,
      "num_tokens": 8317260.0,
      "step": 4790
    },
    {
      "entropy": 5.7800510883331295,
      "epoch": 0.37307916747714454,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990178584033474,
      "loss": 5.6156,
      "mean_token_accuracy": 0.1566510796546936,
      "num_tokens": 8325887.0,
      "step": 4795
    },
    {
      "entropy": 5.842481851577759,
      "epoch": 0.37346819684886207,
      "grad_norm": 1.0,
      "learning_rate": 0.0004990152699182089,
      "loss": 5.6152,
      "mean_token_accuracy": 0.15425875931978225,
      "num_tokens": 8333974.0,
      "step": 4800
    },
    {
      "entropy": 5.722154998779297,
      "epoch": 0.37385722622057965,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004990126780339953,
      "loss": 5.5918,
      "mean_token_accuracy": 0.15548155158758165,
      "num_tokens": 8343199.0,
      "step": 4805
    },
    {
      "entropy": 5.685928583145142,
      "epoch": 0.37424625559229724,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990100827507459,
      "loss": 5.3899,
      "mean_token_accuracy": 0.16845352202653885,
      "num_tokens": 8351202.0,
      "step": 4810
    },
    {
      "entropy": 5.811168527603149,
      "epoch": 0.37463528496401477,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990074840684999,
      "loss": 5.7413,
      "mean_token_accuracy": 0.14929891377687454,
      "num_tokens": 8360011.0,
      "step": 4815
    },
    {
      "entropy": 5.75562105178833,
      "epoch": 0.37502431433573236,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990048819872969,
      "loss": 5.6051,
      "mean_token_accuracy": 0.16131770461797715,
      "num_tokens": 8367874.0,
      "step": 4820
    },
    {
      "entropy": 5.723214483261108,
      "epoch": 0.3754133437074499,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990022765071765,
      "loss": 5.57,
      "mean_token_accuracy": 0.15871281325817108,
      "num_tokens": 8376708.0,
      "step": 4825
    },
    {
      "entropy": 5.7772377014160154,
      "epoch": 0.3758023730791675,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000498999667628178,
      "loss": 5.5918,
      "mean_token_accuracy": 0.16533441841602325,
      "num_tokens": 8385467.0,
      "step": 4830
    },
    {
      "entropy": 5.749735116958618,
      "epoch": 0.37619140245088506,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989970553503411,
      "loss": 5.5598,
      "mean_token_accuracy": 0.15902363508939743,
      "num_tokens": 8393574.0,
      "step": 4835
    },
    {
      "entropy": 5.69984188079834,
      "epoch": 0.3765804318226026,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004989944396737054,
      "loss": 5.6006,
      "mean_token_accuracy": 0.16227839142084122,
      "num_tokens": 8402857.0,
      "step": 4840
    },
    {
      "entropy": 5.737270069122315,
      "epoch": 0.3769694611943202,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989918205983106,
      "loss": 5.599,
      "mean_token_accuracy": 0.16925140619277954,
      "num_tokens": 8411728.0,
      "step": 4845
    },
    {
      "entropy": 5.779333019256592,
      "epoch": 0.37735849056603776,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004989891981241964,
      "loss": 5.5787,
      "mean_token_accuracy": 0.1665400505065918,
      "num_tokens": 8420307.0,
      "step": 4850
    },
    {
      "entropy": 5.787651681900025,
      "epoch": 0.3777475199377553,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004989865722514027,
      "loss": 5.5925,
      "mean_token_accuracy": 0.15569850653409958,
      "num_tokens": 8429442.0,
      "step": 4855
    },
    {
      "entropy": 5.721146774291992,
      "epoch": 0.3781365493094729,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989839429799692,
      "loss": 5.4756,
      "mean_token_accuracy": 0.17139440774917603,
      "num_tokens": 8437945.0,
      "step": 4860
    },
    {
      "entropy": 5.701009464263916,
      "epoch": 0.3785255786811904,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000498981310309936,
      "loss": 5.468,
      "mean_token_accuracy": 0.1617519348859787,
      "num_tokens": 8445743.0,
      "step": 4865
    },
    {
      "entropy": 5.6702821254730225,
      "epoch": 0.378914608052908,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989786742413428,
      "loss": 5.5636,
      "mean_token_accuracy": 0.15807968527078628,
      "num_tokens": 8454091.0,
      "step": 4870
    },
    {
      "entropy": 5.8274698734283445,
      "epoch": 0.3793036374246256,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989760347742298,
      "loss": 5.5638,
      "mean_token_accuracy": 0.1589045539498329,
      "num_tokens": 8461992.0,
      "step": 4875
    },
    {
      "entropy": 5.684429550170899,
      "epoch": 0.3796926667963431,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004989733919086369,
      "loss": 5.5048,
      "mean_token_accuracy": 0.16163448840379716,
      "num_tokens": 8470730.0,
      "step": 4880
    },
    {
      "entropy": 5.702088356018066,
      "epoch": 0.3800816961680607,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004989707456446043,
      "loss": 5.5565,
      "mean_token_accuracy": 0.16036150306463243,
      "num_tokens": 8479947.0,
      "step": 4885
    },
    {
      "entropy": 5.825870656967163,
      "epoch": 0.38047072553977823,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989680959821721,
      "loss": 5.6376,
      "mean_token_accuracy": 0.15936389118432998,
      "num_tokens": 8488237.0,
      "step": 4890
    },
    {
      "entropy": 5.738702583312988,
      "epoch": 0.3808597549114958,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004989654429213805,
      "loss": 5.5045,
      "mean_token_accuracy": 0.1659325733780861,
      "num_tokens": 8497146.0,
      "step": 4895
    },
    {
      "entropy": 5.747964525222779,
      "epoch": 0.3812487842832134,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004989627864622699,
      "loss": 5.5789,
      "mean_token_accuracy": 0.16467523127794265,
      "num_tokens": 8505736.0,
      "step": 4900
    },
    {
      "entropy": 5.765132761001587,
      "epoch": 0.38163781365493094,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004989601266048804,
      "loss": 5.5924,
      "mean_token_accuracy": 0.1583853229880333,
      "num_tokens": 8515368.0,
      "step": 4905
    },
    {
      "entropy": 5.72173228263855,
      "epoch": 0.3820268430266485,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989574633492525,
      "loss": 5.5642,
      "mean_token_accuracy": 0.1621312364935875,
      "num_tokens": 8524141.0,
      "step": 4910
    },
    {
      "entropy": 5.718557834625244,
      "epoch": 0.38241587239836605,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989547966954266,
      "loss": 5.4734,
      "mean_token_accuracy": 0.16636828035116197,
      "num_tokens": 8533116.0,
      "step": 4915
    },
    {
      "entropy": 5.712999486923218,
      "epoch": 0.38280490177008364,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498952126643443,
      "loss": 5.6072,
      "mean_token_accuracy": 0.16635651737451554,
      "num_tokens": 8541919.0,
      "step": 4920
    },
    {
      "entropy": 5.7109620571136475,
      "epoch": 0.3831939311418012,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989494531933424,
      "loss": 5.4876,
      "mean_token_accuracy": 0.17107266783714295,
      "num_tokens": 8550372.0,
      "step": 4925
    },
    {
      "entropy": 5.757292413711548,
      "epoch": 0.38358296051351876,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989467763451652,
      "loss": 5.5827,
      "mean_token_accuracy": 0.15917742997407913,
      "num_tokens": 8559246.0,
      "step": 4930
    },
    {
      "entropy": 5.6922876834869385,
      "epoch": 0.38397198988523634,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989440960989523,
      "loss": 5.3482,
      "mean_token_accuracy": 0.16702754199504852,
      "num_tokens": 8567005.0,
      "step": 4935
    },
    {
      "entropy": 5.64597373008728,
      "epoch": 0.3843610192569539,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498941412454744,
      "loss": 5.5681,
      "mean_token_accuracy": 0.1547877758741379,
      "num_tokens": 8575399.0,
      "step": 4940
    },
    {
      "entropy": 5.817316627502441,
      "epoch": 0.38475004862867146,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989387254125813,
      "loss": 5.6216,
      "mean_token_accuracy": 0.15826431214809417,
      "num_tokens": 8583994.0,
      "step": 4945
    },
    {
      "entropy": 5.731871128082275,
      "epoch": 0.38513907800038905,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989360349725049,
      "loss": 5.5533,
      "mean_token_accuracy": 0.16471867114305497,
      "num_tokens": 8593446.0,
      "step": 4950
    },
    {
      "entropy": 5.743245267868042,
      "epoch": 0.3855281073721066,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989333411345555,
      "loss": 5.5482,
      "mean_token_accuracy": 0.16647694557905196,
      "num_tokens": 8602821.0,
      "step": 4955
    },
    {
      "entropy": 5.751190042495727,
      "epoch": 0.38591713674382416,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989306438987742,
      "loss": 5.6471,
      "mean_token_accuracy": 0.15571539402008056,
      "num_tokens": 8611831.0,
      "step": 4960
    },
    {
      "entropy": 5.795993375778198,
      "epoch": 0.38630616611554175,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989279432652018,
      "loss": 5.5403,
      "mean_token_accuracy": 0.16009591668844222,
      "num_tokens": 8620986.0,
      "step": 4965
    },
    {
      "entropy": 5.733548927307129,
      "epoch": 0.3866951954872593,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989252392338791,
      "loss": 5.4854,
      "mean_token_accuracy": 0.1636480860412121,
      "num_tokens": 8629822.0,
      "step": 4970
    },
    {
      "entropy": 5.724105978012085,
      "epoch": 0.38708422485897687,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989225318048475,
      "loss": 5.6001,
      "mean_token_accuracy": 0.15980519205331803,
      "num_tokens": 8637679.0,
      "step": 4975
    },
    {
      "entropy": 5.76774034500122,
      "epoch": 0.3874732542306944,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989198209781478,
      "loss": 5.6142,
      "mean_token_accuracy": 0.1648936003446579,
      "num_tokens": 8646705.0,
      "step": 4980
    },
    {
      "entropy": 5.794110631942749,
      "epoch": 0.387862283602412,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004989171067538211,
      "loss": 5.5435,
      "mean_token_accuracy": 0.16352889388799668,
      "num_tokens": 8655599.0,
      "step": 4985
    },
    {
      "entropy": 5.69699764251709,
      "epoch": 0.38825131297412957,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000498914389131909,
      "loss": 5.5651,
      "mean_token_accuracy": 0.15984487384557725,
      "num_tokens": 8665025.0,
      "step": 4990
    },
    {
      "entropy": 5.715613317489624,
      "epoch": 0.3886403423458471,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989116681124523,
      "loss": 5.4493,
      "mean_token_accuracy": 0.1684253543615341,
      "num_tokens": 8672994.0,
      "step": 4995
    },
    {
      "entropy": 5.686432600021362,
      "epoch": 0.3890293717175647,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989089436954924,
      "loss": 5.5803,
      "mean_token_accuracy": 0.15855107009410857,
      "num_tokens": 8682457.0,
      "step": 5000
    },
    {
      "entropy": 5.764044570922851,
      "epoch": 0.3894184010892822,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989062158810706,
      "loss": 5.5553,
      "mean_token_accuracy": 0.16128483563661575,
      "num_tokens": 8690888.0,
      "step": 5005
    },
    {
      "entropy": 5.811047554016113,
      "epoch": 0.3898074304609998,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989034846692284,
      "loss": 5.656,
      "mean_token_accuracy": 0.15784021466970444,
      "num_tokens": 8699632.0,
      "step": 5010
    },
    {
      "entropy": 5.699138116836548,
      "epoch": 0.3901964598327174,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989007500600073,
      "loss": 5.4907,
      "mean_token_accuracy": 0.16482501178979875,
      "num_tokens": 8707955.0,
      "step": 5015
    },
    {
      "entropy": 5.753307962417603,
      "epoch": 0.3905854892044349,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988980120534486,
      "loss": 5.6395,
      "mean_token_accuracy": 0.16032529175281524,
      "num_tokens": 8716103.0,
      "step": 5020
    },
    {
      "entropy": 5.694552755355835,
      "epoch": 0.3909745185761525,
      "grad_norm": 1.0,
      "learning_rate": 0.000498895270649594,
      "loss": 5.495,
      "mean_token_accuracy": 0.1711027666926384,
      "num_tokens": 8724499.0,
      "step": 5025
    },
    {
      "entropy": 5.667240476608276,
      "epoch": 0.39136354794787004,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498892525848485,
      "loss": 5.515,
      "mean_token_accuracy": 0.1710180327296257,
      "num_tokens": 8732829.0,
      "step": 5030
    },
    {
      "entropy": 5.62990984916687,
      "epoch": 0.3917525773195876,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988897776501633,
      "loss": 5.3801,
      "mean_token_accuracy": 0.17283668592572213,
      "num_tokens": 8741617.0,
      "step": 5035
    },
    {
      "entropy": 5.655297517776489,
      "epoch": 0.3921416066913052,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004988870260546705,
      "loss": 5.369,
      "mean_token_accuracy": 0.17152581065893174,
      "num_tokens": 8750146.0,
      "step": 5040
    },
    {
      "entropy": 5.684403562545777,
      "epoch": 0.39253063606302274,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988842710620486,
      "loss": 5.5443,
      "mean_token_accuracy": 0.16621591448783873,
      "num_tokens": 8758744.0,
      "step": 5045
    },
    {
      "entropy": 5.667189884185791,
      "epoch": 0.39291966543474033,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004988815126723391,
      "loss": 5.5707,
      "mean_token_accuracy": 0.1603170469403267,
      "num_tokens": 8768219.0,
      "step": 5050
    },
    {
      "entropy": 5.7467005252838135,
      "epoch": 0.3933086948064579,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988787508855841,
      "loss": 5.5639,
      "mean_token_accuracy": 0.15709830820560455,
      "num_tokens": 8777047.0,
      "step": 5055
    },
    {
      "entropy": 5.678100967407227,
      "epoch": 0.39369772417817545,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988759857018253,
      "loss": 5.4517,
      "mean_token_accuracy": 0.16117836385965348,
      "num_tokens": 8785918.0,
      "step": 5060
    },
    {
      "entropy": 5.648393821716309,
      "epoch": 0.39408675354989303,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988732171211048,
      "loss": 5.5265,
      "mean_token_accuracy": 0.15852733552455903,
      "num_tokens": 8794253.0,
      "step": 5065
    },
    {
      "entropy": 5.688629388809204,
      "epoch": 0.39447578292161056,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988704451434644,
      "loss": 5.5785,
      "mean_token_accuracy": 0.16885487884283065,
      "num_tokens": 8802352.0,
      "step": 5070
    },
    {
      "entropy": 5.782166004180908,
      "epoch": 0.39486481229332815,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004988676697689465,
      "loss": 5.5381,
      "mean_token_accuracy": 0.15879994332790376,
      "num_tokens": 8810956.0,
      "step": 5075
    },
    {
      "entropy": 5.6235700130462645,
      "epoch": 0.39525384166504574,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498864890997593,
      "loss": 5.4612,
      "mean_token_accuracy": 0.16021384447813034,
      "num_tokens": 8819062.0,
      "step": 5080
    },
    {
      "entropy": 5.668540382385254,
      "epoch": 0.39564287103676327,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988621088294461,
      "loss": 5.5664,
      "mean_token_accuracy": 0.1544269561767578,
      "num_tokens": 8827470.0,
      "step": 5085
    },
    {
      "entropy": 5.756826543807984,
      "epoch": 0.39603190040848085,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498859323264548,
      "loss": 5.5979,
      "mean_token_accuracy": 0.1592550739645958,
      "num_tokens": 8837003.0,
      "step": 5090
    },
    {
      "entropy": 5.724267387390137,
      "epoch": 0.3964209297801984,
      "grad_norm": 0.921875,
      "learning_rate": 0.000498856534302941,
      "loss": 5.5996,
      "mean_token_accuracy": 0.15522586405277253,
      "num_tokens": 8846605.0,
      "step": 5095
    },
    {
      "entropy": 5.802182674407959,
      "epoch": 0.39680995915191597,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988537419446673,
      "loss": 5.6184,
      "mean_token_accuracy": 0.15980854481458664,
      "num_tokens": 8855232.0,
      "step": 5100
    },
    {
      "entropy": 5.745633840560913,
      "epoch": 0.39719898852363356,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988509461897694,
      "loss": 5.5063,
      "mean_token_accuracy": 0.1694345474243164,
      "num_tokens": 8864176.0,
      "step": 5105
    },
    {
      "entropy": 5.700151252746582,
      "epoch": 0.3975880178953511,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004988481470382897,
      "loss": 5.581,
      "mean_token_accuracy": 0.16570933312177658,
      "num_tokens": 8872671.0,
      "step": 5110
    },
    {
      "entropy": 5.633589887619019,
      "epoch": 0.3979770472670687,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988453444902708,
      "loss": 5.4773,
      "mean_token_accuracy": 0.1687018722295761,
      "num_tokens": 8881316.0,
      "step": 5115
    },
    {
      "entropy": 5.768270206451416,
      "epoch": 0.3983660766387862,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004988425385457549,
      "loss": 5.6128,
      "mean_token_accuracy": 0.16246161833405495,
      "num_tokens": 8890114.0,
      "step": 5120
    },
    {
      "entropy": 5.7693624019622805,
      "epoch": 0.3987551060105038,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004988397292047848,
      "loss": 5.5367,
      "mean_token_accuracy": 0.16662347167730332,
      "num_tokens": 8899598.0,
      "step": 5125
    },
    {
      "entropy": 5.5953779220581055,
      "epoch": 0.3991441353822214,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988369164674032,
      "loss": 5.3957,
      "mean_token_accuracy": 0.16895558983087539,
      "num_tokens": 8908244.0,
      "step": 5130
    },
    {
      "entropy": 5.67269926071167,
      "epoch": 0.3995331647539389,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988341003336526,
      "loss": 5.4557,
      "mean_token_accuracy": 0.16561708748340606,
      "num_tokens": 8916280.0,
      "step": 5135
    },
    {
      "entropy": 5.776008939743042,
      "epoch": 0.3999221941256565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988312808035757,
      "loss": 5.5072,
      "mean_token_accuracy": 0.1566791757941246,
      "num_tokens": 8923947.0,
      "step": 5140
    },
    {
      "entropy": 5.646599531173706,
      "epoch": 0.400311223497374,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004988284578772155,
      "loss": 5.5424,
      "mean_token_accuracy": 0.15635721981525422,
      "num_tokens": 8933374.0,
      "step": 5145
    },
    {
      "entropy": 5.6722198009490965,
      "epoch": 0.4007002528690916,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004988256315546146,
      "loss": 5.4755,
      "mean_token_accuracy": 0.1645133301615715,
      "num_tokens": 8941757.0,
      "step": 5150
    },
    {
      "entropy": 5.66157374382019,
      "epoch": 0.4010892822408092,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988228018358161,
      "loss": 5.3833,
      "mean_token_accuracy": 0.16919830590486526,
      "num_tokens": 8950128.0,
      "step": 5155
    },
    {
      "entropy": 5.734227657318115,
      "epoch": 0.40147831161252673,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988199687208628,
      "loss": 5.5318,
      "mean_token_accuracy": 0.15678182244300842,
      "num_tokens": 8958939.0,
      "step": 5160
    },
    {
      "entropy": 5.7235174655914305,
      "epoch": 0.4018673409842443,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988171322097977,
      "loss": 5.4884,
      "mean_token_accuracy": 0.166029854118824,
      "num_tokens": 8967519.0,
      "step": 5165
    },
    {
      "entropy": 5.623489952087402,
      "epoch": 0.4022563703559619,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988142923026638,
      "loss": 5.3649,
      "mean_token_accuracy": 0.17290344089269638,
      "num_tokens": 8976215.0,
      "step": 5170
    },
    {
      "entropy": 5.739109516143799,
      "epoch": 0.40264539972767943,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988114489995044,
      "loss": 5.5925,
      "mean_token_accuracy": 0.15214981213212014,
      "num_tokens": 8984060.0,
      "step": 5175
    },
    {
      "entropy": 5.726056623458862,
      "epoch": 0.403034429099397,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988086023003624,
      "loss": 5.5359,
      "mean_token_accuracy": 0.15775070786476136,
      "num_tokens": 8992301.0,
      "step": 5180
    },
    {
      "entropy": 5.763833379745483,
      "epoch": 0.40342345847111455,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988057522052811,
      "loss": 5.5542,
      "mean_token_accuracy": 0.16253505051136016,
      "num_tokens": 9000517.0,
      "step": 5185
    },
    {
      "entropy": 5.663650035858154,
      "epoch": 0.40381248784283214,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988028987143037,
      "loss": 5.5185,
      "mean_token_accuracy": 0.15818402916193008,
      "num_tokens": 9008826.0,
      "step": 5190
    },
    {
      "entropy": 5.710343217849731,
      "epoch": 0.4042015172145497,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004988000418274736,
      "loss": 5.5037,
      "mean_token_accuracy": 0.15956878066062927,
      "num_tokens": 9017952.0,
      "step": 5195
    },
    {
      "entropy": 5.698844003677368,
      "epoch": 0.40459054658626725,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004987971815448341,
      "loss": 5.5108,
      "mean_token_accuracy": 0.16802990138530732,
      "num_tokens": 9026688.0,
      "step": 5200
    },
    {
      "entropy": 5.6753887176513675,
      "epoch": 0.40497957595798484,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004987943178664285,
      "loss": 5.4604,
      "mean_token_accuracy": 0.16948893219232558,
      "num_tokens": 9035761.0,
      "step": 5205
    },
    {
      "entropy": 5.711546516418457,
      "epoch": 0.40536860532970237,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004987914507923004,
      "loss": 5.6052,
      "mean_token_accuracy": 0.15351177304983138,
      "num_tokens": 9044510.0,
      "step": 5210
    },
    {
      "entropy": 5.688313055038452,
      "epoch": 0.40575763470141996,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987885803224931,
      "loss": 5.4623,
      "mean_token_accuracy": 0.16280150562524795,
      "num_tokens": 9053031.0,
      "step": 5215
    },
    {
      "entropy": 5.616592454910278,
      "epoch": 0.40614666407313754,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987857064570505,
      "loss": 5.4429,
      "mean_token_accuracy": 0.1689184382557869,
      "num_tokens": 9061580.0,
      "step": 5220
    },
    {
      "entropy": 5.711910200119019,
      "epoch": 0.4065356934448551,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004987828291960158,
      "loss": 5.5666,
      "mean_token_accuracy": 0.16402611136436462,
      "num_tokens": 9069741.0,
      "step": 5225
    },
    {
      "entropy": 5.8263026714324955,
      "epoch": 0.40692472281657266,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498779948539433,
      "loss": 5.5922,
      "mean_token_accuracy": 0.15314422994852067,
      "num_tokens": 9079214.0,
      "step": 5230
    },
    {
      "entropy": 5.779076194763183,
      "epoch": 0.4073137521882902,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004987770644873455,
      "loss": 5.5444,
      "mean_token_accuracy": 0.16699129492044448,
      "num_tokens": 9088201.0,
      "step": 5235
    },
    {
      "entropy": 5.714314365386963,
      "epoch": 0.4077027815600078,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987741770397974,
      "loss": 5.5828,
      "mean_token_accuracy": 0.15950050503015517,
      "num_tokens": 9096829.0,
      "step": 5240
    },
    {
      "entropy": 5.693706512451172,
      "epoch": 0.40809181093172536,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987712861968323,
      "loss": 5.5217,
      "mean_token_accuracy": 0.16581078618764877,
      "num_tokens": 9105515.0,
      "step": 5245
    },
    {
      "entropy": 5.708260154724121,
      "epoch": 0.4084808403034429,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498768391958494,
      "loss": 5.4997,
      "mean_token_accuracy": 0.16316778659820558,
      "num_tokens": 9113614.0,
      "step": 5250
    },
    {
      "entropy": 5.705586671829224,
      "epoch": 0.4088698696751605,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004987654943248266,
      "loss": 5.5814,
      "mean_token_accuracy": 0.15525239408016206,
      "num_tokens": 9122337.0,
      "step": 5255
    },
    {
      "entropy": 5.752633762359619,
      "epoch": 0.40925889904687807,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987625932958739,
      "loss": 5.4856,
      "mean_token_accuracy": 0.16904787570238114,
      "num_tokens": 9130417.0,
      "step": 5260
    },
    {
      "entropy": 5.7156274795532225,
      "epoch": 0.4096479284185956,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987596888716801,
      "loss": 5.5429,
      "mean_token_accuracy": 0.15976496934890747,
      "num_tokens": 9139317.0,
      "step": 5265
    },
    {
      "entropy": 5.701608371734619,
      "epoch": 0.4100369577903132,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987567810522892,
      "loss": 5.5256,
      "mean_token_accuracy": 0.16300612986087798,
      "num_tokens": 9147808.0,
      "step": 5270
    },
    {
      "entropy": 5.785950708389282,
      "epoch": 0.4104259871620307,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004987538698377451,
      "loss": 5.6164,
      "mean_token_accuracy": 0.1566879279911518,
      "num_tokens": 9157117.0,
      "step": 5275
    },
    {
      "entropy": 5.738699245452881,
      "epoch": 0.4108150165337483,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987509552280924,
      "loss": 5.4742,
      "mean_token_accuracy": 0.15899330377578735,
      "num_tokens": 9165297.0,
      "step": 5280
    },
    {
      "entropy": 5.749457693099975,
      "epoch": 0.4112040459054659,
      "grad_norm": 1.0625,
      "learning_rate": 0.000498748037223375,
      "loss": 5.605,
      "mean_token_accuracy": 0.16066010892391205,
      "num_tokens": 9172945.0,
      "step": 5285
    },
    {
      "entropy": 5.722745656967163,
      "epoch": 0.4115930752771834,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987451158236372,
      "loss": 5.5321,
      "mean_token_accuracy": 0.16613929271697997,
      "num_tokens": 9181445.0,
      "step": 5290
    },
    {
      "entropy": 5.635143423080445,
      "epoch": 0.411982104648901,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987421910289234,
      "loss": 5.3782,
      "mean_token_accuracy": 0.17040496617555617,
      "num_tokens": 9189535.0,
      "step": 5295
    },
    {
      "entropy": 5.684240674972534,
      "epoch": 0.41237113402061853,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987392628392781,
      "loss": 5.6002,
      "mean_token_accuracy": 0.14962754249572754,
      "num_tokens": 9198328.0,
      "step": 5300
    },
    {
      "entropy": 5.749529075622559,
      "epoch": 0.4127601633923361,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004987363312547456,
      "loss": 5.486,
      "mean_token_accuracy": 0.16676264256238937,
      "num_tokens": 9206840.0,
      "step": 5305
    },
    {
      "entropy": 5.644282436370849,
      "epoch": 0.4131491927640537,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987333962753703,
      "loss": 5.3891,
      "mean_token_accuracy": 0.1677182286977768,
      "num_tokens": 9215155.0,
      "step": 5310
    },
    {
      "entropy": 5.636444759368897,
      "epoch": 0.41353822213577124,
      "grad_norm": 1.0,
      "learning_rate": 0.0004987304579011967,
      "loss": 5.5709,
      "mean_token_accuracy": 0.16278092861175536,
      "num_tokens": 9224426.0,
      "step": 5315
    },
    {
      "entropy": 5.75674786567688,
      "epoch": 0.4139272515074888,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987275161322697,
      "loss": 5.5879,
      "mean_token_accuracy": 0.16320908665657044,
      "num_tokens": 9233785.0,
      "step": 5320
    },
    {
      "entropy": 5.726920557022095,
      "epoch": 0.41431628087920636,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987245709686337,
      "loss": 5.5433,
      "mean_token_accuracy": 0.1654205784201622,
      "num_tokens": 9242168.0,
      "step": 5325
    },
    {
      "entropy": 5.675925016403198,
      "epoch": 0.41470531025092394,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987216224103334,
      "loss": 5.5575,
      "mean_token_accuracy": 0.165568046271801,
      "num_tokens": 9250862.0,
      "step": 5330
    },
    {
      "entropy": 5.745130348205566,
      "epoch": 0.41509433962264153,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004987186704574136,
      "loss": 5.5718,
      "mean_token_accuracy": 0.1554880768060684,
      "num_tokens": 9259957.0,
      "step": 5335
    },
    {
      "entropy": 5.6644360542297365,
      "epoch": 0.41548336899435906,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004987157151099189,
      "loss": 5.4438,
      "mean_token_accuracy": 0.1625848740339279,
      "num_tokens": 9268590.0,
      "step": 5340
    },
    {
      "entropy": 5.731019973754883,
      "epoch": 0.41587239836607665,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987127563678945,
      "loss": 5.5041,
      "mean_token_accuracy": 0.16602175831794738,
      "num_tokens": 9276872.0,
      "step": 5345
    },
    {
      "entropy": 5.660704183578491,
      "epoch": 0.4162614277377942,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987097942313852,
      "loss": 5.3897,
      "mean_token_accuracy": 0.16859978586435317,
      "num_tokens": 9284556.0,
      "step": 5350
    },
    {
      "entropy": 5.658067750930786,
      "epoch": 0.41665045710951176,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987068287004355,
      "loss": 5.5496,
      "mean_token_accuracy": 0.16045835316181184,
      "num_tokens": 9293315.0,
      "step": 5355
    },
    {
      "entropy": 5.743117427825927,
      "epoch": 0.41703948648122935,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987038597750909,
      "loss": 5.5163,
      "mean_token_accuracy": 0.16282073706388472,
      "num_tokens": 9302923.0,
      "step": 5360
    },
    {
      "entropy": 5.745809125900268,
      "epoch": 0.4174285158529469,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987008874553964,
      "loss": 5.4611,
      "mean_token_accuracy": 0.16524637937545777,
      "num_tokens": 9311388.0,
      "step": 5365
    },
    {
      "entropy": 5.652669095993042,
      "epoch": 0.41781754522466447,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004986979117413969,
      "loss": 5.4161,
      "mean_token_accuracy": 0.16866108775138855,
      "num_tokens": 9319880.0,
      "step": 5370
    },
    {
      "entropy": 5.64964451789856,
      "epoch": 0.41820657459638205,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986949326331376,
      "loss": 5.5533,
      "mean_token_accuracy": 0.16394897997379304,
      "num_tokens": 9328374.0,
      "step": 5375
    },
    {
      "entropy": 5.802340698242188,
      "epoch": 0.4185956039680996,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986919501306638,
      "loss": 5.5123,
      "mean_token_accuracy": 0.1731323316693306,
      "num_tokens": 9336428.0,
      "step": 5380
    },
    {
      "entropy": 5.649902200698852,
      "epoch": 0.41898463333981717,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004986889642340207,
      "loss": 5.5978,
      "mean_token_accuracy": 0.16033813655376433,
      "num_tokens": 9345794.0,
      "step": 5385
    },
    {
      "entropy": 5.750280857086182,
      "epoch": 0.4193736627115347,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004986859749432536,
      "loss": 5.5594,
      "mean_token_accuracy": 0.16315516829490662,
      "num_tokens": 9354724.0,
      "step": 5390
    },
    {
      "entropy": 5.756238889694214,
      "epoch": 0.4197626920832523,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986829822584078,
      "loss": 5.5656,
      "mean_token_accuracy": 0.15743476301431655,
      "num_tokens": 9364008.0,
      "step": 5395
    },
    {
      "entropy": 5.707322835922241,
      "epoch": 0.4201517214549699,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986799861795289,
      "loss": 5.5339,
      "mean_token_accuracy": 0.1696193978190422,
      "num_tokens": 9372409.0,
      "step": 5400
    },
    {
      "entropy": 5.7119945049285885,
      "epoch": 0.4205407508266874,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004986769867066622,
      "loss": 5.4935,
      "mean_token_accuracy": 0.15928569734096526,
      "num_tokens": 9381172.0,
      "step": 5405
    },
    {
      "entropy": 5.687317514419556,
      "epoch": 0.420929780198405,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004986739838398532,
      "loss": 5.498,
      "mean_token_accuracy": 0.16979421973228453,
      "num_tokens": 9389245.0,
      "step": 5410
    },
    {
      "entropy": 5.674244928359985,
      "epoch": 0.4213188095701225,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004986709775791475,
      "loss": 5.5252,
      "mean_token_accuracy": 0.1654878482222557,
      "num_tokens": 9397954.0,
      "step": 5415
    },
    {
      "entropy": 5.690672397613525,
      "epoch": 0.4217078389418401,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986679679245907,
      "loss": 5.6287,
      "mean_token_accuracy": 0.16026378571987152,
      "num_tokens": 9406986.0,
      "step": 5420
    },
    {
      "entropy": 5.717019557952881,
      "epoch": 0.4220968683135577,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986649548762286,
      "loss": 5.5742,
      "mean_token_accuracy": 0.1631506010890007,
      "num_tokens": 9415163.0,
      "step": 5425
    },
    {
      "entropy": 5.661070251464844,
      "epoch": 0.4224858976852752,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986619384341066,
      "loss": 5.4088,
      "mean_token_accuracy": 0.1779765322804451,
      "num_tokens": 9423422.0,
      "step": 5430
    },
    {
      "entropy": 5.645610094070435,
      "epoch": 0.4228749270569928,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004986589185982708,
      "loss": 5.4398,
      "mean_token_accuracy": 0.1771764948964119,
      "num_tokens": 9431349.0,
      "step": 5435
    },
    {
      "entropy": 5.688213777542114,
      "epoch": 0.42326395642871034,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004986558953687671,
      "loss": 5.5416,
      "mean_token_accuracy": 0.1582077220082283,
      "num_tokens": 9439715.0,
      "step": 5440
    },
    {
      "entropy": 5.704529523849487,
      "epoch": 0.42365298580042793,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004986528687456409,
      "loss": 5.5438,
      "mean_token_accuracy": 0.16103184521198272,
      "num_tokens": 9448216.0,
      "step": 5445
    },
    {
      "entropy": 5.721759462356568,
      "epoch": 0.4240420151721455,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986498387289384,
      "loss": 5.4833,
      "mean_token_accuracy": 0.16496723741292954,
      "num_tokens": 9456346.0,
      "step": 5450
    },
    {
      "entropy": 5.695312547683716,
      "epoch": 0.42443104454386305,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986468053187058,
      "loss": 5.5504,
      "mean_token_accuracy": 0.16503108739852906,
      "num_tokens": 9465476.0,
      "step": 5455
    },
    {
      "entropy": 5.706718254089355,
      "epoch": 0.42482007391558063,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004986437685149887,
      "loss": 5.612,
      "mean_token_accuracy": 0.15952038690447806,
      "num_tokens": 9474582.0,
      "step": 5460
    },
    {
      "entropy": 5.746453380584716,
      "epoch": 0.4252091032872982,
      "grad_norm": 1.0,
      "learning_rate": 0.0004986407283178334,
      "loss": 5.4859,
      "mean_token_accuracy": 0.1622675821185112,
      "num_tokens": 9482801.0,
      "step": 5465
    },
    {
      "entropy": 5.641479969024658,
      "epoch": 0.42559813265901575,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986376847272861,
      "loss": 5.5206,
      "mean_token_accuracy": 0.16077911108732224,
      "num_tokens": 9490894.0,
      "step": 5470
    },
    {
      "entropy": 5.651269340515137,
      "epoch": 0.42598716203073334,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986346377433929,
      "loss": 5.4734,
      "mean_token_accuracy": 0.16265116333961488,
      "num_tokens": 9499414.0,
      "step": 5475
    },
    {
      "entropy": 5.661998271942139,
      "epoch": 0.42637619140245087,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004986315873661999,
      "loss": 5.442,
      "mean_token_accuracy": 0.16423789411783218,
      "num_tokens": 9508945.0,
      "step": 5480
    },
    {
      "entropy": 5.670542764663696,
      "epoch": 0.42676522077416845,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004986285335957536,
      "loss": 5.5177,
      "mean_token_accuracy": 0.16872450709342957,
      "num_tokens": 9517833.0,
      "step": 5485
    },
    {
      "entropy": 5.726272010803223,
      "epoch": 0.42715425014588604,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986254764321002,
      "loss": 5.6532,
      "mean_token_accuracy": 0.15438937693834304,
      "num_tokens": 9526551.0,
      "step": 5490
    },
    {
      "entropy": 5.655402517318725,
      "epoch": 0.42754327951760357,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004986224158752861,
      "loss": 5.4257,
      "mean_token_accuracy": 0.16619780510663987,
      "num_tokens": 9535224.0,
      "step": 5495
    },
    {
      "entropy": 5.705122661590576,
      "epoch": 0.42793230888932116,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004986193519253578,
      "loss": 5.5366,
      "mean_token_accuracy": 0.154195199906826,
      "num_tokens": 9544466.0,
      "step": 5500
    },
    {
      "entropy": 5.6616448879241945,
      "epoch": 0.4283213382610387,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986162845823618,
      "loss": 5.5371,
      "mean_token_accuracy": 0.16176573187112808,
      "num_tokens": 9553520.0,
      "step": 5505
    },
    {
      "entropy": 5.68340654373169,
      "epoch": 0.4287103676327563,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986132138463446,
      "loss": 5.4413,
      "mean_token_accuracy": 0.17152554541826248,
      "num_tokens": 9562053.0,
      "step": 5510
    },
    {
      "entropy": 5.66190538406372,
      "epoch": 0.42909939700447386,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004986101397173528,
      "loss": 5.4353,
      "mean_token_accuracy": 0.16798286885023117,
      "num_tokens": 9570632.0,
      "step": 5515
    },
    {
      "entropy": 5.724105596542358,
      "epoch": 0.4294884263761914,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498607062195433,
      "loss": 5.4348,
      "mean_token_accuracy": 0.17020454853773118,
      "num_tokens": 9578739.0,
      "step": 5520
    },
    {
      "entropy": 5.612820720672607,
      "epoch": 0.429877455747909,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498603981280632,
      "loss": 5.4649,
      "mean_token_accuracy": 0.16295552998781204,
      "num_tokens": 9587314.0,
      "step": 5525
    },
    {
      "entropy": 5.702166795730591,
      "epoch": 0.4302664851196265,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004986008969729964,
      "loss": 5.5272,
      "mean_token_accuracy": 0.16558773517608644,
      "num_tokens": 9595954.0,
      "step": 5530
    },
    {
      "entropy": 5.751034450531006,
      "epoch": 0.4306555144913441,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985978092725731,
      "loss": 5.5447,
      "mean_token_accuracy": 0.15605991929769517,
      "num_tokens": 9604767.0,
      "step": 5535
    },
    {
      "entropy": 5.632840204238891,
      "epoch": 0.4310445438630617,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498594718179409,
      "loss": 5.4079,
      "mean_token_accuracy": 0.16811801493167877,
      "num_tokens": 9613173.0,
      "step": 5540
    },
    {
      "entropy": 5.7372047901153564,
      "epoch": 0.4314335732347792,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004985916236935508,
      "loss": 5.5031,
      "mean_token_accuracy": 0.16015706434845925,
      "num_tokens": 9622777.0,
      "step": 5545
    },
    {
      "entropy": 5.673089981079102,
      "epoch": 0.4318226026064968,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004985885258150458,
      "loss": 5.4563,
      "mean_token_accuracy": 0.16085392683744432,
      "num_tokens": 9631019.0,
      "step": 5550
    },
    {
      "entropy": 5.653888320922851,
      "epoch": 0.43221163197821433,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985854245439408,
      "loss": 5.5169,
      "mean_token_accuracy": 0.15890958905220032,
      "num_tokens": 9639797.0,
      "step": 5555
    },
    {
      "entropy": 5.694570541381836,
      "epoch": 0.4326006613499319,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985823198802827,
      "loss": 5.4995,
      "mean_token_accuracy": 0.16874769777059556,
      "num_tokens": 9648393.0,
      "step": 5560
    },
    {
      "entropy": 5.712398386001587,
      "epoch": 0.4329896907216495,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004985792118241188,
      "loss": 5.4821,
      "mean_token_accuracy": 0.1667462021112442,
      "num_tokens": 9657260.0,
      "step": 5565
    },
    {
      "entropy": 5.654960584640503,
      "epoch": 0.43337872009336703,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985761003754961,
      "loss": 5.5491,
      "mean_token_accuracy": 0.16649328768253327,
      "num_tokens": 9665883.0,
      "step": 5570
    },
    {
      "entropy": 5.7316896438598635,
      "epoch": 0.4337677494650846,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985729855344622,
      "loss": 5.5645,
      "mean_token_accuracy": 0.1632181152701378,
      "num_tokens": 9674100.0,
      "step": 5575
    },
    {
      "entropy": 5.70767560005188,
      "epoch": 0.4341567788368022,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000498569867301064,
      "loss": 5.5275,
      "mean_token_accuracy": 0.1664811134338379,
      "num_tokens": 9683236.0,
      "step": 5580
    },
    {
      "entropy": 5.661159086227417,
      "epoch": 0.43454580820851973,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004985667456753489,
      "loss": 5.4882,
      "mean_token_accuracy": 0.1633784294128418,
      "num_tokens": 9692076.0,
      "step": 5585
    },
    {
      "entropy": 5.708376789093018,
      "epoch": 0.4349348375802373,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985636206573642,
      "loss": 5.5367,
      "mean_token_accuracy": 0.16655746847391129,
      "num_tokens": 9700266.0,
      "step": 5590
    },
    {
      "entropy": 5.7091821193695065,
      "epoch": 0.43532386695195485,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004985604922471575,
      "loss": 5.4921,
      "mean_token_accuracy": 0.16823821514844894,
      "num_tokens": 9708643.0,
      "step": 5595
    },
    {
      "entropy": 5.710525894165039,
      "epoch": 0.43571289632367244,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004985573604447761,
      "loss": 5.5321,
      "mean_token_accuracy": 0.16186349242925643,
      "num_tokens": 9717459.0,
      "step": 5600
    },
    {
      "entropy": 5.641446352005005,
      "epoch": 0.43610192569539,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004985542252502676,
      "loss": 5.4471,
      "mean_token_accuracy": 0.16886800825595855,
      "num_tokens": 9726251.0,
      "step": 5605
    },
    {
      "entropy": 5.5887237071990965,
      "epoch": 0.43649095506710756,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985510866636796,
      "loss": 5.4351,
      "mean_token_accuracy": 0.16684530675411224,
      "num_tokens": 9735327.0,
      "step": 5610
    },
    {
      "entropy": 5.692466068267822,
      "epoch": 0.43687998443882514,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985479446850596,
      "loss": 5.5955,
      "mean_token_accuracy": 0.1570721372961998,
      "num_tokens": 9743642.0,
      "step": 5615
    },
    {
      "entropy": 5.708640003204346,
      "epoch": 0.4372690138105427,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004985447993144554,
      "loss": 5.4528,
      "mean_token_accuracy": 0.16921658664941788,
      "num_tokens": 9752235.0,
      "step": 5620
    },
    {
      "entropy": 5.721048498153687,
      "epoch": 0.43765804318226026,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985416505519147,
      "loss": 5.5436,
      "mean_token_accuracy": 0.15602369606494904,
      "num_tokens": 9761012.0,
      "step": 5625
    },
    {
      "entropy": 5.660058546066284,
      "epoch": 0.43804707255397785,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004985384983974853,
      "loss": 5.4967,
      "mean_token_accuracy": 0.16976085305213928,
      "num_tokens": 9769810.0,
      "step": 5630
    },
    {
      "entropy": 5.696043157577515,
      "epoch": 0.4384361019256954,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004985353428512148,
      "loss": 5.4427,
      "mean_token_accuracy": 0.16465951800346373,
      "num_tokens": 9778186.0,
      "step": 5635
    },
    {
      "entropy": 5.6428123950958256,
      "epoch": 0.43882513129741296,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985321839131514,
      "loss": 5.5107,
      "mean_token_accuracy": 0.1715571776032448,
      "num_tokens": 9786456.0,
      "step": 5640
    },
    {
      "entropy": 5.6313409328460695,
      "epoch": 0.4392141606691305,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004985290215833428,
      "loss": 5.5105,
      "mean_token_accuracy": 0.16007644087076187,
      "num_tokens": 9794613.0,
      "step": 5645
    },
    {
      "entropy": 5.772847032546997,
      "epoch": 0.4396031900408481,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985258558618372,
      "loss": 5.4253,
      "mean_token_accuracy": 0.16906558126211166,
      "num_tokens": 9803388.0,
      "step": 5650
    },
    {
      "entropy": 5.686429882049561,
      "epoch": 0.43999221941256567,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004985226867486825,
      "loss": 5.4927,
      "mean_token_accuracy": 0.16778526455163956,
      "num_tokens": 9812450.0,
      "step": 5655
    },
    {
      "entropy": 5.697888708114624,
      "epoch": 0.4403812487842832,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985195142439267,
      "loss": 5.5822,
      "mean_token_accuracy": 0.15718315094709395,
      "num_tokens": 9821377.0,
      "step": 5660
    },
    {
      "entropy": 5.705540800094605,
      "epoch": 0.4407702781560008,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985163383476181,
      "loss": 5.4875,
      "mean_token_accuracy": 0.1577681854367256,
      "num_tokens": 9829761.0,
      "step": 5665
    },
    {
      "entropy": 5.733972978591919,
      "epoch": 0.44115930752771837,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985131590598048,
      "loss": 5.5511,
      "mean_token_accuracy": 0.1609368309378624,
      "num_tokens": 9838921.0,
      "step": 5670
    },
    {
      "entropy": 5.6359227180480955,
      "epoch": 0.4415483368994359,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985099763805352,
      "loss": 5.3777,
      "mean_token_accuracy": 0.17368233948946,
      "num_tokens": 9846822.0,
      "step": 5675
    },
    {
      "entropy": 5.72635440826416,
      "epoch": 0.4419373662711535,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985067903098574,
      "loss": 5.5993,
      "mean_token_accuracy": 0.16365385204553604,
      "num_tokens": 9855951.0,
      "step": 5680
    },
    {
      "entropy": 5.744476556777954,
      "epoch": 0.442326395642871,
      "grad_norm": 1.0,
      "learning_rate": 0.0004985036008478197,
      "loss": 5.5551,
      "mean_token_accuracy": 0.1636480838060379,
      "num_tokens": 9864875.0,
      "step": 5685
    },
    {
      "entropy": 5.712133169174194,
      "epoch": 0.4427154250145886,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985004079944707,
      "loss": 5.5163,
      "mean_token_accuracy": 0.16914052963256837,
      "num_tokens": 9873785.0,
      "step": 5690
    },
    {
      "entropy": 5.688980293273926,
      "epoch": 0.4431044543863062,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004984972117498587,
      "loss": 5.5177,
      "mean_token_accuracy": 0.16533356308937072,
      "num_tokens": 9883214.0,
      "step": 5695
    },
    {
      "entropy": 5.645994234085083,
      "epoch": 0.4434934837580237,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004984940121140323,
      "loss": 5.479,
      "mean_token_accuracy": 0.16787913143634797,
      "num_tokens": 9891346.0,
      "step": 5700
    },
    {
      "entropy": 5.633553791046142,
      "epoch": 0.4438825131297413,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049849080908704,
      "loss": 5.4031,
      "mean_token_accuracy": 0.16850567013025283,
      "num_tokens": 9900646.0,
      "step": 5705
    },
    {
      "entropy": 5.717324352264404,
      "epoch": 0.44427154250145884,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004984876026689302,
      "loss": 5.5555,
      "mean_token_accuracy": 0.16231814920902252,
      "num_tokens": 9910851.0,
      "step": 5710
    },
    {
      "entropy": 5.761937665939331,
      "epoch": 0.4446605718731764,
      "grad_norm": 1.03125,
      "learning_rate": 0.000498484392859752,
      "loss": 5.5562,
      "mean_token_accuracy": 0.16021174788475037,
      "num_tokens": 9919919.0,
      "step": 5715
    },
    {
      "entropy": 5.69112377166748,
      "epoch": 0.445049601244894,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004984811796595538,
      "loss": 5.4923,
      "mean_token_accuracy": 0.16262091994285582,
      "num_tokens": 9928831.0,
      "step": 5720
    },
    {
      "entropy": 5.654398727416992,
      "epoch": 0.44543863061661154,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984779630683843,
      "loss": 5.4547,
      "mean_token_accuracy": 0.1687433198094368,
      "num_tokens": 9936840.0,
      "step": 5725
    },
    {
      "entropy": 5.6957135677337645,
      "epoch": 0.44582765998832913,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004984747430862924,
      "loss": 5.5298,
      "mean_token_accuracy": 0.16438896656036378,
      "num_tokens": 9945845.0,
      "step": 5730
    },
    {
      "entropy": 5.732899951934814,
      "epoch": 0.44621668936004666,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004984715197133271,
      "loss": 5.5126,
      "mean_token_accuracy": 0.16476755142211913,
      "num_tokens": 9955176.0,
      "step": 5735
    },
    {
      "entropy": 5.601716375350952,
      "epoch": 0.44660571873176425,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984682929495371,
      "loss": 5.4027,
      "mean_token_accuracy": 0.16928163319826126,
      "num_tokens": 9963182.0,
      "step": 5740
    },
    {
      "entropy": 5.685400867462159,
      "epoch": 0.44699474810348183,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984650627949715,
      "loss": 5.5542,
      "mean_token_accuracy": 0.16373046189546586,
      "num_tokens": 9971239.0,
      "step": 5745
    },
    {
      "entropy": 5.712329864501953,
      "epoch": 0.44738377747519936,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004984618292496792,
      "loss": 5.5029,
      "mean_token_accuracy": 0.16495627909898758,
      "num_tokens": 9980285.0,
      "step": 5750
    },
    {
      "entropy": 5.7070457458496096,
      "epoch": 0.44777280684691695,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004984585923137093,
      "loss": 5.516,
      "mean_token_accuracy": 0.16542391180992128,
      "num_tokens": 9988922.0,
      "step": 5755
    },
    {
      "entropy": 5.711098575592041,
      "epoch": 0.4481618362186345,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498455351987111,
      "loss": 5.5938,
      "mean_token_accuracy": 0.1608327105641365,
      "num_tokens": 9998538.0,
      "step": 5760
    },
    {
      "entropy": 5.81380786895752,
      "epoch": 0.44855086559035207,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004984521082699333,
      "loss": 5.4965,
      "mean_token_accuracy": 0.16169293075799943,
      "num_tokens": 10006646.0,
      "step": 5765
    },
    {
      "entropy": 5.657651472091675,
      "epoch": 0.44893989496206965,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004984488611622256,
      "loss": 5.4764,
      "mean_token_accuracy": 0.1706524446606636,
      "num_tokens": 10016511.0,
      "step": 5770
    },
    {
      "entropy": 5.626505661010742,
      "epoch": 0.4493289243337872,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498445610664037,
      "loss": 5.3888,
      "mean_token_accuracy": 0.16900736689567566,
      "num_tokens": 10025223.0,
      "step": 5775
    },
    {
      "entropy": 5.618741846084594,
      "epoch": 0.44971795370550477,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498442356775417,
      "loss": 5.4751,
      "mean_token_accuracy": 0.16690605729818345,
      "num_tokens": 10033678.0,
      "step": 5780
    },
    {
      "entropy": 5.664579105377197,
      "epoch": 0.45010698307722236,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004984390994964148,
      "loss": 5.3964,
      "mean_token_accuracy": 0.16832560449838638,
      "num_tokens": 10042028.0,
      "step": 5785
    },
    {
      "entropy": 5.711661767959595,
      "epoch": 0.4504960124489399,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00049843583882708,
      "loss": 5.5481,
      "mean_token_accuracy": 0.16350454241037368,
      "num_tokens": 10050758.0,
      "step": 5790
    },
    {
      "entropy": 5.591366386413574,
      "epoch": 0.4508850418206575,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498432574767462,
      "loss": 5.3269,
      "mean_token_accuracy": 0.17831166684627534,
      "num_tokens": 10058688.0,
      "step": 5795
    },
    {
      "entropy": 5.621751403808593,
      "epoch": 0.451274071192375,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004984293073176103,
      "loss": 5.4197,
      "mean_token_accuracy": 0.17680519819259644,
      "num_tokens": 10067302.0,
      "step": 5800
    },
    {
      "entropy": 5.620537376403808,
      "epoch": 0.4516631005640926,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004984260364775744,
      "loss": 5.4157,
      "mean_token_accuracy": 0.1698002278804779,
      "num_tokens": 10076424.0,
      "step": 5805
    },
    {
      "entropy": 5.689474725723267,
      "epoch": 0.4520521299358102,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000498422762247404,
      "loss": 5.5609,
      "mean_token_accuracy": 0.16164169758558272,
      "num_tokens": 10086307.0,
      "step": 5810
    },
    {
      "entropy": 5.6550195693969725,
      "epoch": 0.4524411593075277,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004984194846271489,
      "loss": 5.4124,
      "mean_token_accuracy": 0.16617335975170136,
      "num_tokens": 10094954.0,
      "step": 5815
    },
    {
      "entropy": 5.625373649597168,
      "epoch": 0.4528301886792453,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004984162036168586,
      "loss": 5.5063,
      "mean_token_accuracy": 0.1653664916753769,
      "num_tokens": 10104401.0,
      "step": 5820
    },
    {
      "entropy": 5.585896444320679,
      "epoch": 0.4532192180509628,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004984129192165831,
      "loss": 5.5119,
      "mean_token_accuracy": 0.16613335609436036,
      "num_tokens": 10112596.0,
      "step": 5825
    },
    {
      "entropy": 5.704642724990845,
      "epoch": 0.4536082474226804,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004984096314263722,
      "loss": 5.5643,
      "mean_token_accuracy": 0.16454292684793473,
      "num_tokens": 10121311.0,
      "step": 5830
    },
    {
      "entropy": 5.630156230926514,
      "epoch": 0.453997276794398,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004984063402462757,
      "loss": 5.3555,
      "mean_token_accuracy": 0.17521204501390458,
      "num_tokens": 10130010.0,
      "step": 5835
    },
    {
      "entropy": 5.6026153564453125,
      "epoch": 0.45438630616611553,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004984030456763435,
      "loss": 5.4706,
      "mean_token_accuracy": 0.1734953925013542,
      "num_tokens": 10138473.0,
      "step": 5840
    },
    {
      "entropy": 5.762540626525879,
      "epoch": 0.4547753355378331,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983997477166257,
      "loss": 5.54,
      "mean_token_accuracy": 0.16557761579751967,
      "num_tokens": 10146843.0,
      "step": 5845
    },
    {
      "entropy": 5.585455226898193,
      "epoch": 0.45516436490955064,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004983964463671723,
      "loss": 5.4082,
      "mean_token_accuracy": 0.17130711823701858,
      "num_tokens": 10155319.0,
      "step": 5850
    },
    {
      "entropy": 5.660451078414917,
      "epoch": 0.45555339428126823,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983931416280334,
      "loss": 5.5784,
      "mean_token_accuracy": 0.15584017261862754,
      "num_tokens": 10164308.0,
      "step": 5855
    },
    {
      "entropy": 5.755096054077148,
      "epoch": 0.4559424236529858,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004983898334992591,
      "loss": 5.5853,
      "mean_token_accuracy": 0.16000744700431824,
      "num_tokens": 10173173.0,
      "step": 5860
    },
    {
      "entropy": 5.678262710571289,
      "epoch": 0.45633145302470335,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983865219808998,
      "loss": 5.489,
      "mean_token_accuracy": 0.16422337144613267,
      "num_tokens": 10181376.0,
      "step": 5865
    },
    {
      "entropy": 5.746049976348877,
      "epoch": 0.45672048239642093,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983832070730055,
      "loss": 5.5995,
      "mean_token_accuracy": 0.16115534156560898,
      "num_tokens": 10189701.0,
      "step": 5870
    },
    {
      "entropy": 5.6774920463562015,
      "epoch": 0.4571095117681385,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983798887756265,
      "loss": 5.4579,
      "mean_token_accuracy": 0.1671434983611107,
      "num_tokens": 10198721.0,
      "step": 5875
    },
    {
      "entropy": 5.598591899871826,
      "epoch": 0.45749854113985605,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004983765670888133,
      "loss": 5.3462,
      "mean_token_accuracy": 0.17291613519191742,
      "num_tokens": 10206428.0,
      "step": 5880
    },
    {
      "entropy": 5.562051773071289,
      "epoch": 0.45788757051157364,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983732420126163,
      "loss": 5.4937,
      "mean_token_accuracy": 0.16735068559646607,
      "num_tokens": 10214941.0,
      "step": 5885
    },
    {
      "entropy": 5.686462688446045,
      "epoch": 0.45827659988329117,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983699135470858,
      "loss": 5.4916,
      "mean_token_accuracy": 0.16933013796806334,
      "num_tokens": 10223491.0,
      "step": 5890
    },
    {
      "entropy": 5.685241460800171,
      "epoch": 0.45866562925500876,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983665816922723,
      "loss": 5.4304,
      "mean_token_accuracy": 0.16738610714673996,
      "num_tokens": 10231820.0,
      "step": 5895
    },
    {
      "entropy": 5.642031812667847,
      "epoch": 0.45905465862672634,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004983632464482267,
      "loss": 5.4196,
      "mean_token_accuracy": 0.17414960265159607,
      "num_tokens": 10239781.0,
      "step": 5900
    },
    {
      "entropy": 5.657448434829712,
      "epoch": 0.4594436879984439,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983599078149991,
      "loss": 5.4475,
      "mean_token_accuracy": 0.1785872086882591,
      "num_tokens": 10248093.0,
      "step": 5905
    },
    {
      "entropy": 5.702689075469971,
      "epoch": 0.45983271737016146,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983565657926405,
      "loss": 5.4636,
      "mean_token_accuracy": 0.16945738792419435,
      "num_tokens": 10256370.0,
      "step": 5910
    },
    {
      "entropy": 5.615186977386474,
      "epoch": 0.460221746741879,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004983532203812017,
      "loss": 5.4399,
      "mean_token_accuracy": 0.16275497525930405,
      "num_tokens": 10264924.0,
      "step": 5915
    },
    {
      "entropy": 5.643068838119507,
      "epoch": 0.4606107761135966,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983498715807331,
      "loss": 5.3804,
      "mean_token_accuracy": 0.1671242117881775,
      "num_tokens": 10273488.0,
      "step": 5920
    },
    {
      "entropy": 5.668789386749268,
      "epoch": 0.46099980548531416,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983465193912857,
      "loss": 5.4632,
      "mean_token_accuracy": 0.16634155362844466,
      "num_tokens": 10281534.0,
      "step": 5925
    },
    {
      "entropy": 5.668724584579468,
      "epoch": 0.4613888348570317,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983431638129104,
      "loss": 5.5026,
      "mean_token_accuracy": 0.16119295507669448,
      "num_tokens": 10289613.0,
      "step": 5930
    },
    {
      "entropy": 5.717075490951538,
      "epoch": 0.4617778642287493,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983398048456581,
      "loss": 5.5491,
      "mean_token_accuracy": 0.1646980419754982,
      "num_tokens": 10299027.0,
      "step": 5935
    },
    {
      "entropy": 5.717835378646851,
      "epoch": 0.4621668936004668,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004983364424895796,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16248458176851271,
      "num_tokens": 10308727.0,
      "step": 5940
    },
    {
      "entropy": 5.725033092498779,
      "epoch": 0.4625559229721844,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983330767447262,
      "loss": 5.5373,
      "mean_token_accuracy": 0.16755590587854385,
      "num_tokens": 10318129.0,
      "step": 5945
    },
    {
      "entropy": 5.624452114105225,
      "epoch": 0.462944952343902,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004983297076111489,
      "loss": 5.3765,
      "mean_token_accuracy": 0.17445236146450044,
      "num_tokens": 10326504.0,
      "step": 5950
    },
    {
      "entropy": 5.594799041748047,
      "epoch": 0.4633339817156195,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004983263350888987,
      "loss": 5.4356,
      "mean_token_accuracy": 0.17249158918857574,
      "num_tokens": 10334585.0,
      "step": 5955
    },
    {
      "entropy": 5.618805265426635,
      "epoch": 0.4637230110873371,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983229591780268,
      "loss": 5.4586,
      "mean_token_accuracy": 0.1713235542178154,
      "num_tokens": 10343328.0,
      "step": 5960
    },
    {
      "entropy": 5.618730592727661,
      "epoch": 0.46411204045905463,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004983195798785844,
      "loss": 5.4736,
      "mean_token_accuracy": 0.1642005518078804,
      "num_tokens": 10352289.0,
      "step": 5965
    },
    {
      "entropy": 5.745180225372314,
      "epoch": 0.4645010698307722,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004983161971906228,
      "loss": 5.5841,
      "mean_token_accuracy": 0.15788466483354568,
      "num_tokens": 10360847.0,
      "step": 5970
    },
    {
      "entropy": 5.670765161514282,
      "epoch": 0.4648900992024898,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004983128111141935,
      "loss": 5.4975,
      "mean_token_accuracy": 0.16726861894130707,
      "num_tokens": 10368911.0,
      "step": 5975
    },
    {
      "entropy": 5.621411466598511,
      "epoch": 0.46527912857420733,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983094216493475,
      "loss": 5.4432,
      "mean_token_accuracy": 0.17352777272462844,
      "num_tokens": 10377257.0,
      "step": 5980
    },
    {
      "entropy": 5.66666350364685,
      "epoch": 0.4656681579459249,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004983060287961367,
      "loss": 5.4989,
      "mean_token_accuracy": 0.1640869066119194,
      "num_tokens": 10386167.0,
      "step": 5985
    },
    {
      "entropy": 5.660472679138183,
      "epoch": 0.4660571873176425,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004983026325546123,
      "loss": 5.3912,
      "mean_token_accuracy": 0.164736832678318,
      "num_tokens": 10395802.0,
      "step": 5990
    },
    {
      "entropy": 5.6535501956939695,
      "epoch": 0.46644621668936004,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982992329248257,
      "loss": 5.5194,
      "mean_token_accuracy": 0.16232506781816483,
      "num_tokens": 10404836.0,
      "step": 5995
    },
    {
      "entropy": 5.737968635559082,
      "epoch": 0.4668352460610776,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004982958299068289,
      "loss": 5.493,
      "mean_token_accuracy": 0.16429309397935868,
      "num_tokens": 10413394.0,
      "step": 6000
    },
    {
      "epoch": 0.4668352460610776,
      "eval_entropy": 5.495044304110662,
      "eval_loss": 5.507410049438477,
      "eval_mean_token_accuracy": 0.17327504654821005,
      "eval_num_tokens": 10413394.0,
      "eval_runtime": 21.5979,
      "eval_samples_per_second": 1924.173,
      "eval_steps_per_second": 240.533,
      "step": 6000
    },
    {
      "entropy": 5.611471319198609,
      "epoch": 0.46722427543279516,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498292423500673,
      "loss": 5.3977,
      "mean_token_accuracy": 0.17447368502616883,
      "num_tokens": 10421916.0,
      "step": 6005
    },
    {
      "entropy": 5.626347494125366,
      "epoch": 0.46761330480451274,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004982890137064102,
      "loss": 5.4753,
      "mean_token_accuracy": 0.16418679058551788,
      "num_tokens": 10430703.0,
      "step": 6010
    },
    {
      "entropy": 5.704648208618164,
      "epoch": 0.46800233417623033,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498285600524092,
      "loss": 5.5136,
      "mean_token_accuracy": 0.16784841120243071,
      "num_tokens": 10439523.0,
      "step": 6015
    },
    {
      "entropy": 5.648009967803955,
      "epoch": 0.46839136354794786,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982821839537701,
      "loss": 5.4241,
      "mean_token_accuracy": 0.17214948534965516,
      "num_tokens": 10447777.0,
      "step": 6020
    },
    {
      "entropy": 5.638533878326416,
      "epoch": 0.46878039291966545,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004982787639954966,
      "loss": 5.3972,
      "mean_token_accuracy": 0.1707349255681038,
      "num_tokens": 10456105.0,
      "step": 6025
    },
    {
      "entropy": 5.68563323020935,
      "epoch": 0.469169422291383,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004982753406493232,
      "loss": 5.4491,
      "mean_token_accuracy": 0.1683100551366806,
      "num_tokens": 10464590.0,
      "step": 6030
    },
    {
      "entropy": 5.686050891876221,
      "epoch": 0.46955845166310056,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004982719139153018,
      "loss": 5.4714,
      "mean_token_accuracy": 0.1673478290438652,
      "num_tokens": 10472702.0,
      "step": 6035
    },
    {
      "entropy": 5.570622825622559,
      "epoch": 0.46994748103481815,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004982684837934845,
      "loss": 5.4005,
      "mean_token_accuracy": 0.1666673019528389,
      "num_tokens": 10480682.0,
      "step": 6040
    },
    {
      "entropy": 5.721953248977661,
      "epoch": 0.4703365104065357,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004982650502839234,
      "loss": 5.4953,
      "mean_token_accuracy": 0.1655719682574272,
      "num_tokens": 10488873.0,
      "step": 6045
    },
    {
      "entropy": 5.685389995574951,
      "epoch": 0.47072553977825327,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982616133866705,
      "loss": 5.4602,
      "mean_token_accuracy": 0.1705882355570793,
      "num_tokens": 10497077.0,
      "step": 6050
    },
    {
      "entropy": 5.664164638519287,
      "epoch": 0.4711145691499708,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004982581731017779,
      "loss": 5.4511,
      "mean_token_accuracy": 0.16683453023433686,
      "num_tokens": 10506072.0,
      "step": 6055
    },
    {
      "entropy": 5.685359239578247,
      "epoch": 0.4715035985216884,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982547294292979,
      "loss": 5.5081,
      "mean_token_accuracy": 0.16408928632736205,
      "num_tokens": 10515175.0,
      "step": 6060
    },
    {
      "entropy": 5.6141149520874025,
      "epoch": 0.47189262789340597,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004982512823692828,
      "loss": 5.4006,
      "mean_token_accuracy": 0.16687188446521758,
      "num_tokens": 10524800.0,
      "step": 6065
    },
    {
      "entropy": 5.649979209899902,
      "epoch": 0.4722816572651235,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004982478319217848,
      "loss": 5.5852,
      "mean_token_accuracy": 0.15808211714029313,
      "num_tokens": 10534608.0,
      "step": 6070
    },
    {
      "entropy": 5.745941495895385,
      "epoch": 0.4726706866368411,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982443780868562,
      "loss": 5.4897,
      "mean_token_accuracy": 0.16620081812143325,
      "num_tokens": 10542992.0,
      "step": 6075
    },
    {
      "entropy": 5.67744812965393,
      "epoch": 0.4730597160085587,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004982409208645496,
      "loss": 5.4874,
      "mean_token_accuracy": 0.16672162860631942,
      "num_tokens": 10551682.0,
      "step": 6080
    },
    {
      "entropy": 5.611584854125977,
      "epoch": 0.4734487453802762,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004982374602549173,
      "loss": 5.4447,
      "mean_token_accuracy": 0.16964718252420424,
      "num_tokens": 10560556.0,
      "step": 6085
    },
    {
      "entropy": 5.694293689727783,
      "epoch": 0.4738377747519938,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004982339962580119,
      "loss": 5.6142,
      "mean_token_accuracy": 0.15800638645887374,
      "num_tokens": 10569361.0,
      "step": 6090
    },
    {
      "entropy": 5.607173299789428,
      "epoch": 0.4742268041237113,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004982305288738859,
      "loss": 5.4209,
      "mean_token_accuracy": 0.17520027309656144,
      "num_tokens": 10577997.0,
      "step": 6095
    },
    {
      "entropy": 5.687979412078858,
      "epoch": 0.4746158334954289,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004982270581025919,
      "loss": 5.5251,
      "mean_token_accuracy": 0.17112909853458405,
      "num_tokens": 10587005.0,
      "step": 6100
    },
    {
      "entropy": 5.672091341018676,
      "epoch": 0.4750048628671465,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004982235839441826,
      "loss": 5.3696,
      "mean_token_accuracy": 0.17774538993835448,
      "num_tokens": 10595196.0,
      "step": 6105
    },
    {
      "entropy": 5.69416036605835,
      "epoch": 0.475393892238864,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004982201063987108,
      "loss": 5.5435,
      "mean_token_accuracy": 0.1597681984305382,
      "num_tokens": 10603796.0,
      "step": 6110
    },
    {
      "entropy": 5.689189910888672,
      "epoch": 0.4757829216105816,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982166254662292,
      "loss": 5.5103,
      "mean_token_accuracy": 0.1657646656036377,
      "num_tokens": 10612895.0,
      "step": 6115
    },
    {
      "entropy": 5.6923877716064455,
      "epoch": 0.47617195098229914,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982131411467904,
      "loss": 5.5651,
      "mean_token_accuracy": 0.16675678938627242,
      "num_tokens": 10621310.0,
      "step": 6120
    },
    {
      "entropy": 5.605409955978393,
      "epoch": 0.4765609803540167,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982096534404476,
      "loss": 5.3864,
      "mean_token_accuracy": 0.16726448982954026,
      "num_tokens": 10629598.0,
      "step": 6125
    },
    {
      "entropy": 5.692520093917847,
      "epoch": 0.4769500097257343,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982061623472535,
      "loss": 5.4087,
      "mean_token_accuracy": 0.17069395780563354,
      "num_tokens": 10638218.0,
      "step": 6130
    },
    {
      "entropy": 5.691115808486939,
      "epoch": 0.47733903909745184,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982026678672612,
      "loss": 5.516,
      "mean_token_accuracy": 0.16282628774642943,
      "num_tokens": 10647215.0,
      "step": 6135
    },
    {
      "entropy": 5.677426338195801,
      "epoch": 0.47772806846916943,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004981991700005238,
      "loss": 5.6324,
      "mean_token_accuracy": 0.15961235165596008,
      "num_tokens": 10655755.0,
      "step": 6140
    },
    {
      "entropy": 5.759636735916137,
      "epoch": 0.47811709784088696,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498195668747094,
      "loss": 5.5084,
      "mean_token_accuracy": 0.16573580950498581,
      "num_tokens": 10663969.0,
      "step": 6145
    },
    {
      "entropy": 5.6918775081634525,
      "epoch": 0.47850612721260455,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981921641070254,
      "loss": 5.3825,
      "mean_token_accuracy": 0.16804722398519517,
      "num_tokens": 10672048.0,
      "step": 6150
    },
    {
      "entropy": 5.608302307128906,
      "epoch": 0.47889515658432213,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981886560803708,
      "loss": 5.4828,
      "mean_token_accuracy": 0.16609022319316863,
      "num_tokens": 10681802.0,
      "step": 6155
    },
    {
      "entropy": 5.6948483943939205,
      "epoch": 0.47928418595603967,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981851446671838,
      "loss": 5.545,
      "mean_token_accuracy": 0.15990679562091828,
      "num_tokens": 10690618.0,
      "step": 6160
    },
    {
      "entropy": 5.67942967414856,
      "epoch": 0.47967321532775725,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004981816298675173,
      "loss": 5.4514,
      "mean_token_accuracy": 0.17377543151378633,
      "num_tokens": 10699080.0,
      "step": 6165
    },
    {
      "entropy": 5.647350740432739,
      "epoch": 0.4800622446994748,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004981781116814248,
      "loss": 5.4539,
      "mean_token_accuracy": 0.16524712294340133,
      "num_tokens": 10708939.0,
      "step": 6170
    },
    {
      "entropy": 5.69949164390564,
      "epoch": 0.48045127407119237,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004981745901089596,
      "loss": 5.3711,
      "mean_token_accuracy": 0.1681235522031784,
      "num_tokens": 10717356.0,
      "step": 6175
    },
    {
      "entropy": 5.69394736289978,
      "epoch": 0.48084030344290996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981710651501753,
      "loss": 5.4605,
      "mean_token_accuracy": 0.16272638440132142,
      "num_tokens": 10725616.0,
      "step": 6180
    },
    {
      "entropy": 5.620643281936646,
      "epoch": 0.4812293328146275,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004981675368051254,
      "loss": 5.4527,
      "mean_token_accuracy": 0.16980734765529631,
      "num_tokens": 10734337.0,
      "step": 6185
    },
    {
      "entropy": 5.645343685150147,
      "epoch": 0.4816183621863451,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004981640050738633,
      "loss": 5.4907,
      "mean_token_accuracy": 0.16745964735746383,
      "num_tokens": 10742720.0,
      "step": 6190
    },
    {
      "entropy": 5.713866233825684,
      "epoch": 0.48200739155806266,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004981604699564426,
      "loss": 5.5143,
      "mean_token_accuracy": 0.16125882863998414,
      "num_tokens": 10751809.0,
      "step": 6195
    },
    {
      "entropy": 5.643674039840699,
      "epoch": 0.4823964209297802,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004981569314529169,
      "loss": 5.3826,
      "mean_token_accuracy": 0.17199781239032746,
      "num_tokens": 10759879.0,
      "step": 6200
    },
    {
      "entropy": 5.555496120452881,
      "epoch": 0.4827854503014978,
      "grad_norm": 1.0,
      "learning_rate": 0.00049815338956334,
      "loss": 5.4444,
      "mean_token_accuracy": 0.16761718839406967,
      "num_tokens": 10769309.0,
      "step": 6205
    },
    {
      "entropy": 5.637994718551636,
      "epoch": 0.4831744796732153,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981498442877656,
      "loss": 5.3933,
      "mean_token_accuracy": 0.17291852235794067,
      "num_tokens": 10777676.0,
      "step": 6210
    },
    {
      "entropy": 5.648993921279907,
      "epoch": 0.4835635090449329,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004981462956262474,
      "loss": 5.5845,
      "mean_token_accuracy": 0.16300415694713594,
      "num_tokens": 10786278.0,
      "step": 6215
    },
    {
      "entropy": 5.711193609237671,
      "epoch": 0.4839525384166505,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004981427435788396,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1662832885980606,
      "num_tokens": 10795323.0,
      "step": 6220
    },
    {
      "entropy": 5.695321989059448,
      "epoch": 0.484341567788368,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004981391881455957,
      "loss": 5.3936,
      "mean_token_accuracy": 0.1736224412918091,
      "num_tokens": 10803978.0,
      "step": 6225
    },
    {
      "entropy": 5.671928787231446,
      "epoch": 0.4847305971600856,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004981356293265696,
      "loss": 5.4905,
      "mean_token_accuracy": 0.16860363334417344,
      "num_tokens": 10812488.0,
      "step": 6230
    },
    {
      "entropy": 5.690825319290161,
      "epoch": 0.4851196265318031,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004981320671218157,
      "loss": 5.4825,
      "mean_token_accuracy": 0.16431137025356293,
      "num_tokens": 10821076.0,
      "step": 6235
    },
    {
      "entropy": 5.602234601974487,
      "epoch": 0.4855086559035207,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981285015313877,
      "loss": 5.31,
      "mean_token_accuracy": 0.18187370002269745,
      "num_tokens": 10828634.0,
      "step": 6240
    },
    {
      "entropy": 5.634187126159668,
      "epoch": 0.4858976852752383,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004981249325553399,
      "loss": 5.4464,
      "mean_token_accuracy": 0.17350577116012572,
      "num_tokens": 10837481.0,
      "step": 6245
    },
    {
      "entropy": 5.625172281265259,
      "epoch": 0.48628671464695583,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004981213601937264,
      "loss": 5.5051,
      "mean_token_accuracy": 0.15857475399971008,
      "num_tokens": 10846729.0,
      "step": 6250
    },
    {
      "entropy": 5.709941816329956,
      "epoch": 0.4866757440186734,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004981177844466013,
      "loss": 5.4911,
      "mean_token_accuracy": 0.16200155019760132,
      "num_tokens": 10855344.0,
      "step": 6255
    },
    {
      "entropy": 5.695097494125366,
      "epoch": 0.48706477339039095,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004981142053140192,
      "loss": 5.5255,
      "mean_token_accuracy": 0.1674111381173134,
      "num_tokens": 10863934.0,
      "step": 6260
    },
    {
      "entropy": 5.626110935211182,
      "epoch": 0.48745380276210853,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004981106227960339,
      "loss": 5.5045,
      "mean_token_accuracy": 0.16477670669555664,
      "num_tokens": 10872031.0,
      "step": 6265
    },
    {
      "entropy": 5.718598461151123,
      "epoch": 0.4878428321338261,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004981070368927001,
      "loss": 5.5731,
      "mean_token_accuracy": 0.16409001350402833,
      "num_tokens": 10881385.0,
      "step": 6270
    },
    {
      "entropy": 5.678594493865967,
      "epoch": 0.48823186150554365,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000498103447604072,
      "loss": 5.4697,
      "mean_token_accuracy": 0.1685481011867523,
      "num_tokens": 10890483.0,
      "step": 6275
    },
    {
      "entropy": 5.630790519714355,
      "epoch": 0.48862089087726124,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980998549302044,
      "loss": 5.5031,
      "mean_token_accuracy": 0.17031818777322769,
      "num_tokens": 10898338.0,
      "step": 6280
    },
    {
      "entropy": 5.6111741065979,
      "epoch": 0.4890099202489788,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004980962588711516,
      "loss": 5.4897,
      "mean_token_accuracy": 0.1661595106124878,
      "num_tokens": 10908087.0,
      "step": 6285
    },
    {
      "entropy": 5.747820568084717,
      "epoch": 0.48939894962069636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000498092659426968,
      "loss": 5.4743,
      "mean_token_accuracy": 0.16227587312459946,
      "num_tokens": 10916704.0,
      "step": 6290
    },
    {
      "entropy": 5.674692535400391,
      "epoch": 0.48978797899241394,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980890565977085,
      "loss": 5.439,
      "mean_token_accuracy": 0.1698486715555191,
      "num_tokens": 10925186.0,
      "step": 6295
    },
    {
      "entropy": 5.616642427444458,
      "epoch": 0.4901770083641315,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004980854503834276,
      "loss": 5.4967,
      "mean_token_accuracy": 0.16892740577459336,
      "num_tokens": 10934247.0,
      "step": 6300
    },
    {
      "entropy": 5.74802975654602,
      "epoch": 0.49056603773584906,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980818407841802,
      "loss": 5.5229,
      "mean_token_accuracy": 0.16045583933591842,
      "num_tokens": 10944260.0,
      "step": 6305
    },
    {
      "entropy": 5.697029542922974,
      "epoch": 0.49095506710756665,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004980782278000207,
      "loss": 5.4731,
      "mean_token_accuracy": 0.16220095604658127,
      "num_tokens": 10954065.0,
      "step": 6310
    },
    {
      "entropy": 5.6517290592193605,
      "epoch": 0.4913440964792842,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004980746114310043,
      "loss": 5.5439,
      "mean_token_accuracy": 0.1659497171640396,
      "num_tokens": 10962918.0,
      "step": 6315
    },
    {
      "entropy": 5.617909145355225,
      "epoch": 0.49173312585100176,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004980709916771858,
      "loss": 5.4076,
      "mean_token_accuracy": 0.1703692078590393,
      "num_tokens": 10972356.0,
      "step": 6320
    },
    {
      "entropy": 5.616169452667236,
      "epoch": 0.4921221552227193,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004980673685386198,
      "loss": 5.4644,
      "mean_token_accuracy": 0.1629459321498871,
      "num_tokens": 10980984.0,
      "step": 6325
    },
    {
      "entropy": 5.617296028137207,
      "epoch": 0.4925111845944369,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004980637420153618,
      "loss": 5.4052,
      "mean_token_accuracy": 0.1666889727115631,
      "num_tokens": 10989809.0,
      "step": 6330
    },
    {
      "entropy": 5.63802056312561,
      "epoch": 0.49290021396615447,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004980601121074664,
      "loss": 5.453,
      "mean_token_accuracy": 0.1644183024764061,
      "num_tokens": 10998824.0,
      "step": 6335
    },
    {
      "entropy": 5.722669553756714,
      "epoch": 0.493289243337872,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980564788149889,
      "loss": 5.5222,
      "mean_token_accuracy": 0.16973644942045213,
      "num_tokens": 11006755.0,
      "step": 6340
    },
    {
      "entropy": 5.619312381744384,
      "epoch": 0.4936782727095896,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004980528421379844,
      "loss": 5.4179,
      "mean_token_accuracy": 0.17462491989135742,
      "num_tokens": 11015337.0,
      "step": 6345
    },
    {
      "entropy": 5.562566661834717,
      "epoch": 0.4940673020813071,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498049202076508,
      "loss": 5.3348,
      "mean_token_accuracy": 0.17756202816963196,
      "num_tokens": 11024186.0,
      "step": 6350
    },
    {
      "entropy": 5.6552270412445065,
      "epoch": 0.4944563314530247,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004980455586306149,
      "loss": 5.4615,
      "mean_token_accuracy": 0.1687190517783165,
      "num_tokens": 11032504.0,
      "step": 6355
    },
    {
      "entropy": 5.723153495788575,
      "epoch": 0.4948453608247423,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980419118003605,
      "loss": 5.3969,
      "mean_token_accuracy": 0.16880524754524232,
      "num_tokens": 11042252.0,
      "step": 6360
    },
    {
      "entropy": 5.703668785095215,
      "epoch": 0.4952343901964598,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980382615858001,
      "loss": 5.4826,
      "mean_token_accuracy": 0.16379437744617462,
      "num_tokens": 11051361.0,
      "step": 6365
    },
    {
      "entropy": 5.6041374683380125,
      "epoch": 0.4956234195681774,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004980346079869892,
      "loss": 5.4484,
      "mean_token_accuracy": 0.16445529013872145,
      "num_tokens": 11060822.0,
      "step": 6370
    },
    {
      "entropy": 5.6370484828948975,
      "epoch": 0.49601244893989493,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000498030951003983,
      "loss": 5.5271,
      "mean_token_accuracy": 0.16127680093050004,
      "num_tokens": 11069528.0,
      "step": 6375
    },
    {
      "entropy": 5.634170007705689,
      "epoch": 0.4964014783116125,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004980272906368371,
      "loss": 5.361,
      "mean_token_accuracy": 0.174397711455822,
      "num_tokens": 11077917.0,
      "step": 6380
    },
    {
      "entropy": 5.5964888572692875,
      "epoch": 0.4967905076833301,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004980236268856071,
      "loss": 5.4408,
      "mean_token_accuracy": 0.16141705960035324,
      "num_tokens": 11085938.0,
      "step": 6385
    },
    {
      "entropy": 5.60595817565918,
      "epoch": 0.49717953705504764,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004980199597503487,
      "loss": 5.5015,
      "mean_token_accuracy": 0.16092778593301774,
      "num_tokens": 11094898.0,
      "step": 6390
    },
    {
      "entropy": 5.640070581436158,
      "epoch": 0.4975685664267652,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980162892311171,
      "loss": 5.3664,
      "mean_token_accuracy": 0.1711999848484993,
      "num_tokens": 11103038.0,
      "step": 6395
    },
    {
      "entropy": 5.675674295425415,
      "epoch": 0.4979575957984828,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980126153279684,
      "loss": 5.4478,
      "mean_token_accuracy": 0.16367504447698594,
      "num_tokens": 11111965.0,
      "step": 6400
    },
    {
      "entropy": 5.636992788314819,
      "epoch": 0.49834662517020034,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004980089380409583,
      "loss": 5.4701,
      "mean_token_accuracy": 0.16154250949621202,
      "num_tokens": 11121482.0,
      "step": 6405
    },
    {
      "entropy": 5.66742844581604,
      "epoch": 0.4987356545419179,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004980052573701424,
      "loss": 5.4534,
      "mean_token_accuracy": 0.1701739862561226,
      "num_tokens": 11129742.0,
      "step": 6410
    },
    {
      "entropy": 5.656267976760864,
      "epoch": 0.49912468391363546,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004980015733155767,
      "loss": 5.4676,
      "mean_token_accuracy": 0.16760584563016892,
      "num_tokens": 11138388.0,
      "step": 6415
    },
    {
      "entropy": 5.474362707138061,
      "epoch": 0.49951371328535304,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004979978858773171,
      "loss": 5.2006,
      "mean_token_accuracy": 0.18340540081262588,
      "num_tokens": 11146967.0,
      "step": 6420
    },
    {
      "entropy": 5.641241121292114,
      "epoch": 0.49990274265707063,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004979941950554195,
      "loss": 5.5672,
      "mean_token_accuracy": 0.16281252056360246,
      "num_tokens": 11155888.0,
      "step": 6425
    },
    {
      "entropy": 5.7240630149841305,
      "epoch": 0.5002917720287882,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004979905008499399,
      "loss": 5.4083,
      "mean_token_accuracy": 0.17032212167978286,
      "num_tokens": 11164358.0,
      "step": 6430
    },
    {
      "entropy": 5.611014366149902,
      "epoch": 0.5006808014005057,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979868032609344,
      "loss": 5.2753,
      "mean_token_accuracy": 0.18071538507938384,
      "num_tokens": 11172416.0,
      "step": 6435
    },
    {
      "entropy": 5.640211296081543,
      "epoch": 0.5010698307722233,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004979831022884591,
      "loss": 5.4608,
      "mean_token_accuracy": 0.16421267986297608,
      "num_tokens": 11181159.0,
      "step": 6440
    },
    {
      "entropy": 5.663838481903076,
      "epoch": 0.5014588601439409,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004979793979325701,
      "loss": 5.4998,
      "mean_token_accuracy": 0.16421327739953995,
      "num_tokens": 11189725.0,
      "step": 6445
    },
    {
      "entropy": 5.672861528396607,
      "epoch": 0.5018478895156584,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979756901933236,
      "loss": 5.4741,
      "mean_token_accuracy": 0.16618643403053285,
      "num_tokens": 11197795.0,
      "step": 6450
    },
    {
      "entropy": 5.684984397888184,
      "epoch": 0.502236918887376,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497971979070776,
      "loss": 5.439,
      "mean_token_accuracy": 0.17090869694948196,
      "num_tokens": 11205352.0,
      "step": 6455
    },
    {
      "entropy": 5.633697748184204,
      "epoch": 0.5026259482590936,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979682645649834,
      "loss": 5.4625,
      "mean_token_accuracy": 0.16927655562758445,
      "num_tokens": 11214142.0,
      "step": 6460
    },
    {
      "entropy": 5.6249748229980465,
      "epoch": 0.5030149776308112,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004979645466760025,
      "loss": 5.3582,
      "mean_token_accuracy": 0.17228575199842452,
      "num_tokens": 11222565.0,
      "step": 6465
    },
    {
      "entropy": 5.6179883003234865,
      "epoch": 0.5034040070025287,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004979608254038892,
      "loss": 5.4671,
      "mean_token_accuracy": 0.1600498229265213,
      "num_tokens": 11231829.0,
      "step": 6470
    },
    {
      "entropy": 5.623617362976074,
      "epoch": 0.5037930363742462,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004979571007487003,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1706765592098236,
      "num_tokens": 11240302.0,
      "step": 6475
    },
    {
      "entropy": 5.727696180343628,
      "epoch": 0.5041820657459638,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979533727104924,
      "loss": 5.5714,
      "mean_token_accuracy": 0.15697085857391357,
      "num_tokens": 11248610.0,
      "step": 6480
    },
    {
      "entropy": 5.652313756942749,
      "epoch": 0.5045710951176814,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497949641289322,
      "loss": 5.4685,
      "mean_token_accuracy": 0.17022727876901628,
      "num_tokens": 11257441.0,
      "step": 6485
    },
    {
      "entropy": 5.590007686614991,
      "epoch": 0.504960124489399,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004979459064852456,
      "loss": 5.3465,
      "mean_token_accuracy": 0.16538669615983964,
      "num_tokens": 11265915.0,
      "step": 6490
    },
    {
      "entropy": 5.670529985427857,
      "epoch": 0.5053491538611166,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979421682983197,
      "loss": 5.4751,
      "mean_token_accuracy": 0.16173707395792009,
      "num_tokens": 11274165.0,
      "step": 6495
    },
    {
      "entropy": 5.689936256408691,
      "epoch": 0.505738183232834,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004979384267286015,
      "loss": 5.4048,
      "mean_token_accuracy": 0.17155286222696303,
      "num_tokens": 11283316.0,
      "step": 6500
    },
    {
      "entropy": 5.648772954940796,
      "epoch": 0.5061272126045516,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004979346817761475,
      "loss": 5.5451,
      "mean_token_accuracy": 0.16382537186145782,
      "num_tokens": 11292289.0,
      "step": 6505
    },
    {
      "entropy": 5.662769746780396,
      "epoch": 0.5065162419762692,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979309334410144,
      "loss": 5.4601,
      "mean_token_accuracy": 0.16324011385440826,
      "num_tokens": 11301192.0,
      "step": 6510
    },
    {
      "entropy": 5.7098266124725345,
      "epoch": 0.5069052713479868,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004979271817232594,
      "loss": 5.4808,
      "mean_token_accuracy": 0.17188810110092162,
      "num_tokens": 11309906.0,
      "step": 6515
    },
    {
      "entropy": 5.6908995628356935,
      "epoch": 0.5072943007197044,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979234266229391,
      "loss": 5.4609,
      "mean_token_accuracy": 0.16591265350580214,
      "num_tokens": 11318227.0,
      "step": 6520
    },
    {
      "entropy": 5.543266248703003,
      "epoch": 0.5076833300914219,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004979196681401106,
      "loss": 5.4885,
      "mean_token_accuracy": 0.16181913167238235,
      "num_tokens": 11327524.0,
      "step": 6525
    },
    {
      "entropy": 5.656088352203369,
      "epoch": 0.5080723594631394,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000497915906274831,
      "loss": 5.4004,
      "mean_token_accuracy": 0.17116282731294633,
      "num_tokens": 11336214.0,
      "step": 6530
    },
    {
      "entropy": 5.681649684906006,
      "epoch": 0.508461388834857,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979121410271574,
      "loss": 5.2699,
      "mean_token_accuracy": 0.18444954007863998,
      "num_tokens": 11344551.0,
      "step": 6535
    },
    {
      "entropy": 5.509459018707275,
      "epoch": 0.5088504182065746,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979083723971468,
      "loss": 5.3258,
      "mean_token_accuracy": 0.1782999262213707,
      "num_tokens": 11353180.0,
      "step": 6540
    },
    {
      "entropy": 5.462130117416382,
      "epoch": 0.5092394475782922,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004979046003848564,
      "loss": 5.3146,
      "mean_token_accuracy": 0.17773108780384064,
      "num_tokens": 11361343.0,
      "step": 6545
    },
    {
      "entropy": 5.581699991226197,
      "epoch": 0.5096284769500097,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979008249903435,
      "loss": 5.4102,
      "mean_token_accuracy": 0.17800912261009216,
      "num_tokens": 11369915.0,
      "step": 6550
    },
    {
      "entropy": 5.63182520866394,
      "epoch": 0.5100175063217273,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004978970462136655,
      "loss": 5.4432,
      "mean_token_accuracy": 0.16393458247184753,
      "num_tokens": 11379813.0,
      "step": 6555
    },
    {
      "entropy": 5.6643470287322994,
      "epoch": 0.5104065356934449,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978932640548794,
      "loss": 5.3959,
      "mean_token_accuracy": 0.1692471981048584,
      "num_tokens": 11388029.0,
      "step": 6560
    },
    {
      "entropy": 5.615650320053101,
      "epoch": 0.5107955650651624,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978894785140429,
      "loss": 5.3138,
      "mean_token_accuracy": 0.17617870420217513,
      "num_tokens": 11395852.0,
      "step": 6565
    },
    {
      "entropy": 5.594472980499267,
      "epoch": 0.51118459443688,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004978856895912132,
      "loss": 5.424,
      "mean_token_accuracy": 0.1696908950805664,
      "num_tokens": 11404756.0,
      "step": 6570
    },
    {
      "entropy": 5.736187171936035,
      "epoch": 0.5115736238085975,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004978818972864481,
      "loss": 5.5854,
      "mean_token_accuracy": 0.16184193193912505,
      "num_tokens": 11413301.0,
      "step": 6575
    },
    {
      "entropy": 5.711348819732666,
      "epoch": 0.5119626531803151,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004978781015998048,
      "loss": 5.3515,
      "mean_token_accuracy": 0.1738321989774704,
      "num_tokens": 11421034.0,
      "step": 6580
    },
    {
      "entropy": 5.609139537811279,
      "epoch": 0.5123516825520327,
      "grad_norm": 1.0,
      "learning_rate": 0.0004978743025313413,
      "loss": 5.3942,
      "mean_token_accuracy": 0.17445635795593262,
      "num_tokens": 11429066.0,
      "step": 6585
    },
    {
      "entropy": 5.498823261260986,
      "epoch": 0.5127407119237503,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004978705000811148,
      "loss": 5.4094,
      "mean_token_accuracy": 0.16928456127643585,
      "num_tokens": 11438109.0,
      "step": 6590
    },
    {
      "entropy": 5.658320903778076,
      "epoch": 0.5131297412954678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004978666942491832,
      "loss": 5.4287,
      "mean_token_accuracy": 0.1691526874899864,
      "num_tokens": 11446787.0,
      "step": 6595
    },
    {
      "entropy": 5.59825587272644,
      "epoch": 0.5135187706671853,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004978628850356043,
      "loss": 5.3867,
      "mean_token_accuracy": 0.16849370002746583,
      "num_tokens": 11456262.0,
      "step": 6600
    },
    {
      "entropy": 5.638675928115845,
      "epoch": 0.5139078000389029,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004978590724404358,
      "loss": 5.3882,
      "mean_token_accuracy": 0.17555086463689804,
      "num_tokens": 11464297.0,
      "step": 6605
    },
    {
      "entropy": 5.6504510879516605,
      "epoch": 0.5142968294106205,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004978552564637356,
      "loss": 5.4598,
      "mean_token_accuracy": 0.16988648921251298,
      "num_tokens": 11473519.0,
      "step": 6610
    },
    {
      "entropy": 5.662782049179077,
      "epoch": 0.5146858587823381,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004978514371055616,
      "loss": 5.4553,
      "mean_token_accuracy": 0.16260288804769515,
      "num_tokens": 11482249.0,
      "step": 6615
    },
    {
      "entropy": 5.5754162788391115,
      "epoch": 0.5150748881540557,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004978476143659718,
      "loss": 5.4179,
      "mean_token_accuracy": 0.1698406532406807,
      "num_tokens": 11490687.0,
      "step": 6620
    },
    {
      "entropy": 5.561270093917846,
      "epoch": 0.5154639175257731,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004978437882450241,
      "loss": 5.4276,
      "mean_token_accuracy": 0.1648566707968712,
      "num_tokens": 11500225.0,
      "step": 6625
    },
    {
      "entropy": 5.6289667129516605,
      "epoch": 0.5158529468974907,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004978399587427766,
      "loss": 5.3552,
      "mean_token_accuracy": 0.1658610612154007,
      "num_tokens": 11508116.0,
      "step": 6630
    },
    {
      "entropy": 5.542283058166504,
      "epoch": 0.5162419762692083,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004978361258592874,
      "loss": 5.4362,
      "mean_token_accuracy": 0.16429463773965836,
      "num_tokens": 11516525.0,
      "step": 6635
    },
    {
      "entropy": 5.606814956665039,
      "epoch": 0.5166310056409259,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004978322895946147,
      "loss": 5.4068,
      "mean_token_accuracy": 0.17431761771440507,
      "num_tokens": 11525434.0,
      "step": 6640
    },
    {
      "entropy": 5.676286029815674,
      "epoch": 0.5170200350126435,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004978284499488167,
      "loss": 5.4635,
      "mean_token_accuracy": 0.16793732196092606,
      "num_tokens": 11534318.0,
      "step": 6645
    },
    {
      "entropy": 5.6349205493927,
      "epoch": 0.5174090643843611,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004978246069219516,
      "loss": 5.3908,
      "mean_token_accuracy": 0.1799973353743553,
      "num_tokens": 11542777.0,
      "step": 6650
    },
    {
      "entropy": 5.625831985473633,
      "epoch": 0.5177980937560785,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004978207605140777,
      "loss": 5.3896,
      "mean_token_accuracy": 0.17461561858654023,
      "num_tokens": 11550888.0,
      "step": 6655
    },
    {
      "entropy": 5.626785707473755,
      "epoch": 0.5181871231277961,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004978169107252534,
      "loss": 5.4318,
      "mean_token_accuracy": 0.1628247007727623,
      "num_tokens": 11560047.0,
      "step": 6660
    },
    {
      "entropy": 5.671987104415893,
      "epoch": 0.5185761524995137,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978130575555373,
      "loss": 5.4189,
      "mean_token_accuracy": 0.1725052371621132,
      "num_tokens": 11568862.0,
      "step": 6665
    },
    {
      "entropy": 5.550399494171143,
      "epoch": 0.5189651818712313,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004978092010049877,
      "loss": 5.3735,
      "mean_token_accuracy": 0.1731133446097374,
      "num_tokens": 11577509.0,
      "step": 6670
    },
    {
      "entropy": 5.743370056152344,
      "epoch": 0.5193542112429489,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497805341073663,
      "loss": 5.4857,
      "mean_token_accuracy": 0.17034248411655425,
      "num_tokens": 11585902.0,
      "step": 6675
    },
    {
      "entropy": 5.667796611785889,
      "epoch": 0.5197432406146664,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004978014777616219,
      "loss": 5.4979,
      "mean_token_accuracy": 0.17023638486862183,
      "num_tokens": 11594895.0,
      "step": 6680
    },
    {
      "entropy": 5.5753960609436035,
      "epoch": 0.520132269986384,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004977976110689229,
      "loss": 5.3123,
      "mean_token_accuracy": 0.1708584502339363,
      "num_tokens": 11604080.0,
      "step": 6685
    },
    {
      "entropy": 5.665987968444824,
      "epoch": 0.5205212993581015,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000497793740995625,
      "loss": 5.4386,
      "mean_token_accuracy": 0.16633949279785157,
      "num_tokens": 11613370.0,
      "step": 6690
    },
    {
      "entropy": 5.585418844223023,
      "epoch": 0.5209103287298191,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004977898675417866,
      "loss": 5.4211,
      "mean_token_accuracy": 0.16649907678365708,
      "num_tokens": 11621913.0,
      "step": 6695
    },
    {
      "entropy": 5.642942237854004,
      "epoch": 0.5212993581015367,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004977859907074664,
      "loss": 5.5436,
      "mean_token_accuracy": 0.16430994272232055,
      "num_tokens": 11630393.0,
      "step": 6700
    },
    {
      "entropy": 5.720639753341675,
      "epoch": 0.5216883874732542,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004977821104927236,
      "loss": 5.4928,
      "mean_token_accuracy": 0.1631960764527321,
      "num_tokens": 11639038.0,
      "step": 6705
    },
    {
      "entropy": 5.618670892715454,
      "epoch": 0.5220774168449718,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004977782268976167,
      "loss": 5.406,
      "mean_token_accuracy": 0.16851633042097092,
      "num_tokens": 11647355.0,
      "step": 6710
    },
    {
      "entropy": 5.596138334274292,
      "epoch": 0.5224664462166894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004977743399222048,
      "loss": 5.3802,
      "mean_token_accuracy": 0.1682727590203285,
      "num_tokens": 11655899.0,
      "step": 6715
    },
    {
      "entropy": 5.725822591781617,
      "epoch": 0.522855475588407,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004977704495665471,
      "loss": 5.5674,
      "mean_token_accuracy": 0.15976732969284058,
      "num_tokens": 11664602.0,
      "step": 6720
    },
    {
      "entropy": 5.6776047706604,
      "epoch": 0.5232445049601245,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004977665558307023,
      "loss": 5.3917,
      "mean_token_accuracy": 0.17342305481433867,
      "num_tokens": 11673011.0,
      "step": 6725
    },
    {
      "entropy": 5.750084304809571,
      "epoch": 0.523633534331842,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004977626587147295,
      "loss": 5.5673,
      "mean_token_accuracy": 0.1633073076605797,
      "num_tokens": 11681989.0,
      "step": 6730
    },
    {
      "entropy": 5.703148746490479,
      "epoch": 0.5240225637035596,
      "grad_norm": 1.0,
      "learning_rate": 0.000497758758218688,
      "loss": 5.4916,
      "mean_token_accuracy": 0.16183123290538787,
      "num_tokens": 11690502.0,
      "step": 6735
    },
    {
      "entropy": 5.641050243377686,
      "epoch": 0.5244115930752772,
      "grad_norm": 1.0,
      "learning_rate": 0.0004977548543426368,
      "loss": 5.4935,
      "mean_token_accuracy": 0.166514340788126,
      "num_tokens": 11699594.0,
      "step": 6740
    },
    {
      "entropy": 5.584754896163941,
      "epoch": 0.5248006224469948,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004977509470866353,
      "loss": 5.3668,
      "mean_token_accuracy": 0.16964129209518433,
      "num_tokens": 11708009.0,
      "step": 6745
    },
    {
      "entropy": 5.702879905700684,
      "epoch": 0.5251896518187124,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004977470364507427,
      "loss": 5.5323,
      "mean_token_accuracy": 0.16005744338035582,
      "num_tokens": 11716592.0,
      "step": 6750
    },
    {
      "entropy": 5.688350057601928,
      "epoch": 0.5255786811904298,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004977431224350184,
      "loss": 5.559,
      "mean_token_accuracy": 0.15319210886955262,
      "num_tokens": 11725894.0,
      "step": 6755
    },
    {
      "entropy": 5.726099252700806,
      "epoch": 0.5259677105621474,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977392050395217,
      "loss": 5.4079,
      "mean_token_accuracy": 0.16904182434082032,
      "num_tokens": 11734141.0,
      "step": 6760
    },
    {
      "entropy": 5.658994817733765,
      "epoch": 0.526356739933865,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004977352842643121,
      "loss": 5.5067,
      "mean_token_accuracy": 0.164360873401165,
      "num_tokens": 11742539.0,
      "step": 6765
    },
    {
      "entropy": 5.661802816390991,
      "epoch": 0.5267457693055826,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497731360109449,
      "loss": 5.4757,
      "mean_token_accuracy": 0.16561661511659623,
      "num_tokens": 11751649.0,
      "step": 6770
    },
    {
      "entropy": 5.689319801330567,
      "epoch": 0.5271347986773002,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004977274325749922,
      "loss": 5.4887,
      "mean_token_accuracy": 0.16493163704872132,
      "num_tokens": 11760376.0,
      "step": 6775
    },
    {
      "entropy": 5.683128690719604,
      "epoch": 0.5275238280490177,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004977235016610011,
      "loss": 5.4508,
      "mean_token_accuracy": 0.16838639378547668,
      "num_tokens": 11768644.0,
      "step": 6780
    },
    {
      "entropy": 5.702654647827148,
      "epoch": 0.5279128574207352,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004977195673675353,
      "loss": 5.521,
      "mean_token_accuracy": 0.16457068622112275,
      "num_tokens": 11777282.0,
      "step": 6785
    },
    {
      "entropy": 5.630511379241943,
      "epoch": 0.5283018867924528,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004977156296946545,
      "loss": 5.4285,
      "mean_token_accuracy": 0.16902912259101868,
      "num_tokens": 11786108.0,
      "step": 6790
    },
    {
      "entropy": 5.619226408004761,
      "epoch": 0.5286909161641704,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004977116886424187,
      "loss": 5.4717,
      "mean_token_accuracy": 0.1616060495376587,
      "num_tokens": 11795151.0,
      "step": 6795
    },
    {
      "entropy": 5.745349264144897,
      "epoch": 0.529079945535888,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004977077442108873,
      "loss": 5.4159,
      "mean_token_accuracy": 0.17375196963548661,
      "num_tokens": 11803971.0,
      "step": 6800
    },
    {
      "entropy": 5.609159469604492,
      "epoch": 0.5294689749076055,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004977037964001204,
      "loss": 5.4662,
      "mean_token_accuracy": 0.17144315242767333,
      "num_tokens": 11812483.0,
      "step": 6805
    },
    {
      "entropy": 5.580070114135742,
      "epoch": 0.5298580042793231,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000497699845210178,
      "loss": 5.3558,
      "mean_token_accuracy": 0.17689975798130037,
      "num_tokens": 11821144.0,
      "step": 6810
    },
    {
      "entropy": 5.630408430099488,
      "epoch": 0.5302470336510406,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004976958906411197,
      "loss": 5.3189,
      "mean_token_accuracy": 0.17868188321590422,
      "num_tokens": 11829298.0,
      "step": 6815
    },
    {
      "entropy": 5.668645286560059,
      "epoch": 0.5306360630227582,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004976919326930059,
      "loss": 5.4219,
      "mean_token_accuracy": 0.17015438079833983,
      "num_tokens": 11837558.0,
      "step": 6820
    },
    {
      "entropy": 5.546228361129761,
      "epoch": 0.5310250923944758,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004976879713658964,
      "loss": 5.3114,
      "mean_token_accuracy": 0.17407533675432205,
      "num_tokens": 11845770.0,
      "step": 6825
    },
    {
      "entropy": 5.677445554733277,
      "epoch": 0.5314141217661933,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004976840066598514,
      "loss": 5.4971,
      "mean_token_accuracy": 0.15888483226299285,
      "num_tokens": 11854351.0,
      "step": 6830
    },
    {
      "entropy": 5.77203950881958,
      "epoch": 0.5318031511379109,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976800385749309,
      "loss": 5.4912,
      "mean_token_accuracy": 0.17434257715940477,
      "num_tokens": 11862610.0,
      "step": 6835
    },
    {
      "entropy": 5.652606201171875,
      "epoch": 0.5321921805096285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004976760671111954,
      "loss": 5.4788,
      "mean_token_accuracy": 0.17035529613494874,
      "num_tokens": 11871078.0,
      "step": 6840
    },
    {
      "entropy": 5.609366941452026,
      "epoch": 0.532581209881346,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497672092268705,
      "loss": 5.3485,
      "mean_token_accuracy": 0.17742927223443986,
      "num_tokens": 11879303.0,
      "step": 6845
    },
    {
      "entropy": 5.624922180175782,
      "epoch": 0.5329702392530636,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049766811404752,
      "loss": 5.3957,
      "mean_token_accuracy": 0.17183460891246796,
      "num_tokens": 11887855.0,
      "step": 6850
    },
    {
      "entropy": 5.701442718505859,
      "epoch": 0.5333592686247812,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004976641324477008,
      "loss": 5.3599,
      "mean_token_accuracy": 0.17164410650730133,
      "num_tokens": 11896262.0,
      "step": 6855
    },
    {
      "entropy": 5.511118984222412,
      "epoch": 0.5337482979964987,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004976601474693077,
      "loss": 5.2781,
      "mean_token_accuracy": 0.1787315785884857,
      "num_tokens": 11905140.0,
      "step": 6860
    },
    {
      "entropy": 5.542644214630127,
      "epoch": 0.5341373273682163,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004976561591124014,
      "loss": 5.3198,
      "mean_token_accuracy": 0.17594115883111955,
      "num_tokens": 11913177.0,
      "step": 6865
    },
    {
      "entropy": 5.453828239440918,
      "epoch": 0.5345263567399339,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004976521673770421,
      "loss": 5.3571,
      "mean_token_accuracy": 0.17943618297576905,
      "num_tokens": 11921955.0,
      "step": 6870
    },
    {
      "entropy": 5.603758478164673,
      "epoch": 0.5349153861116515,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004976481722632907,
      "loss": 5.3558,
      "mean_token_accuracy": 0.1703154921531677,
      "num_tokens": 11930976.0,
      "step": 6875
    },
    {
      "entropy": 5.619595861434936,
      "epoch": 0.535304415483369,
      "grad_norm": 1.0,
      "learning_rate": 0.0004976441737712076,
      "loss": 5.2819,
      "mean_token_accuracy": 0.17539516985416412,
      "num_tokens": 11939708.0,
      "step": 6880
    },
    {
      "entropy": 5.6227837085723875,
      "epoch": 0.5356934448550865,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004976401719008536,
      "loss": 5.4718,
      "mean_token_accuracy": 0.16340460404753684,
      "num_tokens": 11949176.0,
      "step": 6885
    },
    {
      "entropy": 5.623278713226318,
      "epoch": 0.5360824742268041,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004976361666522893,
      "loss": 5.3277,
      "mean_token_accuracy": 0.17831586450338363,
      "num_tokens": 11958089.0,
      "step": 6890
    },
    {
      "entropy": 5.647400808334351,
      "epoch": 0.5364715035985217,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976321580255755,
      "loss": 5.4209,
      "mean_token_accuracy": 0.17419404089450835,
      "num_tokens": 11966719.0,
      "step": 6895
    },
    {
      "entropy": 5.651475811004639,
      "epoch": 0.5368605329702393,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004976281460207731,
      "loss": 5.4901,
      "mean_token_accuracy": 0.1658528283238411,
      "num_tokens": 11975959.0,
      "step": 6900
    },
    {
      "entropy": 5.63256721496582,
      "epoch": 0.5372495623419569,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000497624130637943,
      "loss": 5.4354,
      "mean_token_accuracy": 0.1661873921751976,
      "num_tokens": 11984282.0,
      "step": 6905
    },
    {
      "entropy": 5.659536361694336,
      "epoch": 0.5376385917136743,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497620111877146,
      "loss": 5.4367,
      "mean_token_accuracy": 0.1674356371164322,
      "num_tokens": 11993289.0,
      "step": 6910
    },
    {
      "entropy": 5.584860134124756,
      "epoch": 0.5380276210853919,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004976160897384431,
      "loss": 5.3668,
      "mean_token_accuracy": 0.17173303812742233,
      "num_tokens": 12002425.0,
      "step": 6915
    },
    {
      "entropy": 5.628754281997681,
      "epoch": 0.5384166504571095,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004976120642218955,
      "loss": 5.3784,
      "mean_token_accuracy": 0.17149677276611328,
      "num_tokens": 12010435.0,
      "step": 6920
    },
    {
      "entropy": 5.562549209594726,
      "epoch": 0.5388056798288271,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000497608035327564,
      "loss": 5.4306,
      "mean_token_accuracy": 0.1634228304028511,
      "num_tokens": 12018768.0,
      "step": 6925
    },
    {
      "entropy": 5.64377384185791,
      "epoch": 0.5391947092005447,
      "grad_norm": 0.984375,
      "learning_rate": 0.00049760400305551,
      "loss": 5.4494,
      "mean_token_accuracy": 0.17214796990156173,
      "num_tokens": 12027617.0,
      "step": 6930
    },
    {
      "entropy": 5.663191556930542,
      "epoch": 0.5395837385722622,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004975999674057944,
      "loss": 5.4171,
      "mean_token_accuracy": 0.16715274453163148,
      "num_tokens": 12036024.0,
      "step": 6935
    },
    {
      "entropy": 5.671573257446289,
      "epoch": 0.5399727679439797,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004975959283784787,
      "loss": 5.4519,
      "mean_token_accuracy": 0.16841286569833755,
      "num_tokens": 12044206.0,
      "step": 6940
    },
    {
      "entropy": 5.598822069168091,
      "epoch": 0.5403617973156973,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004975918859736241,
      "loss": 5.3335,
      "mean_token_accuracy": 0.17251801639795303,
      "num_tokens": 12052294.0,
      "step": 6945
    },
    {
      "entropy": 5.585722303390503,
      "epoch": 0.5407508266874149,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004975878401912919,
      "loss": 5.453,
      "mean_token_accuracy": 0.16552573442459106,
      "num_tokens": 12060446.0,
      "step": 6950
    },
    {
      "entropy": 5.657906723022461,
      "epoch": 0.5411398560591325,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004975837910315436,
      "loss": 5.5333,
      "mean_token_accuracy": 0.1649618551135063,
      "num_tokens": 12069267.0,
      "step": 6955
    },
    {
      "entropy": 5.696007919311524,
      "epoch": 0.54152888543085,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004975797384944405,
      "loss": 5.4524,
      "mean_token_accuracy": 0.1656179204583168,
      "num_tokens": 12077808.0,
      "step": 6960
    },
    {
      "entropy": 5.620603752136231,
      "epoch": 0.5419179148025676,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497575682580044,
      "loss": 5.4231,
      "mean_token_accuracy": 0.16909924298524856,
      "num_tokens": 12086893.0,
      "step": 6965
    },
    {
      "entropy": 5.599455213546753,
      "epoch": 0.5423069441742852,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497571623288416,
      "loss": 5.3045,
      "mean_token_accuracy": 0.1750321552157402,
      "num_tokens": 12095462.0,
      "step": 6970
    },
    {
      "entropy": 5.646379852294922,
      "epoch": 0.5426959735460027,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004975675606196177,
      "loss": 5.4068,
      "mean_token_accuracy": 0.1716737598180771,
      "num_tokens": 12103477.0,
      "step": 6975
    },
    {
      "entropy": 5.693296003341675,
      "epoch": 0.5430850029177203,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497563494573711,
      "loss": 5.5203,
      "mean_token_accuracy": 0.1662907287478447,
      "num_tokens": 12112752.0,
      "step": 6980
    },
    {
      "entropy": 5.695676946640015,
      "epoch": 0.5434740322894378,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004975594251507575,
      "loss": 5.4567,
      "mean_token_accuracy": 0.1623358443379402,
      "num_tokens": 12121883.0,
      "step": 6985
    },
    {
      "entropy": 5.662042140960693,
      "epoch": 0.5438630616611554,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000497555352350819,
      "loss": 5.3843,
      "mean_token_accuracy": 0.1695976585149765,
      "num_tokens": 12130316.0,
      "step": 6990
    },
    {
      "entropy": 5.607014846801758,
      "epoch": 0.544252091032873,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975512761739572,
      "loss": 5.3817,
      "mean_token_accuracy": 0.1685868486762047,
      "num_tokens": 12138856.0,
      "step": 6995
    },
    {
      "entropy": 5.5710813999176025,
      "epoch": 0.5446411204045906,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497547196620234,
      "loss": 5.449,
      "mean_token_accuracy": 0.1645628333091736,
      "num_tokens": 12147933.0,
      "step": 7000
    },
    {
      "entropy": 5.600865125656128,
      "epoch": 0.5450301497763081,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004975431136897114,
      "loss": 5.3418,
      "mean_token_accuracy": 0.17403218895196915,
      "num_tokens": 12156529.0,
      "step": 7005
    },
    {
      "entropy": 5.6149626731872555,
      "epoch": 0.5454191791480256,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004975390273824512,
      "loss": 5.2957,
      "mean_token_accuracy": 0.17711113393306732,
      "num_tokens": 12165233.0,
      "step": 7010
    },
    {
      "entropy": 5.656652784347534,
      "epoch": 0.5458082085197432,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004975349376985155,
      "loss": 5.4905,
      "mean_token_accuracy": 0.16208152174949647,
      "num_tokens": 12173260.0,
      "step": 7015
    },
    {
      "entropy": 5.633213138580322,
      "epoch": 0.5461972378914608,
      "grad_norm": 1.0,
      "learning_rate": 0.0004975308446379663,
      "loss": 5.4872,
      "mean_token_accuracy": 0.1661377176642418,
      "num_tokens": 12181676.0,
      "step": 7020
    },
    {
      "entropy": 5.663449764251709,
      "epoch": 0.5465862672631784,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004975267482008657,
      "loss": 5.3998,
      "mean_token_accuracy": 0.16634807139635086,
      "num_tokens": 12189485.0,
      "step": 7025
    },
    {
      "entropy": 5.61206283569336,
      "epoch": 0.546975296634896,
      "grad_norm": 0.90625,
      "learning_rate": 0.000497522648387276,
      "loss": 5.318,
      "mean_token_accuracy": 0.17358834743499757,
      "num_tokens": 12198429.0,
      "step": 7030
    },
    {
      "entropy": 5.6052594661712645,
      "epoch": 0.5473643260066134,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004975185451972592,
      "loss": 5.4449,
      "mean_token_accuracy": 0.17202031910419463,
      "num_tokens": 12207404.0,
      "step": 7035
    },
    {
      "entropy": 5.670430994033813,
      "epoch": 0.547753355378331,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004975144386308776,
      "loss": 5.4222,
      "mean_token_accuracy": 0.1661749705672264,
      "num_tokens": 12216301.0,
      "step": 7040
    },
    {
      "entropy": 5.589423131942749,
      "epoch": 0.5481423847500486,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975103286881936,
      "loss": 5.4065,
      "mean_token_accuracy": 0.17458733767271042,
      "num_tokens": 12224873.0,
      "step": 7045
    },
    {
      "entropy": 5.603284597396851,
      "epoch": 0.5485314141217662,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004975062153692696,
      "loss": 5.3784,
      "mean_token_accuracy": 0.17105503529310226,
      "num_tokens": 12233439.0,
      "step": 7050
    },
    {
      "entropy": 5.656129550933838,
      "epoch": 0.5489204434934838,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975020986741678,
      "loss": 5.4493,
      "mean_token_accuracy": 0.16752919703722,
      "num_tokens": 12242235.0,
      "step": 7055
    },
    {
      "entropy": 5.597258377075195,
      "epoch": 0.5493094728652014,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004974979786029509,
      "loss": 5.388,
      "mean_token_accuracy": 0.17538175135850906,
      "num_tokens": 12251295.0,
      "step": 7060
    },
    {
      "entropy": 5.629272842407227,
      "epoch": 0.5496985022369189,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004974938551556814,
      "loss": 5.3133,
      "mean_token_accuracy": 0.17724365890026092,
      "num_tokens": 12260130.0,
      "step": 7065
    },
    {
      "entropy": 5.5907104969024655,
      "epoch": 0.5500875316086364,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004974897283324217,
      "loss": 5.393,
      "mean_token_accuracy": 0.16955282539129257,
      "num_tokens": 12269542.0,
      "step": 7070
    },
    {
      "entropy": 5.630454397201538,
      "epoch": 0.550476560980354,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974855981332343,
      "loss": 5.4673,
      "mean_token_accuracy": 0.17138559967279435,
      "num_tokens": 12278174.0,
      "step": 7075
    },
    {
      "entropy": 5.663634777069092,
      "epoch": 0.5508655903520716,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974814645581823,
      "loss": 5.5483,
      "mean_token_accuracy": 0.16026232093572618,
      "num_tokens": 12287651.0,
      "step": 7080
    },
    {
      "entropy": 5.76173586845398,
      "epoch": 0.5512546197237892,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004974773276073282,
      "loss": 5.5228,
      "mean_token_accuracy": 0.1664297416806221,
      "num_tokens": 12296009.0,
      "step": 7085
    },
    {
      "entropy": 5.6218091487884525,
      "epoch": 0.5516436490955067,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974731872807347,
      "loss": 5.3927,
      "mean_token_accuracy": 0.16824719458818435,
      "num_tokens": 12305406.0,
      "step": 7090
    },
    {
      "entropy": 5.65089635848999,
      "epoch": 0.5520326784672243,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004974690435784647,
      "loss": 5.45,
      "mean_token_accuracy": 0.1661311000585556,
      "num_tokens": 12314169.0,
      "step": 7095
    },
    {
      "entropy": 5.569345140457154,
      "epoch": 0.5524217078389418,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004974648965005811,
      "loss": 5.331,
      "mean_token_accuracy": 0.17652279734611512,
      "num_tokens": 12323019.0,
      "step": 7100
    },
    {
      "entropy": 5.595728015899658,
      "epoch": 0.5528107372106594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974607460471466,
      "loss": 5.435,
      "mean_token_accuracy": 0.16858349591493607,
      "num_tokens": 12331491.0,
      "step": 7105
    },
    {
      "entropy": 5.6596444606781,
      "epoch": 0.553199766582377,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004974565922182246,
      "loss": 5.398,
      "mean_token_accuracy": 0.1735352173447609,
      "num_tokens": 12339647.0,
      "step": 7110
    },
    {
      "entropy": 5.648700618743897,
      "epoch": 0.5535887959540945,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004974524350138777,
      "loss": 5.4002,
      "mean_token_accuracy": 0.1696406900882721,
      "num_tokens": 12348469.0,
      "step": 7115
    },
    {
      "entropy": 5.645704698562622,
      "epoch": 0.5539778253258121,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004974482744341693,
      "loss": 5.3714,
      "mean_token_accuracy": 0.17332192361354828,
      "num_tokens": 12357168.0,
      "step": 7120
    },
    {
      "entropy": 5.5719810962677006,
      "epoch": 0.5543668546975297,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974441104791624,
      "loss": 5.3745,
      "mean_token_accuracy": 0.17122909426689148,
      "num_tokens": 12365750.0,
      "step": 7125
    },
    {
      "entropy": 5.639922761917115,
      "epoch": 0.5547558840692473,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004974399431489201,
      "loss": 5.4018,
      "mean_token_accuracy": 0.17219276428222657,
      "num_tokens": 12374086.0,
      "step": 7130
    },
    {
      "entropy": 5.663942813873291,
      "epoch": 0.5551449134409648,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974357724435057,
      "loss": 5.3642,
      "mean_token_accuracy": 0.1792665883898735,
      "num_tokens": 12381997.0,
      "step": 7135
    },
    {
      "entropy": 5.646653938293457,
      "epoch": 0.5555339428126823,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004974315983629825,
      "loss": 5.4462,
      "mean_token_accuracy": 0.17241283059120177,
      "num_tokens": 12391083.0,
      "step": 7140
    },
    {
      "entropy": 5.634599733352661,
      "epoch": 0.5559229721843999,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004974274209074139,
      "loss": 5.371,
      "mean_token_accuracy": 0.17252092063426971,
      "num_tokens": 12400016.0,
      "step": 7145
    },
    {
      "entropy": 5.685489320755005,
      "epoch": 0.5563120015561175,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974232400768632,
      "loss": 5.4818,
      "mean_token_accuracy": 0.16982473880052568,
      "num_tokens": 12409255.0,
      "step": 7150
    },
    {
      "entropy": 5.641375494003296,
      "epoch": 0.5567010309278351,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004974190558713939,
      "loss": 5.3884,
      "mean_token_accuracy": 0.17358123064041137,
      "num_tokens": 12417765.0,
      "step": 7155
    },
    {
      "entropy": 5.669316005706787,
      "epoch": 0.5570900602995527,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974148682910694,
      "loss": 5.6177,
      "mean_token_accuracy": 0.15360698252916336,
      "num_tokens": 12427588.0,
      "step": 7160
    },
    {
      "entropy": 5.59427547454834,
      "epoch": 0.5574790896712701,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004974106773359533,
      "loss": 5.289,
      "mean_token_accuracy": 0.17874647974967955,
      "num_tokens": 12435773.0,
      "step": 7165
    },
    {
      "entropy": 5.605812740325928,
      "epoch": 0.5578681190429877,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004974064830061091,
      "loss": 5.4123,
      "mean_token_accuracy": 0.17299440503120422,
      "num_tokens": 12443837.0,
      "step": 7170
    },
    {
      "entropy": 5.572979688644409,
      "epoch": 0.5582571484147053,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004974022853016006,
      "loss": 5.4106,
      "mean_token_accuracy": 0.17499408423900603,
      "num_tokens": 12452945.0,
      "step": 7175
    },
    {
      "entropy": 5.595126533508301,
      "epoch": 0.5586461777864229,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973980842224913,
      "loss": 5.3213,
      "mean_token_accuracy": 0.18019433468580245,
      "num_tokens": 12461008.0,
      "step": 7180
    },
    {
      "entropy": 5.541195869445801,
      "epoch": 0.5590352071581405,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004973938797688452,
      "loss": 5.3315,
      "mean_token_accuracy": 0.16913951933383942,
      "num_tokens": 12469314.0,
      "step": 7185
    },
    {
      "entropy": 5.607437181472778,
      "epoch": 0.559424236529858,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004973896719407259,
      "loss": 5.3747,
      "mean_token_accuracy": 0.16569244116544724,
      "num_tokens": 12477540.0,
      "step": 7190
    },
    {
      "entropy": 5.611822509765625,
      "epoch": 0.5598132659015755,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004973854607381972,
      "loss": 5.4272,
      "mean_token_accuracy": 0.17706743329763414,
      "num_tokens": 12486894.0,
      "step": 7195
    },
    {
      "entropy": 5.660370969772339,
      "epoch": 0.5602022952732931,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004973812461613232,
      "loss": 5.4215,
      "mean_token_accuracy": 0.16543299257755278,
      "num_tokens": 12496213.0,
      "step": 7200
    },
    {
      "entropy": 5.621604681015015,
      "epoch": 0.5605913246450107,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004973770282101677,
      "loss": 5.3941,
      "mean_token_accuracy": 0.17001705318689347,
      "num_tokens": 12504160.0,
      "step": 7205
    },
    {
      "entropy": 5.5064887523651125,
      "epoch": 0.5609803540167283,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004973728068847947,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17627592831850053,
      "num_tokens": 12511915.0,
      "step": 7210
    },
    {
      "entropy": 5.633753728866577,
      "epoch": 0.5613693833884458,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004973685821852684,
      "loss": 5.3721,
      "mean_token_accuracy": 0.17439102977514268,
      "num_tokens": 12520400.0,
      "step": 7215
    },
    {
      "entropy": 5.592429780960083,
      "epoch": 0.5617584127601634,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004973643541116528,
      "loss": 5.295,
      "mean_token_accuracy": 0.1770132914185524,
      "num_tokens": 12528724.0,
      "step": 7220
    },
    {
      "entropy": 5.6063238143920895,
      "epoch": 0.562147442131881,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000497360122664012,
      "loss": 5.4863,
      "mean_token_accuracy": 0.17149095386266708,
      "num_tokens": 12537378.0,
      "step": 7225
    },
    {
      "entropy": 5.597244024276733,
      "epoch": 0.5625364715035985,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004973558878424104,
      "loss": 5.3994,
      "mean_token_accuracy": 0.17489453703165053,
      "num_tokens": 12546080.0,
      "step": 7230
    },
    {
      "entropy": 5.684036159515381,
      "epoch": 0.5629255008753161,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004973516496469119,
      "loss": 5.4159,
      "mean_token_accuracy": 0.1706525832414627,
      "num_tokens": 12554526.0,
      "step": 7235
    },
    {
      "entropy": 5.616761255264282,
      "epoch": 0.5633145302470336,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004973474080775811,
      "loss": 5.3895,
      "mean_token_accuracy": 0.17201378792524338,
      "num_tokens": 12563018.0,
      "step": 7240
    },
    {
      "entropy": 5.597792434692383,
      "epoch": 0.5637035596187512,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004973431631344824,
      "loss": 5.4425,
      "mean_token_accuracy": 0.1725260868668556,
      "num_tokens": 12571611.0,
      "step": 7245
    },
    {
      "entropy": 5.68755464553833,
      "epoch": 0.5640925889904688,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049733891481768,
      "loss": 5.5059,
      "mean_token_accuracy": 0.1666056454181671,
      "num_tokens": 12580168.0,
      "step": 7250
    },
    {
      "entropy": 5.59826512336731,
      "epoch": 0.5644816183621864,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004973346631272384,
      "loss": 5.4141,
      "mean_token_accuracy": 0.17382865399122238,
      "num_tokens": 12589843.0,
      "step": 7255
    },
    {
      "entropy": 5.685404348373413,
      "epoch": 0.5648706477339039,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004973304080632222,
      "loss": 5.4488,
      "mean_token_accuracy": 0.16553327292203904,
      "num_tokens": 12598219.0,
      "step": 7260
    },
    {
      "entropy": 5.603097867965698,
      "epoch": 0.5652596771056215,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000497326149625696,
      "loss": 5.4304,
      "mean_token_accuracy": 0.16851159632205964,
      "num_tokens": 12606665.0,
      "step": 7265
    },
    {
      "entropy": 5.603170919418335,
      "epoch": 0.565648706477339,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004973218878147244,
      "loss": 5.3684,
      "mean_token_accuracy": 0.17258040308952333,
      "num_tokens": 12615062.0,
      "step": 7270
    },
    {
      "entropy": 5.682673454284668,
      "epoch": 0.5660377358490566,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004973176226303719,
      "loss": 5.4455,
      "mean_token_accuracy": 0.1684245362877846,
      "num_tokens": 12623923.0,
      "step": 7275
    },
    {
      "entropy": 5.619794845581055,
      "epoch": 0.5664267652207742,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004973133540727033,
      "loss": 5.3834,
      "mean_token_accuracy": 0.16963649094104766,
      "num_tokens": 12632579.0,
      "step": 7280
    },
    {
      "entropy": 5.571510648727417,
      "epoch": 0.5668157945924918,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004973090821417835,
      "loss": 5.3358,
      "mean_token_accuracy": 0.17332665473222733,
      "num_tokens": 12640847.0,
      "step": 7285
    },
    {
      "entropy": 5.655331659317016,
      "epoch": 0.5672048239642093,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004973048068376772,
      "loss": 5.3288,
      "mean_token_accuracy": 0.17651396989822388,
      "num_tokens": 12649020.0,
      "step": 7290
    },
    {
      "entropy": 5.5143170833587645,
      "epoch": 0.5675938533359268,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004973005281604492,
      "loss": 5.2662,
      "mean_token_accuracy": 0.17431186586618425,
      "num_tokens": 12657569.0,
      "step": 7295
    },
    {
      "entropy": 5.692615413665772,
      "epoch": 0.5679828827076444,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972962461101646,
      "loss": 5.5137,
      "mean_token_accuracy": 0.16529098376631737,
      "num_tokens": 12666681.0,
      "step": 7300
    },
    {
      "entropy": 5.616756820678711,
      "epoch": 0.568371912079362,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004972919606868883,
      "loss": 5.3323,
      "mean_token_accuracy": 0.17070917636156083,
      "num_tokens": 12675210.0,
      "step": 7305
    },
    {
      "entropy": 5.661812496185303,
      "epoch": 0.5687609414510796,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004972876718906852,
      "loss": 5.3856,
      "mean_token_accuracy": 0.17086227983236313,
      "num_tokens": 12684019.0,
      "step": 7310
    },
    {
      "entropy": 5.607829952239991,
      "epoch": 0.5691499708227972,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004972833797216206,
      "loss": 5.4218,
      "mean_token_accuracy": 0.1609044313430786,
      "num_tokens": 12692891.0,
      "step": 7315
    },
    {
      "entropy": 5.591648054122925,
      "epoch": 0.5695390001945146,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004972790841797595,
      "loss": 5.3389,
      "mean_token_accuracy": 0.1735210597515106,
      "num_tokens": 12700606.0,
      "step": 7320
    },
    {
      "entropy": 5.58307032585144,
      "epoch": 0.5699280295662322,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497274785265167,
      "loss": 5.4367,
      "mean_token_accuracy": 0.16894268244504929,
      "num_tokens": 12709116.0,
      "step": 7325
    },
    {
      "entropy": 5.636340570449829,
      "epoch": 0.5703170589379498,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004972704829779086,
      "loss": 5.3984,
      "mean_token_accuracy": 0.170769339799881,
      "num_tokens": 12717484.0,
      "step": 7330
    },
    {
      "entropy": 5.540655183792114,
      "epoch": 0.5707060883096674,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972661773180492,
      "loss": 5.3223,
      "mean_token_accuracy": 0.17410922795534134,
      "num_tokens": 12726216.0,
      "step": 7335
    },
    {
      "entropy": 5.556302165985107,
      "epoch": 0.571095117681385,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004972618682856545,
      "loss": 5.3093,
      "mean_token_accuracy": 0.17505342066287993,
      "num_tokens": 12735312.0,
      "step": 7340
    },
    {
      "entropy": 5.59719409942627,
      "epoch": 0.5714841470531025,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004972575558807896,
      "loss": 5.4374,
      "mean_token_accuracy": 0.16841867715120315,
      "num_tokens": 12744641.0,
      "step": 7345
    },
    {
      "entropy": 5.6559624671936035,
      "epoch": 0.57187317642482,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004972532401035201,
      "loss": 5.4043,
      "mean_token_accuracy": 0.1704619497060776,
      "num_tokens": 12753951.0,
      "step": 7350
    },
    {
      "entropy": 5.688356494903564,
      "epoch": 0.5722622057965376,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004972489209539114,
      "loss": 5.418,
      "mean_token_accuracy": 0.16977932006120683,
      "num_tokens": 12762571.0,
      "step": 7355
    },
    {
      "entropy": 5.673377943038941,
      "epoch": 0.5726512351682552,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004972445984320292,
      "loss": 5.484,
      "mean_token_accuracy": 0.16849375218153,
      "num_tokens": 12771275.0,
      "step": 7360
    },
    {
      "entropy": 5.620267629623413,
      "epoch": 0.5730402645399728,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004972402725379388,
      "loss": 5.3931,
      "mean_token_accuracy": 0.17000342756509781,
      "num_tokens": 12780488.0,
      "step": 7365
    },
    {
      "entropy": 5.646515369415283,
      "epoch": 0.5734292939116903,
      "grad_norm": 1.09375,
      "learning_rate": 0.000497235943271706,
      "loss": 5.408,
      "mean_token_accuracy": 0.17405337244272232,
      "num_tokens": 12788588.0,
      "step": 7370
    },
    {
      "entropy": 5.5791943073272705,
      "epoch": 0.5738183232834079,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972316106333966,
      "loss": 5.3517,
      "mean_token_accuracy": 0.16888998448848724,
      "num_tokens": 12797010.0,
      "step": 7375
    },
    {
      "entropy": 5.655337905883789,
      "epoch": 0.5742073526551255,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004972272746230761,
      "loss": 5.442,
      "mean_token_accuracy": 0.17081831395626068,
      "num_tokens": 12805430.0,
      "step": 7380
    },
    {
      "entropy": 5.704648160934449,
      "epoch": 0.574596382026843,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004972229352408104,
      "loss": 5.4147,
      "mean_token_accuracy": 0.17035266011953354,
      "num_tokens": 12813912.0,
      "step": 7385
    },
    {
      "entropy": 5.688452529907226,
      "epoch": 0.5749854113985606,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004972185924866655,
      "loss": 5.4272,
      "mean_token_accuracy": 0.1671118125319481,
      "num_tokens": 12822421.0,
      "step": 7390
    },
    {
      "entropy": 5.53705153465271,
      "epoch": 0.5753744407702781,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004972142463607071,
      "loss": 5.3079,
      "mean_token_accuracy": 0.1757025510072708,
      "num_tokens": 12830670.0,
      "step": 7395
    },
    {
      "entropy": 5.633034515380859,
      "epoch": 0.5757634701419957,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004972098968630012,
      "loss": 5.431,
      "mean_token_accuracy": 0.16999536752700806,
      "num_tokens": 12839887.0,
      "step": 7400
    },
    {
      "entropy": 5.541968441009521,
      "epoch": 0.5761524995137133,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004972055439936137,
      "loss": 5.3334,
      "mean_token_accuracy": 0.1776442050933838,
      "num_tokens": 12848028.0,
      "step": 7405
    },
    {
      "entropy": 5.519324254989624,
      "epoch": 0.5765415288854309,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004972011877526109,
      "loss": 5.3389,
      "mean_token_accuracy": 0.17134140133857728,
      "num_tokens": 12857248.0,
      "step": 7410
    },
    {
      "entropy": 5.639415216445923,
      "epoch": 0.5769305582571485,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004971968281400587,
      "loss": 5.3294,
      "mean_token_accuracy": 0.17492977529764175,
      "num_tokens": 12866352.0,
      "step": 7415
    },
    {
      "entropy": 5.6588198184967045,
      "epoch": 0.5773195876288659,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004971924651560233,
      "loss": 5.4357,
      "mean_token_accuracy": 0.16394705325365067,
      "num_tokens": 12874548.0,
      "step": 7420
    },
    {
      "entropy": 5.5939655780792235,
      "epoch": 0.5777086170005835,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004971880988005709,
      "loss": 5.2928,
      "mean_token_accuracy": 0.17573003321886063,
      "num_tokens": 12883172.0,
      "step": 7425
    },
    {
      "entropy": 5.526439380645752,
      "epoch": 0.5780976463723011,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004971837290737677,
      "loss": 5.3223,
      "mean_token_accuracy": 0.1796867400407791,
      "num_tokens": 12891776.0,
      "step": 7430
    },
    {
      "entropy": 5.583204650878907,
      "epoch": 0.5784866757440187,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004971793559756801,
      "loss": 5.2974,
      "mean_token_accuracy": 0.17945968955755234,
      "num_tokens": 12900530.0,
      "step": 7435
    },
    {
      "entropy": 5.538809108734131,
      "epoch": 0.5788757051157363,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004971749795063744,
      "loss": 5.3276,
      "mean_token_accuracy": 0.17095972299575807,
      "num_tokens": 12909260.0,
      "step": 7440
    },
    {
      "entropy": 5.569579696655273,
      "epoch": 0.5792647344874537,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000497170599665917,
      "loss": 5.3305,
      "mean_token_accuracy": 0.16970742046833037,
      "num_tokens": 12918143.0,
      "step": 7445
    },
    {
      "entropy": 5.609571313858032,
      "epoch": 0.5796537638591713,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004971662164543743,
      "loss": 5.3917,
      "mean_token_accuracy": 0.1706962466239929,
      "num_tokens": 12926361.0,
      "step": 7450
    },
    {
      "entropy": 5.564591646194458,
      "epoch": 0.5800427932308889,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004971618298718131,
      "loss": 5.3313,
      "mean_token_accuracy": 0.17570846676826476,
      "num_tokens": 12935080.0,
      "step": 7455
    },
    {
      "entropy": 5.648508977890015,
      "epoch": 0.5804318226026065,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004971574399182995,
      "loss": 5.4895,
      "mean_token_accuracy": 0.16634377092123032,
      "num_tokens": 12944078.0,
      "step": 7460
    },
    {
      "entropy": 5.659052038192749,
      "epoch": 0.5808208519743241,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004971530465939005,
      "loss": 5.3972,
      "mean_token_accuracy": 0.16960996389389038,
      "num_tokens": 12952800.0,
      "step": 7465
    },
    {
      "entropy": 5.715012264251709,
      "epoch": 0.5812098813460417,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004971486498986825,
      "loss": 5.4961,
      "mean_token_accuracy": 0.16457866877317429,
      "num_tokens": 12962402.0,
      "step": 7470
    },
    {
      "entropy": 5.568425273895263,
      "epoch": 0.5815989107177592,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004971442498327124,
      "loss": 5.2584,
      "mean_token_accuracy": 0.17899834513664245,
      "num_tokens": 12971116.0,
      "step": 7475
    },
    {
      "entropy": 5.5052838802337645,
      "epoch": 0.5819879400894767,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004971398463960568,
      "loss": 5.3686,
      "mean_token_accuracy": 0.1670986607670784,
      "num_tokens": 12979570.0,
      "step": 7480
    },
    {
      "entropy": 5.640371656417846,
      "epoch": 0.5823769694611943,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004971354395887827,
      "loss": 5.3951,
      "mean_token_accuracy": 0.17526982724666595,
      "num_tokens": 12987942.0,
      "step": 7485
    },
    {
      "entropy": 5.622890377044678,
      "epoch": 0.5827659988329119,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004971310294109569,
      "loss": 5.4293,
      "mean_token_accuracy": 0.17028453648090364,
      "num_tokens": 12996499.0,
      "step": 7490
    },
    {
      "entropy": 5.529724359512329,
      "epoch": 0.5831550282046295,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004971266158626461,
      "loss": 5.3675,
      "mean_token_accuracy": 0.17635761350393295,
      "num_tokens": 13005622.0,
      "step": 7495
    },
    {
      "entropy": 5.594239330291748,
      "epoch": 0.583544057576347,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004971221989439177,
      "loss": 5.4032,
      "mean_token_accuracy": 0.16471988409757615,
      "num_tokens": 13014417.0,
      "step": 7500
    },
    {
      "entropy": 5.586181688308716,
      "epoch": 0.5839330869480646,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004971177786548384,
      "loss": 5.3209,
      "mean_token_accuracy": 0.17067594975233077,
      "num_tokens": 13022887.0,
      "step": 7505
    },
    {
      "entropy": 5.564024019241333,
      "epoch": 0.5843221163197821,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004971133549954753,
      "loss": 5.4208,
      "mean_token_accuracy": 0.1676742047071457,
      "num_tokens": 13031676.0,
      "step": 7510
    },
    {
      "entropy": 5.591061210632324,
      "epoch": 0.5847111456914997,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004971089279658954,
      "loss": 5.3809,
      "mean_token_accuracy": 0.17073456943035126,
      "num_tokens": 13040489.0,
      "step": 7515
    },
    {
      "entropy": 5.645109987258911,
      "epoch": 0.5851001750632173,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004971044975661663,
      "loss": 5.298,
      "mean_token_accuracy": 0.17565406411886214,
      "num_tokens": 13048280.0,
      "step": 7520
    },
    {
      "entropy": 5.553524398803711,
      "epoch": 0.5854892044349348,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004971000637963549,
      "loss": 5.2769,
      "mean_token_accuracy": 0.18950254768133162,
      "num_tokens": 13056471.0,
      "step": 7525
    },
    {
      "entropy": 5.616096782684326,
      "epoch": 0.5858782338066524,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004970956266565285,
      "loss": 5.4636,
      "mean_token_accuracy": 0.168794621527195,
      "num_tokens": 13065289.0,
      "step": 7530
    },
    {
      "entropy": 5.677121305465699,
      "epoch": 0.58626726317837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004970911861467544,
      "loss": 5.4795,
      "mean_token_accuracy": 0.17004801630973815,
      "num_tokens": 13073648.0,
      "step": 7535
    },
    {
      "entropy": 5.6547130107879635,
      "epoch": 0.5866562925500876,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004970867422671002,
      "loss": 5.4862,
      "mean_token_accuracy": 0.17260521799325942,
      "num_tokens": 13082205.0,
      "step": 7540
    },
    {
      "entropy": 5.654366731643677,
      "epoch": 0.5870453219218051,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497082295017633,
      "loss": 5.3632,
      "mean_token_accuracy": 0.169638891518116,
      "num_tokens": 13090970.0,
      "step": 7545
    },
    {
      "entropy": 5.633034324645996,
      "epoch": 0.5874343512935226,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970778443984206,
      "loss": 5.2772,
      "mean_token_accuracy": 0.18007944971323014,
      "num_tokens": 13098290.0,
      "step": 7550
    },
    {
      "entropy": 5.531872987747192,
      "epoch": 0.5878233806652402,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004970733904095303,
      "loss": 5.2325,
      "mean_token_accuracy": 0.18482622057199477,
      "num_tokens": 13106971.0,
      "step": 7555
    },
    {
      "entropy": 5.663135814666748,
      "epoch": 0.5882124100369578,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004970689330510298,
      "loss": 5.5028,
      "mean_token_accuracy": 0.16574273109436036,
      "num_tokens": 13115742.0,
      "step": 7560
    },
    {
      "entropy": 5.635857486724854,
      "epoch": 0.5886014394086754,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004970644723229868,
      "loss": 5.4773,
      "mean_token_accuracy": 0.1727609857916832,
      "num_tokens": 13124192.0,
      "step": 7565
    },
    {
      "entropy": 5.580559301376343,
      "epoch": 0.588990468780393,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004970600082254687,
      "loss": 5.3013,
      "mean_token_accuracy": 0.17216529250144957,
      "num_tokens": 13133510.0,
      "step": 7570
    },
    {
      "entropy": 5.604172897338867,
      "epoch": 0.5893794981521104,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004970555407585436,
      "loss": 5.4111,
      "mean_token_accuracy": 0.17594430446624756,
      "num_tokens": 13141704.0,
      "step": 7575
    },
    {
      "entropy": 5.523994970321655,
      "epoch": 0.589768527523828,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497051069922279,
      "loss": 5.2853,
      "mean_token_accuracy": 0.16935604214668273,
      "num_tokens": 13150559.0,
      "step": 7580
    },
    {
      "entropy": 5.558675527572632,
      "epoch": 0.5901575568955456,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004970465957167429,
      "loss": 5.3639,
      "mean_token_accuracy": 0.17506027966737747,
      "num_tokens": 13159428.0,
      "step": 7585
    },
    {
      "entropy": 5.58739275932312,
      "epoch": 0.5905465862672632,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497042118142003,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16214175373315812,
      "num_tokens": 13168299.0,
      "step": 7590
    },
    {
      "entropy": 5.6119704246521,
      "epoch": 0.5909356156389808,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004970376371981275,
      "loss": 5.3608,
      "mean_token_accuracy": 0.17236538380384445,
      "num_tokens": 13176906.0,
      "step": 7595
    },
    {
      "entropy": 5.6741773128509525,
      "epoch": 0.5913246450106983,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004970331528851842,
      "loss": 5.4567,
      "mean_token_accuracy": 0.16617743521928788,
      "num_tokens": 13186180.0,
      "step": 7600
    },
    {
      "entropy": 5.561134624481201,
      "epoch": 0.5917136743824158,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004970286652032412,
      "loss": 5.2933,
      "mean_token_accuracy": 0.1796172395348549,
      "num_tokens": 13195111.0,
      "step": 7605
    },
    {
      "entropy": 5.601275491714477,
      "epoch": 0.5921027037541334,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004970241741523667,
      "loss": 5.3604,
      "mean_token_accuracy": 0.17476796954870225,
      "num_tokens": 13204170.0,
      "step": 7610
    },
    {
      "entropy": 5.5979756832122805,
      "epoch": 0.592491733125851,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004970196797326286,
      "loss": 5.364,
      "mean_token_accuracy": 0.17164454609155655,
      "num_tokens": 13213479.0,
      "step": 7615
    },
    {
      "entropy": 5.593438625335693,
      "epoch": 0.5928807624975686,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004970151819440955,
      "loss": 5.3883,
      "mean_token_accuracy": 0.17189265340566634,
      "num_tokens": 13222182.0,
      "step": 7620
    },
    {
      "entropy": 5.526077461242676,
      "epoch": 0.5932697918692861,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004970106807868351,
      "loss": 5.2641,
      "mean_token_accuracy": 0.17490562349557875,
      "num_tokens": 13230681.0,
      "step": 7625
    },
    {
      "entropy": 5.697098302841186,
      "epoch": 0.5936588212410037,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000497006176260916,
      "loss": 5.446,
      "mean_token_accuracy": 0.1732114225625992,
      "num_tokens": 13239254.0,
      "step": 7630
    },
    {
      "entropy": 5.534705781936646,
      "epoch": 0.5940478506127213,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004970016683664068,
      "loss": 5.3178,
      "mean_token_accuracy": 0.1772586613893509,
      "num_tokens": 13248317.0,
      "step": 7635
    },
    {
      "entropy": 5.653423500061035,
      "epoch": 0.5944368799844388,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004969971571033754,
      "loss": 5.3687,
      "mean_token_accuracy": 0.17507459074258805,
      "num_tokens": 13256977.0,
      "step": 7640
    },
    {
      "entropy": 5.635997629165649,
      "epoch": 0.5948259093561564,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004969926424718906,
      "loss": 5.4371,
      "mean_token_accuracy": 0.17063387334346772,
      "num_tokens": 13266012.0,
      "step": 7645
    },
    {
      "entropy": 5.618613338470459,
      "epoch": 0.5952149387278739,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004969881244720207,
      "loss": 5.4522,
      "mean_token_accuracy": 0.17322476655244828,
      "num_tokens": 13275248.0,
      "step": 7650
    },
    {
      "entropy": 5.6184052467346195,
      "epoch": 0.5956039680995915,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004969836031038343,
      "loss": 5.3183,
      "mean_token_accuracy": 0.16789155900478364,
      "num_tokens": 13284235.0,
      "step": 7655
    },
    {
      "entropy": 5.635403156280518,
      "epoch": 0.5959929974713091,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004969790783674002,
      "loss": 5.4,
      "mean_token_accuracy": 0.1698325291275978,
      "num_tokens": 13292423.0,
      "step": 7660
    },
    {
      "entropy": 5.590408039093018,
      "epoch": 0.5963820268430267,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969745502627868,
      "loss": 5.4235,
      "mean_token_accuracy": 0.167571958899498,
      "num_tokens": 13300936.0,
      "step": 7665
    },
    {
      "entropy": 5.5580121040344235,
      "epoch": 0.5967710562147442,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004969700187900629,
      "loss": 5.2224,
      "mean_token_accuracy": 0.18108144104480745,
      "num_tokens": 13308881.0,
      "step": 7670
    },
    {
      "entropy": 5.607459640502929,
      "epoch": 0.5971600855864618,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969654839492973,
      "loss": 5.4123,
      "mean_token_accuracy": 0.17161239981651305,
      "num_tokens": 13317864.0,
      "step": 7675
    },
    {
      "entropy": 5.620231771469117,
      "epoch": 0.5975491149581793,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004969609457405588,
      "loss": 5.3725,
      "mean_token_accuracy": 0.1709914728999138,
      "num_tokens": 13326990.0,
      "step": 7680
    },
    {
      "entropy": 5.692099857330322,
      "epoch": 0.5979381443298969,
      "grad_norm": 1.0,
      "learning_rate": 0.0004969564041639163,
      "loss": 5.5486,
      "mean_token_accuracy": 0.16365605890750884,
      "num_tokens": 13336437.0,
      "step": 7685
    },
    {
      "entropy": 5.628011226654053,
      "epoch": 0.5983271737016145,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004969518592194386,
      "loss": 5.383,
      "mean_token_accuracy": 0.16583366096019744,
      "num_tokens": 13345368.0,
      "step": 7690
    },
    {
      "entropy": 5.559656143188477,
      "epoch": 0.5987162030733321,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004969473109071946,
      "loss": 5.4253,
      "mean_token_accuracy": 0.1732177883386612,
      "num_tokens": 13354171.0,
      "step": 7695
    },
    {
      "entropy": 5.661171102523804,
      "epoch": 0.5991052324450497,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969427592272535,
      "loss": 5.4133,
      "mean_token_accuracy": 0.16940583735704423,
      "num_tokens": 13362912.0,
      "step": 7700
    },
    {
      "entropy": 5.636720609664917,
      "epoch": 0.5994942618167671,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004969382041796843,
      "loss": 5.3092,
      "mean_token_accuracy": 0.17516656816005707,
      "num_tokens": 13370913.0,
      "step": 7705
    },
    {
      "entropy": 5.62378077507019,
      "epoch": 0.5998832911884847,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000496933645764556,
      "loss": 5.4739,
      "mean_token_accuracy": 0.16723166555166244,
      "num_tokens": 13379557.0,
      "step": 7710
    },
    {
      "entropy": 5.5516557693481445,
      "epoch": 0.6002723205602023,
      "grad_norm": 1.0,
      "learning_rate": 0.0004969290839819381,
      "loss": 5.3666,
      "mean_token_accuracy": 0.1747287005186081,
      "num_tokens": 13388129.0,
      "step": 7715
    },
    {
      "entropy": 5.645891332626343,
      "epoch": 0.6006613499319199,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969245188318994,
      "loss": 5.4869,
      "mean_token_accuracy": 0.17049409300088883,
      "num_tokens": 13396713.0,
      "step": 7720
    },
    {
      "entropy": 5.645609188079834,
      "epoch": 0.6010503793036375,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004969199503145093,
      "loss": 5.4195,
      "mean_token_accuracy": 0.17086775004863738,
      "num_tokens": 13406171.0,
      "step": 7725
    },
    {
      "entropy": 5.566971063613892,
      "epoch": 0.601439408675355,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004969153784298374,
      "loss": 5.3476,
      "mean_token_accuracy": 0.17182642966508865,
      "num_tokens": 13415107.0,
      "step": 7730
    },
    {
      "entropy": 5.60140700340271,
      "epoch": 0.6018284380470725,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004969108031779527,
      "loss": 5.4544,
      "mean_token_accuracy": 0.1706515446305275,
      "num_tokens": 13423533.0,
      "step": 7735
    },
    {
      "entropy": 5.673208951950073,
      "epoch": 0.6022174674187901,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969062245589247,
      "loss": 5.4364,
      "mean_token_accuracy": 0.16617097109556198,
      "num_tokens": 13432798.0,
      "step": 7740
    },
    {
      "entropy": 5.657231378555298,
      "epoch": 0.6026064967905077,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004969016425728231,
      "loss": 5.3837,
      "mean_token_accuracy": 0.17782800644636154,
      "num_tokens": 13441123.0,
      "step": 7745
    },
    {
      "entropy": 5.566253709793091,
      "epoch": 0.6029955261622253,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004968970572197172,
      "loss": 5.304,
      "mean_token_accuracy": 0.17716853320598602,
      "num_tokens": 13449548.0,
      "step": 7750
    },
    {
      "entropy": 5.599495124816895,
      "epoch": 0.6033845555339428,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004968924684996765,
      "loss": 5.3814,
      "mean_token_accuracy": 0.1734702154994011,
      "num_tokens": 13458022.0,
      "step": 7755
    },
    {
      "entropy": 5.591137886047363,
      "epoch": 0.6037735849056604,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000496887876412771,
      "loss": 5.352,
      "mean_token_accuracy": 0.1717812716960907,
      "num_tokens": 13466871.0,
      "step": 7760
    },
    {
      "entropy": 5.614107513427735,
      "epoch": 0.6041626142773779,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004968832809590701,
      "loss": 5.3703,
      "mean_token_accuracy": 0.17414651662111283,
      "num_tokens": 13475659.0,
      "step": 7765
    },
    {
      "entropy": 5.625330400466919,
      "epoch": 0.6045516436490955,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004968786821386435,
      "loss": 5.3962,
      "mean_token_accuracy": 0.16906533837318422,
      "num_tokens": 13484146.0,
      "step": 7770
    },
    {
      "entropy": 5.612520551681518,
      "epoch": 0.6049406730208131,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004968740799515611,
      "loss": 5.3949,
      "mean_token_accuracy": 0.17522106766700746,
      "num_tokens": 13492831.0,
      "step": 7775
    },
    {
      "entropy": 5.607007646560669,
      "epoch": 0.6053297023925306,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004968694743978926,
      "loss": 5.3678,
      "mean_token_accuracy": 0.17839196920394898,
      "num_tokens": 13501640.0,
      "step": 7780
    },
    {
      "entropy": 5.655982160568238,
      "epoch": 0.6057187317642482,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004968648654777081,
      "loss": 5.3845,
      "mean_token_accuracy": 0.1761223405599594,
      "num_tokens": 13509816.0,
      "step": 7785
    },
    {
      "entropy": 5.650158643722534,
      "epoch": 0.6061077611359658,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004968602531910773,
      "loss": 5.4491,
      "mean_token_accuracy": 0.17165745943784713,
      "num_tokens": 13518384.0,
      "step": 7790
    },
    {
      "entropy": 5.587653398513794,
      "epoch": 0.6064967905076833,
      "grad_norm": 1.0,
      "learning_rate": 0.0004968556375380704,
      "loss": 5.3126,
      "mean_token_accuracy": 0.17027076035737992,
      "num_tokens": 13527115.0,
      "step": 7795
    },
    {
      "entropy": 5.6321692943573,
      "epoch": 0.6068858198794009,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004968510185187571,
      "loss": 5.5346,
      "mean_token_accuracy": 0.16949589401483536,
      "num_tokens": 13535743.0,
      "step": 7800
    },
    {
      "entropy": 5.591909599304199,
      "epoch": 0.6072748492511184,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000496846396133208,
      "loss": 5.3394,
      "mean_token_accuracy": 0.1780050903558731,
      "num_tokens": 13544167.0,
      "step": 7805
    },
    {
      "entropy": 5.585002040863037,
      "epoch": 0.607663878622836,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004968417703814928,
      "loss": 5.4004,
      "mean_token_accuracy": 0.1679321452975273,
      "num_tokens": 13552291.0,
      "step": 7810
    },
    {
      "entropy": 5.6585938930511475,
      "epoch": 0.6080529079945536,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004968371412636818,
      "loss": 5.4355,
      "mean_token_accuracy": 0.17032116800546646,
      "num_tokens": 13561185.0,
      "step": 7815
    },
    {
      "entropy": 5.682382202148437,
      "epoch": 0.6084419373662712,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004968325087798453,
      "loss": 5.3238,
      "mean_token_accuracy": 0.18022632747888565,
      "num_tokens": 13569815.0,
      "step": 7820
    },
    {
      "entropy": 5.588699150085449,
      "epoch": 0.6088309667379888,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004968278729300536,
      "loss": 5.4331,
      "mean_token_accuracy": 0.16961714923381804,
      "num_tokens": 13578882.0,
      "step": 7825
    },
    {
      "entropy": 5.57780966758728,
      "epoch": 0.6092199961097062,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004968232337143769,
      "loss": 5.3144,
      "mean_token_accuracy": 0.17948854118585586,
      "num_tokens": 13587479.0,
      "step": 7830
    },
    {
      "entropy": 5.515082788467407,
      "epoch": 0.6096090254814238,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004968185911328858,
      "loss": 5.2152,
      "mean_token_accuracy": 0.17272976338863372,
      "num_tokens": 13596049.0,
      "step": 7835
    },
    {
      "entropy": 5.643992233276367,
      "epoch": 0.6099980548531414,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004968139451856506,
      "loss": 5.4925,
      "mean_token_accuracy": 0.1680475354194641,
      "num_tokens": 13604766.0,
      "step": 7840
    },
    {
      "entropy": 5.6659993648529055,
      "epoch": 0.610387084224859,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004968092958727419,
      "loss": 5.4614,
      "mean_token_accuracy": 0.1690959021449089,
      "num_tokens": 13614051.0,
      "step": 7845
    },
    {
      "entropy": 5.634029293060303,
      "epoch": 0.6107761135965766,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00049680464319423,
      "loss": 5.4037,
      "mean_token_accuracy": 0.17126537561416627,
      "num_tokens": 13623100.0,
      "step": 7850
    },
    {
      "entropy": 5.656065464019775,
      "epoch": 0.6111651429682942,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967999871501858,
      "loss": 5.4086,
      "mean_token_accuracy": 0.17563506215810776,
      "num_tokens": 13631732.0,
      "step": 7855
    },
    {
      "entropy": 5.617393684387207,
      "epoch": 0.6115541723400116,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004967953277406798,
      "loss": 5.3901,
      "mean_token_accuracy": 0.1730097770690918,
      "num_tokens": 13640355.0,
      "step": 7860
    },
    {
      "entropy": 5.635977268218994,
      "epoch": 0.6119432017117292,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004967906649657828,
      "loss": 5.4603,
      "mean_token_accuracy": 0.1736888751387596,
      "num_tokens": 13649108.0,
      "step": 7865
    },
    {
      "entropy": 5.610578441619873,
      "epoch": 0.6123322310834468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004967859988255655,
      "loss": 5.3027,
      "mean_token_accuracy": 0.1875598818063736,
      "num_tokens": 13656765.0,
      "step": 7870
    },
    {
      "entropy": 5.5898693084716795,
      "epoch": 0.6127212604551644,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004967813293200985,
      "loss": 5.3808,
      "mean_token_accuracy": 0.17080039381980897,
      "num_tokens": 13664692.0,
      "step": 7875
    },
    {
      "entropy": 5.491356992721558,
      "epoch": 0.613110289826882,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004967766564494529,
      "loss": 5.2597,
      "mean_token_accuracy": 0.17703975439071656,
      "num_tokens": 13672913.0,
      "step": 7880
    },
    {
      "entropy": 5.547981834411621,
      "epoch": 0.6134993191985995,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004967719802136996,
      "loss": 5.2479,
      "mean_token_accuracy": 0.1814624547958374,
      "num_tokens": 13682427.0,
      "step": 7885
    },
    {
      "entropy": 5.589805030822754,
      "epoch": 0.613888348570317,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967673006129094,
      "loss": 5.352,
      "mean_token_accuracy": 0.17558028995990754,
      "num_tokens": 13690815.0,
      "step": 7890
    },
    {
      "entropy": 5.6545045375823975,
      "epoch": 0.6142773779420346,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004967626176471536,
      "loss": 5.4793,
      "mean_token_accuracy": 0.16500618010759355,
      "num_tokens": 13699522.0,
      "step": 7895
    },
    {
      "entropy": 5.731170749664306,
      "epoch": 0.6146664073137522,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004967579313165028,
      "loss": 5.4809,
      "mean_token_accuracy": 0.16668136566877365,
      "num_tokens": 13708075.0,
      "step": 7900
    },
    {
      "entropy": 5.629131746292114,
      "epoch": 0.6150554366854698,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004967532416210284,
      "loss": 5.3669,
      "mean_token_accuracy": 0.17131840139627458,
      "num_tokens": 13716397.0,
      "step": 7905
    },
    {
      "entropy": 5.569436693191529,
      "epoch": 0.6154444660571873,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004967485485608016,
      "loss": 5.3618,
      "mean_token_accuracy": 0.17754382640123367,
      "num_tokens": 13724724.0,
      "step": 7910
    },
    {
      "entropy": 5.581193113327027,
      "epoch": 0.6158334954289049,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004967438521358934,
      "loss": 5.3233,
      "mean_token_accuracy": 0.183150814473629,
      "num_tokens": 13733389.0,
      "step": 7915
    },
    {
      "entropy": 5.581504392623901,
      "epoch": 0.6162225248006225,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967391523463754,
      "loss": 5.2795,
      "mean_token_accuracy": 0.17858623266220092,
      "num_tokens": 13742569.0,
      "step": 7920
    },
    {
      "entropy": 5.526614665985107,
      "epoch": 0.61661155417234,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004967344491923185,
      "loss": 5.272,
      "mean_token_accuracy": 0.17839372754096985,
      "num_tokens": 13750975.0,
      "step": 7925
    },
    {
      "entropy": 5.558490228652954,
      "epoch": 0.6170005835440576,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004967297426737943,
      "loss": 5.3336,
      "mean_token_accuracy": 0.18053995072841644,
      "num_tokens": 13758782.0,
      "step": 7930
    },
    {
      "entropy": 5.655747127532959,
      "epoch": 0.6173896129157751,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004967250327908742,
      "loss": 5.4636,
      "mean_token_accuracy": 0.16564081460237504,
      "num_tokens": 13767321.0,
      "step": 7935
    },
    {
      "entropy": 5.6967109680175785,
      "epoch": 0.6177786422874927,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004967203195436297,
      "loss": 5.4465,
      "mean_token_accuracy": 0.1746225267648697,
      "num_tokens": 13775992.0,
      "step": 7940
    },
    {
      "entropy": 5.58898811340332,
      "epoch": 0.6181676716592103,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004967156029321322,
      "loss": 5.2813,
      "mean_token_accuracy": 0.18081785142421722,
      "num_tokens": 13784358.0,
      "step": 7945
    },
    {
      "entropy": 5.495432949066162,
      "epoch": 0.6185567010309279,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004967108829564532,
      "loss": 5.2837,
      "mean_token_accuracy": 0.18145979791879654,
      "num_tokens": 13792834.0,
      "step": 7950
    },
    {
      "entropy": 5.627613973617554,
      "epoch": 0.6189457304026454,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004967061596166646,
      "loss": 5.3854,
      "mean_token_accuracy": 0.17125617116689681,
      "num_tokens": 13801414.0,
      "step": 7955
    },
    {
      "entropy": 5.628814077377319,
      "epoch": 0.6193347597743629,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004967014329128378,
      "loss": 5.4099,
      "mean_token_accuracy": 0.17231540232896805,
      "num_tokens": 13810459.0,
      "step": 7960
    },
    {
      "entropy": 5.581355810165405,
      "epoch": 0.6197237891460805,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004966967028450448,
      "loss": 5.2502,
      "mean_token_accuracy": 0.17979186326265334,
      "num_tokens": 13818460.0,
      "step": 7965
    },
    {
      "entropy": 5.602181911468506,
      "epoch": 0.6201128185177981,
      "grad_norm": 1.0,
      "learning_rate": 0.000496691969413357,
      "loss": 5.3718,
      "mean_token_accuracy": 0.17108058035373688,
      "num_tokens": 13827693.0,
      "step": 7970
    },
    {
      "entropy": 5.610408926010132,
      "epoch": 0.6205018478895157,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004966872326178466,
      "loss": 5.3649,
      "mean_token_accuracy": 0.1674000144004822,
      "num_tokens": 13836506.0,
      "step": 7975
    },
    {
      "entropy": 5.558001708984375,
      "epoch": 0.6208908772612333,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004966824924585851,
      "loss": 5.4006,
      "mean_token_accuracy": 0.17037640810012816,
      "num_tokens": 13844977.0,
      "step": 7980
    },
    {
      "entropy": 5.588616037368775,
      "epoch": 0.6212799066329507,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004966777489356448,
      "loss": 5.3664,
      "mean_token_accuracy": 0.17286057770252228,
      "num_tokens": 13854514.0,
      "step": 7985
    },
    {
      "entropy": 5.607264375686645,
      "epoch": 0.6216689360046683,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004966730020490975,
      "loss": 5.3567,
      "mean_token_accuracy": 0.17180452793836593,
      "num_tokens": 13863037.0,
      "step": 7990
    },
    {
      "entropy": 5.52875657081604,
      "epoch": 0.6220579653763859,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496668251799015,
      "loss": 5.3316,
      "mean_token_accuracy": 0.18510827124118806,
      "num_tokens": 13871720.0,
      "step": 7995
    },
    {
      "entropy": 5.5923436164855955,
      "epoch": 0.6224469947481035,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004966634981854699,
      "loss": 5.3831,
      "mean_token_accuracy": 0.1798848181962967,
      "num_tokens": 13880214.0,
      "step": 8000
    },
    {
      "entropy": 5.608328056335449,
      "epoch": 0.6228360241198211,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004966587412085339,
      "loss": 5.4798,
      "mean_token_accuracy": 0.16268185526132584,
      "num_tokens": 13888784.0,
      "step": 8005
    },
    {
      "entropy": 5.62626895904541,
      "epoch": 0.6232250534915386,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004966539808682793,
      "loss": 5.3488,
      "mean_token_accuracy": 0.17655370533466339,
      "num_tokens": 13897466.0,
      "step": 8010
    },
    {
      "entropy": 5.555896949768067,
      "epoch": 0.6236140828632561,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966492171647783,
      "loss": 5.3722,
      "mean_token_accuracy": 0.1709124967455864,
      "num_tokens": 13906556.0,
      "step": 8015
    },
    {
      "entropy": 5.5468177318573,
      "epoch": 0.6240031122349737,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004966444500981032,
      "loss": 5.316,
      "mean_token_accuracy": 0.17478737235069275,
      "num_tokens": 13915473.0,
      "step": 8020
    },
    {
      "entropy": 5.603696870803833,
      "epoch": 0.6243921416066913,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004966396796683265,
      "loss": 5.4901,
      "mean_token_accuracy": 0.16448316276073455,
      "num_tokens": 13924775.0,
      "step": 8025
    },
    {
      "entropy": 5.635816383361816,
      "epoch": 0.6247811709784089,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004966349058755204,
      "loss": 5.3678,
      "mean_token_accuracy": 0.16653134822845458,
      "num_tokens": 13933992.0,
      "step": 8030
    },
    {
      "entropy": 5.6757204055786135,
      "epoch": 0.6251702003501264,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004966301287197573,
      "loss": 5.4984,
      "mean_token_accuracy": 0.17714831978082657,
      "num_tokens": 13942877.0,
      "step": 8035
    },
    {
      "entropy": 5.570053243637085,
      "epoch": 0.625559229721844,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004966253482011098,
      "loss": 5.3822,
      "mean_token_accuracy": 0.17082124948501587,
      "num_tokens": 13952003.0,
      "step": 8040
    },
    {
      "entropy": 5.638340044021606,
      "epoch": 0.6259482590935616,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004966205643196503,
      "loss": 5.3317,
      "mean_token_accuracy": 0.1808565765619278,
      "num_tokens": 13959801.0,
      "step": 8045
    },
    {
      "entropy": 5.574903726577759,
      "epoch": 0.6263372884652791,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966157770754516,
      "loss": 5.3307,
      "mean_token_accuracy": 0.16821729838848115,
      "num_tokens": 13968535.0,
      "step": 8050
    },
    {
      "entropy": 5.579510879516602,
      "epoch": 0.6267263178369967,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496610986468586,
      "loss": 5.3052,
      "mean_token_accuracy": 0.1752540349960327,
      "num_tokens": 13977079.0,
      "step": 8055
    },
    {
      "entropy": 5.544235515594482,
      "epoch": 0.6271153472087143,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004966061924991267,
      "loss": 5.313,
      "mean_token_accuracy": 0.1814327582716942,
      "num_tokens": 13985118.0,
      "step": 8060
    },
    {
      "entropy": 5.574939870834351,
      "epoch": 0.6275043765804318,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000496601395167146,
      "loss": 5.2674,
      "mean_token_accuracy": 0.175836481153965,
      "num_tokens": 13994061.0,
      "step": 8065
    },
    {
      "entropy": 5.588698101043701,
      "epoch": 0.6278934059521494,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004965965944727168,
      "loss": 5.4225,
      "mean_token_accuracy": 0.16943728923797607,
      "num_tokens": 14002879.0,
      "step": 8070
    },
    {
      "entropy": 5.6054764747619625,
      "epoch": 0.628282435323867,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004965917904159121,
      "loss": 5.3928,
      "mean_token_accuracy": 0.17182569801807404,
      "num_tokens": 14011570.0,
      "step": 8075
    },
    {
      "entropy": 5.540693473815918,
      "epoch": 0.6286714646955845,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004965869829968046,
      "loss": 5.2401,
      "mean_token_accuracy": 0.17718469053506852,
      "num_tokens": 14019282.0,
      "step": 8080
    },
    {
      "entropy": 5.535258197784424,
      "epoch": 0.6290604940673021,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004965821722154674,
      "loss": 5.2634,
      "mean_token_accuracy": 0.18298577517271042,
      "num_tokens": 14027774.0,
      "step": 8085
    },
    {
      "entropy": 5.539744234085083,
      "epoch": 0.6294495234390196,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004965773580719734,
      "loss": 5.2434,
      "mean_token_accuracy": 0.17935188710689545,
      "num_tokens": 14036754.0,
      "step": 8090
    },
    {
      "entropy": 5.625921058654785,
      "epoch": 0.6298385528107372,
      "grad_norm": 1.0,
      "learning_rate": 0.0004965725405663957,
      "loss": 5.4197,
      "mean_token_accuracy": 0.17523077130317688,
      "num_tokens": 14045714.0,
      "step": 8095
    },
    {
      "entropy": 5.662188243865967,
      "epoch": 0.6302275821824548,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004965677196988073,
      "loss": 5.4539,
      "mean_token_accuracy": 0.16636739373207093,
      "num_tokens": 14054973.0,
      "step": 8100
    },
    {
      "entropy": 5.639545726776123,
      "epoch": 0.6306166115541724,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004965628954692815,
      "loss": 5.3587,
      "mean_token_accuracy": 0.1728656142950058,
      "num_tokens": 14064037.0,
      "step": 8105
    },
    {
      "entropy": 5.64419846534729,
      "epoch": 0.63100564092589,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004965580678778913,
      "loss": 5.3895,
      "mean_token_accuracy": 0.17385393232107163,
      "num_tokens": 14072269.0,
      "step": 8110
    },
    {
      "entropy": 5.581698369979859,
      "epoch": 0.6313946702976074,
      "grad_norm": 1.0,
      "learning_rate": 0.0004965532369247102,
      "loss": 5.34,
      "mean_token_accuracy": 0.17125858664512633,
      "num_tokens": 14080595.0,
      "step": 8115
    },
    {
      "entropy": 5.579161024093628,
      "epoch": 0.631783699669325,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004965484026098111,
      "loss": 5.3099,
      "mean_token_accuracy": 0.17416503578424453,
      "num_tokens": 14089159.0,
      "step": 8120
    },
    {
      "entropy": 5.63450517654419,
      "epoch": 0.6321727290410426,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004965435649332679,
      "loss": 5.4103,
      "mean_token_accuracy": 0.17195792496204376,
      "num_tokens": 14098192.0,
      "step": 8125
    },
    {
      "entropy": 5.590230131149292,
      "epoch": 0.6325617584127602,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004965387238951536,
      "loss": 5.2306,
      "mean_token_accuracy": 0.1833092600107193,
      "num_tokens": 14106552.0,
      "step": 8130
    },
    {
      "entropy": 5.554232931137085,
      "epoch": 0.6329507877844778,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004965338794955418,
      "loss": 5.2961,
      "mean_token_accuracy": 0.17610278874635696,
      "num_tokens": 14114783.0,
      "step": 8135
    },
    {
      "entropy": 5.547451019287109,
      "epoch": 0.6333398171561953,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004965290317345061,
      "loss": 5.3602,
      "mean_token_accuracy": 0.16638550609350206,
      "num_tokens": 14124182.0,
      "step": 8140
    },
    {
      "entropy": 5.656743097305298,
      "epoch": 0.6337288465279128,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004965241806121198,
      "loss": 5.3824,
      "mean_token_accuracy": 0.16936965882778168,
      "num_tokens": 14133037.0,
      "step": 8145
    },
    {
      "entropy": 5.612976551055908,
      "epoch": 0.6341178758996304,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004965193261284567,
      "loss": 5.364,
      "mean_token_accuracy": 0.17326362580060958,
      "num_tokens": 14141677.0,
      "step": 8150
    },
    {
      "entropy": 5.518466806411743,
      "epoch": 0.634506905271348,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965144682835904,
      "loss": 5.3133,
      "mean_token_accuracy": 0.17942651063203813,
      "num_tokens": 14149980.0,
      "step": 8155
    },
    {
      "entropy": 5.577079343795776,
      "epoch": 0.6348959346430656,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004965096070775946,
      "loss": 5.3674,
      "mean_token_accuracy": 0.18032984137535096,
      "num_tokens": 14158298.0,
      "step": 8160
    },
    {
      "entropy": 5.589863634109497,
      "epoch": 0.6352849640147831,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965047425105431,
      "loss": 5.3223,
      "mean_token_accuracy": 0.18497874438762665,
      "num_tokens": 14166321.0,
      "step": 8165
    },
    {
      "entropy": 5.48642930984497,
      "epoch": 0.6356739933865007,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004964998745825097,
      "loss": 5.3027,
      "mean_token_accuracy": 0.18269782811403273,
      "num_tokens": 14174619.0,
      "step": 8170
    },
    {
      "entropy": 5.537335062026978,
      "epoch": 0.6360630227582182,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964950032935682,
      "loss": 5.2955,
      "mean_token_accuracy": 0.17796382308006287,
      "num_tokens": 14183548.0,
      "step": 8175
    },
    {
      "entropy": 5.7116475105285645,
      "epoch": 0.6364520521299358,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004964901286437927,
      "loss": 5.494,
      "mean_token_accuracy": 0.17413605600595475,
      "num_tokens": 14192306.0,
      "step": 8180
    },
    {
      "entropy": 5.5339843273162845,
      "epoch": 0.6368410815016534,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004964852506332568,
      "loss": 5.1908,
      "mean_token_accuracy": 0.18867799490690232,
      "num_tokens": 14200280.0,
      "step": 8185
    },
    {
      "entropy": 5.540139293670654,
      "epoch": 0.6372301108733709,
      "grad_norm": 1.046875,
      "learning_rate": 0.000496480369262035,
      "loss": 5.4294,
      "mean_token_accuracy": 0.1720765322446823,
      "num_tokens": 14208484.0,
      "step": 8190
    },
    {
      "entropy": 5.579662942886353,
      "epoch": 0.6376191402450885,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000496475484530201,
      "loss": 5.4106,
      "mean_token_accuracy": 0.17133046686649323,
      "num_tokens": 14217368.0,
      "step": 8195
    },
    {
      "entropy": 5.689886903762817,
      "epoch": 0.6380081696168061,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004964705964378292,
      "loss": 5.438,
      "mean_token_accuracy": 0.166541887819767,
      "num_tokens": 14226663.0,
      "step": 8200
    },
    {
      "entropy": 5.6033202648162845,
      "epoch": 0.6383971989885237,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004964657049849934,
      "loss": 5.2964,
      "mean_token_accuracy": 0.17614648044109343,
      "num_tokens": 14235153.0,
      "step": 8205
    },
    {
      "entropy": 5.556395196914673,
      "epoch": 0.6387862283602412,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004964608101717681,
      "loss": 5.4151,
      "mean_token_accuracy": 0.1731547698378563,
      "num_tokens": 14244245.0,
      "step": 8210
    },
    {
      "entropy": 5.605984258651733,
      "epoch": 0.6391752577319587,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004964559119982275,
      "loss": 5.3254,
      "mean_token_accuracy": 0.17295570224523543,
      "num_tokens": 14252666.0,
      "step": 8215
    },
    {
      "entropy": 5.631249046325683,
      "epoch": 0.6395642871036763,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496451010464446,
      "loss": 5.4016,
      "mean_token_accuracy": 0.17393245548009872,
      "num_tokens": 14262232.0,
      "step": 8220
    },
    {
      "entropy": 5.62627592086792,
      "epoch": 0.6399533164753939,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004964461055704978,
      "loss": 5.3398,
      "mean_token_accuracy": 0.17456934452056885,
      "num_tokens": 14270496.0,
      "step": 8225
    },
    {
      "entropy": 5.5464574813842775,
      "epoch": 0.6403423458471115,
      "grad_norm": 1.0,
      "learning_rate": 0.0004964411973164574,
      "loss": 5.2623,
      "mean_token_accuracy": 0.18351061046123504,
      "num_tokens": 14278597.0,
      "step": 8230
    },
    {
      "entropy": 5.544543647766114,
      "epoch": 0.6407313752188291,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004964362857023993,
      "loss": 5.3248,
      "mean_token_accuracy": 0.17445123195648193,
      "num_tokens": 14287444.0,
      "step": 8235
    },
    {
      "entropy": 5.535120153427124,
      "epoch": 0.6411204045905465,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496431370728398,
      "loss": 5.2631,
      "mean_token_accuracy": 0.17866938263177873,
      "num_tokens": 14296041.0,
      "step": 8240
    },
    {
      "entropy": 5.575885438919068,
      "epoch": 0.6415094339622641,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004964264523945281,
      "loss": 5.2654,
      "mean_token_accuracy": 0.17801269739866257,
      "num_tokens": 14304700.0,
      "step": 8245
    },
    {
      "entropy": 5.615669870376587,
      "epoch": 0.6418984633339817,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004964215307008643,
      "loss": 5.3976,
      "mean_token_accuracy": 0.17064619660377503,
      "num_tokens": 14312873.0,
      "step": 8250
    },
    {
      "entropy": 5.623337507247925,
      "epoch": 0.6422874927056993,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004964166056474811,
      "loss": 5.3849,
      "mean_token_accuracy": 0.1692052811384201,
      "num_tokens": 14321697.0,
      "step": 8255
    },
    {
      "entropy": 5.58499116897583,
      "epoch": 0.6426765220774169,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004964116772344534,
      "loss": 5.3774,
      "mean_token_accuracy": 0.16862674802541733,
      "num_tokens": 14330074.0,
      "step": 8260
    },
    {
      "entropy": 5.659800958633423,
      "epoch": 0.6430655514491345,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004964067454618559,
      "loss": 5.4435,
      "mean_token_accuracy": 0.16576170325279235,
      "num_tokens": 14339485.0,
      "step": 8265
    },
    {
      "entropy": 5.680325651168824,
      "epoch": 0.6434545808208519,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004964018103297634,
      "loss": 5.4143,
      "mean_token_accuracy": 0.1698738992214203,
      "num_tokens": 14348182.0,
      "step": 8270
    },
    {
      "entropy": 5.692117309570312,
      "epoch": 0.6438436101925695,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004963968718382509,
      "loss": 5.4582,
      "mean_token_accuracy": 0.1681407019495964,
      "num_tokens": 14357202.0,
      "step": 8275
    },
    {
      "entropy": 5.604986333847046,
      "epoch": 0.6442326395642871,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004963919299873931,
      "loss": 5.3812,
      "mean_token_accuracy": 0.1657526060938835,
      "num_tokens": 14366001.0,
      "step": 8280
    },
    {
      "entropy": 5.664509963989258,
      "epoch": 0.6446216689360047,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004963869847772653,
      "loss": 5.4065,
      "mean_token_accuracy": 0.16915652602910997,
      "num_tokens": 14374894.0,
      "step": 8285
    },
    {
      "entropy": 5.562790155410767,
      "epoch": 0.6450106983077223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004963820362079423,
      "loss": 5.3809,
      "mean_token_accuracy": 0.17515988796949386,
      "num_tokens": 14383396.0,
      "step": 8290
    },
    {
      "entropy": 5.58872127532959,
      "epoch": 0.6453997276794398,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004963770842794994,
      "loss": 5.2961,
      "mean_token_accuracy": 0.18302446156740187,
      "num_tokens": 14392514.0,
      "step": 8295
    },
    {
      "entropy": 5.607506370544433,
      "epoch": 0.6457887570511573,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004963721289920115,
      "loss": 5.3395,
      "mean_token_accuracy": 0.1733679324388504,
      "num_tokens": 14401916.0,
      "step": 8300
    },
    {
      "entropy": 5.592385196685791,
      "epoch": 0.6461777864228749,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004963671703455538,
      "loss": 5.406,
      "mean_token_accuracy": 0.17042556703090667,
      "num_tokens": 14410476.0,
      "step": 8305
    },
    {
      "entropy": 5.610119152069092,
      "epoch": 0.6465668157945925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004963622083402018,
      "loss": 5.3981,
      "mean_token_accuracy": 0.17096285223960878,
      "num_tokens": 14418437.0,
      "step": 8310
    },
    {
      "entropy": 5.557107162475586,
      "epoch": 0.6469558451663101,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004963572429760305,
      "loss": 5.3384,
      "mean_token_accuracy": 0.17626968324184417,
      "num_tokens": 14426978.0,
      "step": 8315
    },
    {
      "entropy": 5.637966680526733,
      "epoch": 0.6473448745380276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004963522742531155,
      "loss": 5.3646,
      "mean_token_accuracy": 0.17541678845882416,
      "num_tokens": 14435043.0,
      "step": 8320
    },
    {
      "entropy": 5.583234119415283,
      "epoch": 0.6477339039097452,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000496347302171532,
      "loss": 5.2759,
      "mean_token_accuracy": 0.18421368151903153,
      "num_tokens": 14443397.0,
      "step": 8325
    },
    {
      "entropy": 5.574254941940308,
      "epoch": 0.6481229332814628,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004963423267313554,
      "loss": 5.3712,
      "mean_token_accuracy": 0.17410993427038193,
      "num_tokens": 14452536.0,
      "step": 8330
    },
    {
      "entropy": 5.595439624786377,
      "epoch": 0.6485119626531803,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004963373479326614,
      "loss": 5.4407,
      "mean_token_accuracy": 0.16464628130197526,
      "num_tokens": 14461901.0,
      "step": 8335
    },
    {
      "entropy": 5.591050386428833,
      "epoch": 0.6489009920248979,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004963323657755255,
      "loss": 5.3376,
      "mean_token_accuracy": 0.17729219943284988,
      "num_tokens": 14470547.0,
      "step": 8340
    },
    {
      "entropy": 5.528455781936645,
      "epoch": 0.6492900213966154,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963273802600231,
      "loss": 5.2297,
      "mean_token_accuracy": 0.17649297267198563,
      "num_tokens": 14478439.0,
      "step": 8345
    },
    {
      "entropy": 5.617388200759888,
      "epoch": 0.649679050768333,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049632239138623,
      "loss": 5.2801,
      "mean_token_accuracy": 0.18349619209766388,
      "num_tokens": 14486621.0,
      "step": 8350
    },
    {
      "entropy": 5.562037467956543,
      "epoch": 0.6500680801400506,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004963173991542219,
      "loss": 5.2874,
      "mean_token_accuracy": 0.17734003514051438,
      "num_tokens": 14495313.0,
      "step": 8355
    },
    {
      "entropy": 5.617541790008545,
      "epoch": 0.6504571095117682,
      "grad_norm": 1.0,
      "learning_rate": 0.0004963124035640746,
      "loss": 5.3475,
      "mean_token_accuracy": 0.1766296923160553,
      "num_tokens": 14503647.0,
      "step": 8360
    },
    {
      "entropy": 5.613102531433105,
      "epoch": 0.6508461388834857,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004963074046158637,
      "loss": 5.2805,
      "mean_token_accuracy": 0.1789421707391739,
      "num_tokens": 14512350.0,
      "step": 8365
    },
    {
      "entropy": 5.6533385753631595,
      "epoch": 0.6512351682552032,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004963024023096652,
      "loss": 5.4291,
      "mean_token_accuracy": 0.1655445083975792,
      "num_tokens": 14521801.0,
      "step": 8370
    },
    {
      "entropy": 5.554292678833008,
      "epoch": 0.6516241976269208,
      "grad_norm": 1.0,
      "learning_rate": 0.0004962973966455551,
      "loss": 5.259,
      "mean_token_accuracy": 0.1753644213080406,
      "num_tokens": 14530045.0,
      "step": 8375
    },
    {
      "entropy": 5.60089373588562,
      "epoch": 0.6520132269986384,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004962923876236092,
      "loss": 5.3808,
      "mean_token_accuracy": 0.1721917897462845,
      "num_tokens": 14538944.0,
      "step": 8380
    },
    {
      "entropy": 5.680092430114746,
      "epoch": 0.652402256370356,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962873752439035,
      "loss": 5.3918,
      "mean_token_accuracy": 0.17222664952278138,
      "num_tokens": 14548180.0,
      "step": 8385
    },
    {
      "entropy": 5.645467519760132,
      "epoch": 0.6527912857420736,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004962823595065141,
      "loss": 5.4098,
      "mean_token_accuracy": 0.17034562826156616,
      "num_tokens": 14556907.0,
      "step": 8390
    },
    {
      "entropy": 5.578256464004516,
      "epoch": 0.653180315113791,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004962773404115172,
      "loss": 5.3816,
      "mean_token_accuracy": 0.17010515034198762,
      "num_tokens": 14565317.0,
      "step": 8395
    },
    {
      "entropy": 5.6273193359375,
      "epoch": 0.6535693444855086,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004962723179589886,
      "loss": 5.4375,
      "mean_token_accuracy": 0.17320182919502258,
      "num_tokens": 14574485.0,
      "step": 8400
    },
    {
      "entropy": 5.6831926822662355,
      "epoch": 0.6539583738572262,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496267292149005,
      "loss": 5.3366,
      "mean_token_accuracy": 0.17173146307468415,
      "num_tokens": 14584002.0,
      "step": 8405
    },
    {
      "entropy": 5.620897436141968,
      "epoch": 0.6543474032289438,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004962622629816423,
      "loss": 5.2866,
      "mean_token_accuracy": 0.17372749596834183,
      "num_tokens": 14592033.0,
      "step": 8410
    },
    {
      "entropy": 5.575413608551026,
      "epoch": 0.6547364326006614,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000496257230456977,
      "loss": 5.4229,
      "mean_token_accuracy": 0.16647793501615524,
      "num_tokens": 14600084.0,
      "step": 8415
    },
    {
      "entropy": 5.549050331115723,
      "epoch": 0.6551254619723789,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004962521945750855,
      "loss": 5.3345,
      "mean_token_accuracy": 0.18473218381404877,
      "num_tokens": 14608380.0,
      "step": 8420
    },
    {
      "entropy": 5.601489877700805,
      "epoch": 0.6555144913440965,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496247155336044,
      "loss": 5.3939,
      "mean_token_accuracy": 0.17354901134967804,
      "num_tokens": 14616864.0,
      "step": 8425
    },
    {
      "entropy": 5.623040390014649,
      "epoch": 0.655903520715814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004962421127399291,
      "loss": 5.4296,
      "mean_token_accuracy": 0.1710478574037552,
      "num_tokens": 14627054.0,
      "step": 8430
    },
    {
      "entropy": 5.670633983612061,
      "epoch": 0.6562925500875316,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962370667868172,
      "loss": 5.3229,
      "mean_token_accuracy": 0.17703424990177155,
      "num_tokens": 14634985.0,
      "step": 8435
    },
    {
      "entropy": 5.643927383422851,
      "epoch": 0.6566815794592492,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004962320174767851,
      "loss": 5.3916,
      "mean_token_accuracy": 0.18009753823280333,
      "num_tokens": 14644135.0,
      "step": 8440
    },
    {
      "entropy": 5.552023220062256,
      "epoch": 0.6570706088309667,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000496226964809909,
      "loss": 5.3505,
      "mean_token_accuracy": 0.17871672064065933,
      "num_tokens": 14652997.0,
      "step": 8445
    },
    {
      "entropy": 5.575016212463379,
      "epoch": 0.6574596382026843,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000496221908786266,
      "loss": 5.4318,
      "mean_token_accuracy": 0.1723785251379013,
      "num_tokens": 14661961.0,
      "step": 8450
    },
    {
      "entropy": 5.677073335647583,
      "epoch": 0.6578486675744019,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004962168494059327,
      "loss": 5.5334,
      "mean_token_accuracy": 0.1619396537542343,
      "num_tokens": 14671407.0,
      "step": 8455
    },
    {
      "entropy": 5.653186559677124,
      "epoch": 0.6582376969461194,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004962117866689857,
      "loss": 5.3801,
      "mean_token_accuracy": 0.17210789620876313,
      "num_tokens": 14680015.0,
      "step": 8460
    },
    {
      "entropy": 5.530022239685058,
      "epoch": 0.658626726317837,
      "grad_norm": 1.078125,
      "learning_rate": 0.000496206720575502,
      "loss": 5.2363,
      "mean_token_accuracy": 0.17737187445163727,
      "num_tokens": 14688444.0,
      "step": 8465
    },
    {
      "entropy": 5.562881183624268,
      "epoch": 0.6590157556895546,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962016511255584,
      "loss": 5.3704,
      "mean_token_accuracy": 0.1729799672961235,
      "num_tokens": 14697246.0,
      "step": 8470
    },
    {
      "entropy": 5.561779022216797,
      "epoch": 0.6594047850612721,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004961965783192318,
      "loss": 5.2517,
      "mean_token_accuracy": 0.18331620693206788,
      "num_tokens": 14706066.0,
      "step": 8475
    },
    {
      "entropy": 5.601294326782226,
      "epoch": 0.6597938144329897,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004961915021565992,
      "loss": 5.3758,
      "mean_token_accuracy": 0.16975661367177963,
      "num_tokens": 14714327.0,
      "step": 8480
    },
    {
      "entropy": 5.552827501296997,
      "epoch": 0.6601828438047073,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004961864226377377,
      "loss": 5.2382,
      "mean_token_accuracy": 0.17383477240800857,
      "num_tokens": 14723053.0,
      "step": 8485
    },
    {
      "entropy": 5.542729759216309,
      "epoch": 0.6605718731764249,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004961813397627241,
      "loss": 5.2263,
      "mean_token_accuracy": 0.1823290318250656,
      "num_tokens": 14731825.0,
      "step": 8490
    },
    {
      "entropy": 5.584241247177124,
      "epoch": 0.6609609025481424,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004961762535316358,
      "loss": 5.4327,
      "mean_token_accuracy": 0.16857006251811982,
      "num_tokens": 14740751.0,
      "step": 8495
    },
    {
      "entropy": 5.608772897720337,
      "epoch": 0.6613499319198599,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004961711639445499,
      "loss": 5.3014,
      "mean_token_accuracy": 0.1776784747838974,
      "num_tokens": 14748994.0,
      "step": 8500
    },
    {
      "entropy": 5.606518363952636,
      "epoch": 0.6617389612915775,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004961660710015435,
      "loss": 5.2935,
      "mean_token_accuracy": 0.17489899098873138,
      "num_tokens": 14757388.0,
      "step": 8505
    },
    {
      "entropy": 5.676743650436402,
      "epoch": 0.6621279906632951,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004961609747026943,
      "loss": 5.4395,
      "mean_token_accuracy": 0.16976004093885422,
      "num_tokens": 14766654.0,
      "step": 8510
    },
    {
      "entropy": 5.580006885528564,
      "epoch": 0.6625170200350127,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004961558750480791,
      "loss": 5.3578,
      "mean_token_accuracy": 0.17017442137002944,
      "num_tokens": 14775561.0,
      "step": 8515
    },
    {
      "entropy": 5.593653726577759,
      "epoch": 0.6629060494067303,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004961507720377756,
      "loss": 5.4202,
      "mean_token_accuracy": 0.16423940807580947,
      "num_tokens": 14783596.0,
      "step": 8520
    },
    {
      "entropy": 5.634108591079712,
      "epoch": 0.6632950787784477,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004961456656718611,
      "loss": 5.4294,
      "mean_token_accuracy": 0.16964395791292192,
      "num_tokens": 14791932.0,
      "step": 8525
    },
    {
      "entropy": 5.652415704727173,
      "epoch": 0.6636841081501653,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000496140555950413,
      "loss": 5.4536,
      "mean_token_accuracy": 0.16912026405334474,
      "num_tokens": 14801261.0,
      "step": 8530
    },
    {
      "entropy": 5.6047392845153805,
      "epoch": 0.6640731375218829,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004961354428735091,
      "loss": 5.3357,
      "mean_token_accuracy": 0.17305805534124374,
      "num_tokens": 14809788.0,
      "step": 8535
    },
    {
      "entropy": 5.545555257797242,
      "epoch": 0.6644621668936005,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004961303264412267,
      "loss": 5.3274,
      "mean_token_accuracy": 0.17021722495555877,
      "num_tokens": 14818324.0,
      "step": 8540
    },
    {
      "entropy": 5.526040124893188,
      "epoch": 0.6648511962653181,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004961252066536437,
      "loss": 5.2805,
      "mean_token_accuracy": 0.17774665653705596,
      "num_tokens": 14827020.0,
      "step": 8545
    },
    {
      "entropy": 5.564234209060669,
      "epoch": 0.6652402256370356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961200835108375,
      "loss": 5.3871,
      "mean_token_accuracy": 0.1685611218214035,
      "num_tokens": 14835522.0,
      "step": 8550
    },
    {
      "entropy": 5.63460865020752,
      "epoch": 0.6656292550087531,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496114957012886,
      "loss": 5.3465,
      "mean_token_accuracy": 0.17625492215156555,
      "num_tokens": 14844676.0,
      "step": 8555
    },
    {
      "entropy": 5.653724718093872,
      "epoch": 0.6660182843804707,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961098271598671,
      "loss": 5.3852,
      "mean_token_accuracy": 0.17274842262268067,
      "num_tokens": 14853996.0,
      "step": 8560
    },
    {
      "entropy": 5.612109565734864,
      "epoch": 0.6664073137521883,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961046939518585,
      "loss": 5.3729,
      "mean_token_accuracy": 0.17253222763538362,
      "num_tokens": 14862981.0,
      "step": 8565
    },
    {
      "entropy": 5.613665246963501,
      "epoch": 0.6667963431239059,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004960995573889379,
      "loss": 5.4015,
      "mean_token_accuracy": 0.16996358931064606,
      "num_tokens": 14872378.0,
      "step": 8570
    },
    {
      "entropy": 5.558696269989014,
      "epoch": 0.6671853724956234,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004960944174711836,
      "loss": 5.3487,
      "mean_token_accuracy": 0.1734008237719536,
      "num_tokens": 14881662.0,
      "step": 8575
    },
    {
      "entropy": 5.628997802734375,
      "epoch": 0.667574401867341,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004960892741986734,
      "loss": 5.2375,
      "mean_token_accuracy": 0.17901071906089783,
      "num_tokens": 14890211.0,
      "step": 8580
    },
    {
      "entropy": 5.663716506958008,
      "epoch": 0.6679634312390585,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004960841275714853,
      "loss": 5.4768,
      "mean_token_accuracy": 0.165939624607563,
      "num_tokens": 14899865.0,
      "step": 8585
    },
    {
      "entropy": 5.59187388420105,
      "epoch": 0.6683524606107761,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004960789775896975,
      "loss": 5.375,
      "mean_token_accuracy": 0.1709680363535881,
      "num_tokens": 14908334.0,
      "step": 8590
    },
    {
      "entropy": 5.546440458297729,
      "epoch": 0.6687414899824937,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004960738242533881,
      "loss": 5.3263,
      "mean_token_accuracy": 0.17018923163414001,
      "num_tokens": 14916477.0,
      "step": 8595
    },
    {
      "entropy": 5.534609317779541,
      "epoch": 0.6691305193542112,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004960686675626353,
      "loss": 5.2612,
      "mean_token_accuracy": 0.18023026138544082,
      "num_tokens": 14925223.0,
      "step": 8600
    },
    {
      "entropy": 5.615374946594239,
      "epoch": 0.6695195487259288,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004960635075175173,
      "loss": 5.4224,
      "mean_token_accuracy": 0.1766976684331894,
      "num_tokens": 14934427.0,
      "step": 8605
    },
    {
      "entropy": 5.660296106338501,
      "epoch": 0.6699085780976464,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004960583441181124,
      "loss": 5.3294,
      "mean_token_accuracy": 0.17061750888824462,
      "num_tokens": 14943175.0,
      "step": 8610
    },
    {
      "entropy": 5.641629028320312,
      "epoch": 0.670297607469364,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004960531773644992,
      "loss": 5.3658,
      "mean_token_accuracy": 0.17371981889009475,
      "num_tokens": 14951933.0,
      "step": 8615
    },
    {
      "entropy": 5.539992618560791,
      "epoch": 0.6706866368410815,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960480072567557,
      "loss": 5.3346,
      "mean_token_accuracy": 0.17478349655866623,
      "num_tokens": 14960735.0,
      "step": 8620
    },
    {
      "entropy": 5.596373748779297,
      "epoch": 0.671075666212799,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004960428337949604,
      "loss": 5.4398,
      "mean_token_accuracy": 0.16128688901662827,
      "num_tokens": 14970326.0,
      "step": 8625
    },
    {
      "entropy": 5.629201078414917,
      "epoch": 0.6714646955845166,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004960376569791921,
      "loss": 5.2938,
      "mean_token_accuracy": 0.17815542072057725,
      "num_tokens": 14979822.0,
      "step": 8630
    },
    {
      "entropy": 5.6754405975341795,
      "epoch": 0.6718537249562342,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960324768095291,
      "loss": 5.4085,
      "mean_token_accuracy": 0.1730176329612732,
      "num_tokens": 14989350.0,
      "step": 8635
    },
    {
      "entropy": 5.6606629371643065,
      "epoch": 0.6722427543279518,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004960272932860502,
      "loss": 5.4147,
      "mean_token_accuracy": 0.16840969920158386,
      "num_tokens": 14998763.0,
      "step": 8640
    },
    {
      "entropy": 5.59739351272583,
      "epoch": 0.6726317836996694,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004960221064088338,
      "loss": 5.2809,
      "mean_token_accuracy": 0.17504533380270004,
      "num_tokens": 15006497.0,
      "step": 8645
    },
    {
      "entropy": 5.513886976242065,
      "epoch": 0.6730208130713868,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004960169161779588,
      "loss": 5.2495,
      "mean_token_accuracy": 0.17931035459041594,
      "num_tokens": 15014397.0,
      "step": 8650
    },
    {
      "entropy": 5.548941230773925,
      "epoch": 0.6734098424431044,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004960117225935038,
      "loss": 5.2764,
      "mean_token_accuracy": 0.18180057406425476,
      "num_tokens": 15022726.0,
      "step": 8655
    },
    {
      "entropy": 5.680811405181885,
      "epoch": 0.673798871814822,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004960065256555477,
      "loss": 5.4944,
      "mean_token_accuracy": 0.1663055121898651,
      "num_tokens": 15032317.0,
      "step": 8660
    },
    {
      "entropy": 5.596422576904297,
      "epoch": 0.6741879011865396,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004960013253641695,
      "loss": 5.3526,
      "mean_token_accuracy": 0.17427897900342942,
      "num_tokens": 15041091.0,
      "step": 8665
    },
    {
      "entropy": 5.630930995941162,
      "epoch": 0.6745769305582572,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004959961217194477,
      "loss": 5.495,
      "mean_token_accuracy": 0.16269925981760025,
      "num_tokens": 15049797.0,
      "step": 8670
    },
    {
      "entropy": 5.549777984619141,
      "epoch": 0.6749659599299748,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004959909147214615,
      "loss": 5.1909,
      "mean_token_accuracy": 0.18470783829689025,
      "num_tokens": 15057736.0,
      "step": 8675
    },
    {
      "entropy": 5.5708211898803714,
      "epoch": 0.6753549893016922,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004959857043702902,
      "loss": 5.362,
      "mean_token_accuracy": 0.1793429061770439,
      "num_tokens": 15066072.0,
      "step": 8680
    },
    {
      "entropy": 5.529650020599365,
      "epoch": 0.6757440186734098,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004959804906660124,
      "loss": 5.2662,
      "mean_token_accuracy": 0.1785495787858963,
      "num_tokens": 15073918.0,
      "step": 8685
    },
    {
      "entropy": 5.613301610946655,
      "epoch": 0.6761330480451274,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004959752736087073,
      "loss": 5.3614,
      "mean_token_accuracy": 0.1788905903697014,
      "num_tokens": 15082937.0,
      "step": 8690
    },
    {
      "entropy": 5.559736108779907,
      "epoch": 0.676522077416845,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004959700531984543,
      "loss": 5.3177,
      "mean_token_accuracy": 0.18061764240264894,
      "num_tokens": 15091218.0,
      "step": 8695
    },
    {
      "entropy": 5.553287220001221,
      "epoch": 0.6769111067885626,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004959648294353324,
      "loss": 5.2474,
      "mean_token_accuracy": 0.18204528391361235,
      "num_tokens": 15099640.0,
      "step": 8700
    },
    {
      "entropy": 5.599412107467652,
      "epoch": 0.6773001361602801,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004959596023194208,
      "loss": 5.3279,
      "mean_token_accuracy": 0.17507115006446838,
      "num_tokens": 15108710.0,
      "step": 8705
    },
    {
      "entropy": 5.5698775291442875,
      "epoch": 0.6776891655319977,
      "grad_norm": 1.015625,
      "learning_rate": 0.000495954371850799,
      "loss": 5.2904,
      "mean_token_accuracy": 0.17398254275321962,
      "num_tokens": 15117074.0,
      "step": 8710
    },
    {
      "entropy": 5.6180814743042,
      "epoch": 0.6780781949037152,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004959491380295463,
      "loss": 5.2742,
      "mean_token_accuracy": 0.17939819693565368,
      "num_tokens": 15124714.0,
      "step": 8715
    },
    {
      "entropy": 5.544650268554688,
      "epoch": 0.6784672242754328,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004959439008557422,
      "loss": 5.3062,
      "mean_token_accuracy": 0.1737214908003807,
      "num_tokens": 15134177.0,
      "step": 8720
    },
    {
      "entropy": 5.614916658401489,
      "epoch": 0.6788562536471504,
      "grad_norm": 1.046875,
      "learning_rate": 0.000495938660329466,
      "loss": 5.3171,
      "mean_token_accuracy": 0.17673608660697937,
      "num_tokens": 15142659.0,
      "step": 8725
    },
    {
      "entropy": 5.581689453125,
      "epoch": 0.6792452830188679,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004959334164507973,
      "loss": 5.4753,
      "mean_token_accuracy": 0.1690713033080101,
      "num_tokens": 15151423.0,
      "step": 8730
    },
    {
      "entropy": 5.582829475402832,
      "epoch": 0.6796343123905855,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004959281692198155,
      "loss": 5.2525,
      "mean_token_accuracy": 0.17850179523229598,
      "num_tokens": 15159986.0,
      "step": 8735
    },
    {
      "entropy": 5.616036605834961,
      "epoch": 0.6800233417623031,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004959229186366007,
      "loss": 5.3328,
      "mean_token_accuracy": 0.17353588789701463,
      "num_tokens": 15168183.0,
      "step": 8740
    },
    {
      "entropy": 5.525445938110352,
      "epoch": 0.6804123711340206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000495917664701232,
      "loss": 5.243,
      "mean_token_accuracy": 0.18078750669956206,
      "num_tokens": 15176476.0,
      "step": 8745
    },
    {
      "entropy": 5.673672342300415,
      "epoch": 0.6808014005057382,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004959124074137894,
      "loss": 5.5157,
      "mean_token_accuracy": 0.16620651483535767,
      "num_tokens": 15186873.0,
      "step": 8750
    },
    {
      "entropy": 5.639481735229492,
      "epoch": 0.6811904298774557,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004959071467743526,
      "loss": 5.309,
      "mean_token_accuracy": 0.16986880004405974,
      "num_tokens": 15194965.0,
      "step": 8755
    },
    {
      "entropy": 5.551694726943969,
      "epoch": 0.6815794592491733,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004959018827830016,
      "loss": 5.1768,
      "mean_token_accuracy": 0.18066548109054564,
      "num_tokens": 15204028.0,
      "step": 8760
    },
    {
      "entropy": 5.494821929931641,
      "epoch": 0.6819684886208909,
      "grad_norm": 1.0625,
      "learning_rate": 0.000495896615439816,
      "loss": 5.2047,
      "mean_token_accuracy": 0.1873243436217308,
      "num_tokens": 15212011.0,
      "step": 8765
    },
    {
      "entropy": 5.519406366348266,
      "epoch": 0.6823575179926085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004958913447448759,
      "loss": 5.288,
      "mean_token_accuracy": 0.17712078392505645,
      "num_tokens": 15220088.0,
      "step": 8770
    },
    {
      "entropy": 5.612569952011109,
      "epoch": 0.682746547364326,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958860706982613,
      "loss": 5.2461,
      "mean_token_accuracy": 0.1817832812666893,
      "num_tokens": 15228676.0,
      "step": 8775
    },
    {
      "entropy": 5.540105104446411,
      "epoch": 0.6831355767360435,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000495880793300052,
      "loss": 5.2864,
      "mean_token_accuracy": 0.1815814346075058,
      "num_tokens": 15237392.0,
      "step": 8780
    },
    {
      "entropy": 5.54796781539917,
      "epoch": 0.6835246061077611,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004958755125503284,
      "loss": 5.4185,
      "mean_token_accuracy": 0.1767105430364609,
      "num_tokens": 15246738.0,
      "step": 8785
    },
    {
      "entropy": 5.647914218902588,
      "epoch": 0.6839136354794787,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958702284491703,
      "loss": 5.3897,
      "mean_token_accuracy": 0.1756160721182823,
      "num_tokens": 15255154.0,
      "step": 8790
    },
    {
      "entropy": 5.57085337638855,
      "epoch": 0.6843026648511963,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004958649409966581,
      "loss": 5.2768,
      "mean_token_accuracy": 0.18078256845474244,
      "num_tokens": 15264178.0,
      "step": 8795
    },
    {
      "entropy": 5.605639028549194,
      "epoch": 0.6846916942229139,
      "grad_norm": 1.046875,
      "learning_rate": 0.000495859650192872,
      "loss": 5.3878,
      "mean_token_accuracy": 0.17433761656284333,
      "num_tokens": 15272591.0,
      "step": 8800
    },
    {
      "entropy": 5.59878191947937,
      "epoch": 0.6850807235946313,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004958543560378922,
      "loss": 5.2939,
      "mean_token_accuracy": 0.16908714026212693,
      "num_tokens": 15281280.0,
      "step": 8805
    },
    {
      "entropy": 5.593587160110474,
      "epoch": 0.6854697529663489,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004958490585317991,
      "loss": 5.2382,
      "mean_token_accuracy": 0.1843152090907097,
      "num_tokens": 15290639.0,
      "step": 8810
    },
    {
      "entropy": 5.52235860824585,
      "epoch": 0.6858587823380665,
      "grad_norm": 1.0625,
      "learning_rate": 0.000495843757674673,
      "loss": 5.2296,
      "mean_token_accuracy": 0.1753144696354866,
      "num_tokens": 15298877.0,
      "step": 8815
    },
    {
      "entropy": 5.596991682052613,
      "epoch": 0.6862478117097841,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004958384534665945,
      "loss": 5.3783,
      "mean_token_accuracy": 0.17458462715148926,
      "num_tokens": 15307710.0,
      "step": 8820
    },
    {
      "entropy": 5.633730268478393,
      "epoch": 0.6866368410815017,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004958331459076438,
      "loss": 5.3891,
      "mean_token_accuracy": 0.17663782089948654,
      "num_tokens": 15315749.0,
      "step": 8825
    },
    {
      "entropy": 5.661491775512696,
      "epoch": 0.6870258704532192,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004958278349979018,
      "loss": 5.3432,
      "mean_token_accuracy": 0.16726212650537492,
      "num_tokens": 15324161.0,
      "step": 8830
    },
    {
      "entropy": 5.476969242095947,
      "epoch": 0.6874148998249368,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958225207374488,
      "loss": 5.2393,
      "mean_token_accuracy": 0.17856181263923646,
      "num_tokens": 15333414.0,
      "step": 8835
    },
    {
      "entropy": 5.547606134414673,
      "epoch": 0.6878039291966543,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004958172031263655,
      "loss": 5.3028,
      "mean_token_accuracy": 0.17507703602313995,
      "num_tokens": 15342369.0,
      "step": 8840
    },
    {
      "entropy": 5.619305086135864,
      "epoch": 0.6881929585683719,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004958118821647326,
      "loss": 5.3836,
      "mean_token_accuracy": 0.17099674493074418,
      "num_tokens": 15351109.0,
      "step": 8845
    },
    {
      "entropy": 5.574743413925171,
      "epoch": 0.6885819879400895,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004958065578526308,
      "loss": 5.2711,
      "mean_token_accuracy": 0.1795060008764267,
      "num_tokens": 15359837.0,
      "step": 8850
    },
    {
      "entropy": 5.53717565536499,
      "epoch": 0.688971017311807,
      "grad_norm": 0.9375,
      "learning_rate": 0.000495801230190141,
      "loss": 5.3116,
      "mean_token_accuracy": 0.18208779990673066,
      "num_tokens": 15368776.0,
      "step": 8855
    },
    {
      "entropy": 5.568300533294678,
      "epoch": 0.6893600466835246,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004957958991773441,
      "loss": 5.3055,
      "mean_token_accuracy": 0.17928243428468704,
      "num_tokens": 15378127.0,
      "step": 8860
    },
    {
      "entropy": 5.579755067825317,
      "epoch": 0.6897490760552422,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004957905648143206,
      "loss": 5.2787,
      "mean_token_accuracy": 0.17940096259117128,
      "num_tokens": 15386580.0,
      "step": 8865
    },
    {
      "entropy": 5.584767055511475,
      "epoch": 0.6901381054269597,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004957852271011519,
      "loss": 5.405,
      "mean_token_accuracy": 0.17330296337604523,
      "num_tokens": 15395973.0,
      "step": 8870
    },
    {
      "entropy": 5.606221294403076,
      "epoch": 0.6905271347986773,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004957798860379187,
      "loss": 5.3118,
      "mean_token_accuracy": 0.1782598003745079,
      "num_tokens": 15404738.0,
      "step": 8875
    },
    {
      "entropy": 5.622283554077148,
      "epoch": 0.6909161641703949,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004957745416247022,
      "loss": 5.4441,
      "mean_token_accuracy": 0.1751920148730278,
      "num_tokens": 15413530.0,
      "step": 8880
    },
    {
      "entropy": 5.633462619781494,
      "epoch": 0.6913051935421124,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004957691938615833,
      "loss": 5.4183,
      "mean_token_accuracy": 0.1715339332818985,
      "num_tokens": 15423268.0,
      "step": 8885
    },
    {
      "entropy": 5.574162769317627,
      "epoch": 0.69169422291383,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004957638427486434,
      "loss": 5.2495,
      "mean_token_accuracy": 0.18264658004045486,
      "num_tokens": 15431690.0,
      "step": 8890
    },
    {
      "entropy": 5.598479509353638,
      "epoch": 0.6920832522855476,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004957584882859636,
      "loss": 5.3982,
      "mean_token_accuracy": 0.16933944076299667,
      "num_tokens": 15440110.0,
      "step": 8895
    },
    {
      "entropy": 5.663942480087281,
      "epoch": 0.6924722816572652,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004957531304736251,
      "loss": 5.4078,
      "mean_token_accuracy": 0.17168613970279695,
      "num_tokens": 15449022.0,
      "step": 8900
    },
    {
      "entropy": 5.677679395675659,
      "epoch": 0.6928613110289827,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004957477693117091,
      "loss": 5.4359,
      "mean_token_accuracy": 0.16897011548280716,
      "num_tokens": 15457826.0,
      "step": 8905
    },
    {
      "entropy": 5.600473356246948,
      "epoch": 0.6932503404007002,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004957424048002971,
      "loss": 5.3552,
      "mean_token_accuracy": 0.17295342981815337,
      "num_tokens": 15466631.0,
      "step": 8910
    },
    {
      "entropy": 5.5296543598175045,
      "epoch": 0.6936393697724178,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004957370369394706,
      "loss": 5.2514,
      "mean_token_accuracy": 0.17798096388578416,
      "num_tokens": 15474666.0,
      "step": 8915
    },
    {
      "entropy": 5.53621883392334,
      "epoch": 0.6940283991441354,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004957316657293107,
      "loss": 5.2409,
      "mean_token_accuracy": 0.1825147047638893,
      "num_tokens": 15483576.0,
      "step": 8920
    },
    {
      "entropy": 5.751343250274658,
      "epoch": 0.694417428515853,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004957262911698992,
      "loss": 5.3941,
      "mean_token_accuracy": 0.17334622144699097,
      "num_tokens": 15491313.0,
      "step": 8925
    },
    {
      "entropy": 5.640914344787598,
      "epoch": 0.6948064578875706,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004957209132613175,
      "loss": 5.3363,
      "mean_token_accuracy": 0.17856969833374023,
      "num_tokens": 15500638.0,
      "step": 8930
    },
    {
      "entropy": 5.565763854980469,
      "epoch": 0.695195487259288,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004957155320036473,
      "loss": 5.3285,
      "mean_token_accuracy": 0.17599815875291824,
      "num_tokens": 15509170.0,
      "step": 8935
    },
    {
      "entropy": 5.678889417648316,
      "epoch": 0.6955845166310056,
      "grad_norm": 1.0,
      "learning_rate": 0.0004957101473969702,
      "loss": 5.4244,
      "mean_token_accuracy": 0.16916741132736207,
      "num_tokens": 15518108.0,
      "step": 8940
    },
    {
      "entropy": 5.578175735473633,
      "epoch": 0.6959735460027232,
      "grad_norm": 1.0,
      "learning_rate": 0.000495704759441368,
      "loss": 5.29,
      "mean_token_accuracy": 0.1825377270579338,
      "num_tokens": 15526558.0,
      "step": 8945
    },
    {
      "entropy": 5.577849960327148,
      "epoch": 0.6963625753744408,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004956993681369221,
      "loss": 5.2647,
      "mean_token_accuracy": 0.17751545161008836,
      "num_tokens": 15535685.0,
      "step": 8950
    },
    {
      "entropy": 5.555082225799561,
      "epoch": 0.6967516047461584,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004956939734837148,
      "loss": 5.3069,
      "mean_token_accuracy": 0.18180457949638368,
      "num_tokens": 15544072.0,
      "step": 8955
    },
    {
      "entropy": 5.575622272491455,
      "epoch": 0.6971406341178759,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004956885754818277,
      "loss": 5.4028,
      "mean_token_accuracy": 0.1679277539253235,
      "num_tokens": 15553303.0,
      "step": 8960
    },
    {
      "entropy": 5.61887001991272,
      "epoch": 0.6975296634895934,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004956831741313427,
      "loss": 5.3316,
      "mean_token_accuracy": 0.17559644281864167,
      "num_tokens": 15561942.0,
      "step": 8965
    },
    {
      "entropy": 5.5660655975341795,
      "epoch": 0.697918692861311,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004956777694323418,
      "loss": 5.1906,
      "mean_token_accuracy": 0.18667613863945007,
      "num_tokens": 15569988.0,
      "step": 8970
    },
    {
      "entropy": 5.58125638961792,
      "epoch": 0.6983077222330286,
      "grad_norm": 1.015625,
      "learning_rate": 0.000495672361384907,
      "loss": 5.3878,
      "mean_token_accuracy": 0.1742572844028473,
      "num_tokens": 15579816.0,
      "step": 8975
    },
    {
      "entropy": 5.611002349853516,
      "epoch": 0.6986967516047462,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956669499891202,
      "loss": 5.2717,
      "mean_token_accuracy": 0.17696754336357118,
      "num_tokens": 15587843.0,
      "step": 8980
    },
    {
      "entropy": 5.579117298126221,
      "epoch": 0.6990857809764637,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004956615352450639,
      "loss": 5.3207,
      "mean_token_accuracy": 0.17747793793678285,
      "num_tokens": 15596798.0,
      "step": 8985
    },
    {
      "entropy": 5.588120603561402,
      "epoch": 0.6994748103481813,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004956561171528198,
      "loss": 5.4235,
      "mean_token_accuracy": 0.1786605969071388,
      "num_tokens": 15605360.0,
      "step": 8990
    },
    {
      "entropy": 5.573551416397095,
      "epoch": 0.6998638397198989,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004956506957124704,
      "loss": 5.2779,
      "mean_token_accuracy": 0.18233997970819474,
      "num_tokens": 15613557.0,
      "step": 8995
    },
    {
      "entropy": 5.566599798202515,
      "epoch": 0.7002528690916164,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000495645270924098,
      "loss": 5.3547,
      "mean_token_accuracy": 0.17617263793945312,
      "num_tokens": 15622144.0,
      "step": 9000
    },
    {
      "epoch": 0.7002528690916164,
      "eval_entropy": 5.384195727218907,
      "eval_loss": 5.35413932800293,
      "eval_mean_token_accuracy": 0.18322673444568938,
      "eval_num_tokens": 15622144.0,
      "eval_runtime": 21.6228,
      "eval_samples_per_second": 1921.957,
      "eval_steps_per_second": 240.256,
      "step": 9000
    },
    {
      "entropy": 5.5418754577636715,
      "epoch": 0.700641898463334,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004956398427877847,
      "loss": 5.2871,
      "mean_token_accuracy": 0.17854682952165604,
      "num_tokens": 15631119.0,
      "step": 9005
    },
    {
      "entropy": 5.5199097156524655,
      "epoch": 0.7010309278350515,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000495634411303613,
      "loss": 5.204,
      "mean_token_accuracy": 0.18505824506282806,
      "num_tokens": 15639969.0,
      "step": 9010
    },
    {
      "entropy": 5.551648426055908,
      "epoch": 0.7014199572067691,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004956289764716654,
      "loss": 5.3377,
      "mean_token_accuracy": 0.17793741226196289,
      "num_tokens": 15648522.0,
      "step": 9015
    },
    {
      "entropy": 5.524827766418457,
      "epoch": 0.7018089865784867,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004956235382920241,
      "loss": 5.2618,
      "mean_token_accuracy": 0.17254749685525894,
      "num_tokens": 15656834.0,
      "step": 9020
    },
    {
      "entropy": 5.66479754447937,
      "epoch": 0.7021980159502043,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956180967647717,
      "loss": 5.4573,
      "mean_token_accuracy": 0.1677004277706146,
      "num_tokens": 15665066.0,
      "step": 9025
    },
    {
      "entropy": 5.605295324325562,
      "epoch": 0.7025870453219218,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004956126518899911,
      "loss": 5.3422,
      "mean_token_accuracy": 0.17819261848926543,
      "num_tokens": 15673733.0,
      "step": 9030
    },
    {
      "entropy": 5.6864011764526365,
      "epoch": 0.7029760746936393,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956072036677644,
      "loss": 5.3775,
      "mean_token_accuracy": 0.1767784833908081,
      "num_tokens": 15682212.0,
      "step": 9035
    },
    {
      "entropy": 5.5967779636383055,
      "epoch": 0.7033651040653569,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004956017520981746,
      "loss": 5.3616,
      "mean_token_accuracy": 0.1693396046757698,
      "num_tokens": 15690518.0,
      "step": 9040
    },
    {
      "entropy": 5.656971168518067,
      "epoch": 0.7037541334370745,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004955962971813044,
      "loss": 5.41,
      "mean_token_accuracy": 0.1654631972312927,
      "num_tokens": 15699474.0,
      "step": 9045
    },
    {
      "entropy": 5.683543634414673,
      "epoch": 0.7041431628087921,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004955908389172364,
      "loss": 5.4212,
      "mean_token_accuracy": 0.17195746153593064,
      "num_tokens": 15708571.0,
      "step": 9050
    },
    {
      "entropy": 5.708956527709961,
      "epoch": 0.7045321921805097,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004955853773060536,
      "loss": 5.4089,
      "mean_token_accuracy": 0.1737460657954216,
      "num_tokens": 15717307.0,
      "step": 9055
    },
    {
      "entropy": 5.600407600402832,
      "epoch": 0.7049212215522271,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004955799123478388,
      "loss": 5.1901,
      "mean_token_accuracy": 0.18662470281124116,
      "num_tokens": 15725502.0,
      "step": 9060
    },
    {
      "entropy": 5.536356353759766,
      "epoch": 0.7053102509239447,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004955744440426748,
      "loss": 5.4074,
      "mean_token_accuracy": 0.1703765869140625,
      "num_tokens": 15734805.0,
      "step": 9065
    },
    {
      "entropy": 5.601238012313843,
      "epoch": 0.7056992802956623,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004955689723906448,
      "loss": 5.258,
      "mean_token_accuracy": 0.1776318818330765,
      "num_tokens": 15743313.0,
      "step": 9070
    },
    {
      "entropy": 5.553364896774292,
      "epoch": 0.7060883096673799,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004955634973918318,
      "loss": 5.2177,
      "mean_token_accuracy": 0.18231946974992752,
      "num_tokens": 15752552.0,
      "step": 9075
    },
    {
      "entropy": 5.623722982406616,
      "epoch": 0.7064773390390975,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004955580190463186,
      "loss": 5.3135,
      "mean_token_accuracy": 0.1862163260579109,
      "num_tokens": 15761404.0,
      "step": 9080
    },
    {
      "entropy": 5.648780250549317,
      "epoch": 0.7068663684108151,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004955525373541886,
      "loss": 5.3778,
      "mean_token_accuracy": 0.17349406629800795,
      "num_tokens": 15770605.0,
      "step": 9085
    },
    {
      "entropy": 5.516891813278198,
      "epoch": 0.7072553977825325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004955470523155249,
      "loss": 5.268,
      "mean_token_accuracy": 0.17934121787548066,
      "num_tokens": 15778047.0,
      "step": 9090
    },
    {
      "entropy": 5.537967300415039,
      "epoch": 0.7076444271542501,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004955415639304107,
      "loss": 5.3323,
      "mean_token_accuracy": 0.17830815613269807,
      "num_tokens": 15788304.0,
      "step": 9095
    },
    {
      "entropy": 5.617676544189453,
      "epoch": 0.7080334565259677,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004955360721989292,
      "loss": 5.3359,
      "mean_token_accuracy": 0.17249158769845963,
      "num_tokens": 15796448.0,
      "step": 9100
    },
    {
      "entropy": 5.555719661712646,
      "epoch": 0.7084224858976853,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004955305771211641,
      "loss": 5.2904,
      "mean_token_accuracy": 0.17670710235834122,
      "num_tokens": 15805072.0,
      "step": 9105
    },
    {
      "entropy": 5.572696495056152,
      "epoch": 0.7088115152694029,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004955250786971982,
      "loss": 5.3442,
      "mean_token_accuracy": 0.1760936662554741,
      "num_tokens": 15813918.0,
      "step": 9110
    },
    {
      "entropy": 5.7184289455413815,
      "epoch": 0.7092005446411204,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004955195769271154,
      "loss": 5.4707,
      "mean_token_accuracy": 0.17003964632749557,
      "num_tokens": 15822675.0,
      "step": 9115
    },
    {
      "entropy": 5.6127406597137455,
      "epoch": 0.709589574012838,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004955140718109991,
      "loss": 5.356,
      "mean_token_accuracy": 0.17116720974445343,
      "num_tokens": 15831056.0,
      "step": 9120
    },
    {
      "entropy": 5.630828237533569,
      "epoch": 0.7099786033845555,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004955085633489326,
      "loss": 5.4178,
      "mean_token_accuracy": 0.16970840990543365,
      "num_tokens": 15839948.0,
      "step": 9125
    },
    {
      "entropy": 5.632455205917358,
      "epoch": 0.7103676327562731,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004955030515409997,
      "loss": 5.316,
      "mean_token_accuracy": 0.1776179775595665,
      "num_tokens": 15848738.0,
      "step": 9130
    },
    {
      "entropy": 5.609870433807373,
      "epoch": 0.7107566621279907,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004954975363872838,
      "loss": 5.3776,
      "mean_token_accuracy": 0.1716231659054756,
      "num_tokens": 15859073.0,
      "step": 9135
    },
    {
      "entropy": 5.601574182510376,
      "epoch": 0.7111456914997082,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004954920178878689,
      "loss": 5.2674,
      "mean_token_accuracy": 0.18010449558496475,
      "num_tokens": 15868146.0,
      "step": 9140
    },
    {
      "entropy": 5.528825283050537,
      "epoch": 0.7115347208714258,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004954864960428385,
      "loss": 5.2252,
      "mean_token_accuracy": 0.1761257156729698,
      "num_tokens": 15876915.0,
      "step": 9145
    },
    {
      "entropy": 5.556982183456421,
      "epoch": 0.7119237502431434,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004954809708522765,
      "loss": 5.309,
      "mean_token_accuracy": 0.17798453122377395,
      "num_tokens": 15884947.0,
      "step": 9150
    },
    {
      "entropy": 5.611155080795288,
      "epoch": 0.712312779614861,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004954754423162667,
      "loss": 5.3015,
      "mean_token_accuracy": 0.1726186603307724,
      "num_tokens": 15894595.0,
      "step": 9155
    },
    {
      "entropy": 5.616395807266235,
      "epoch": 0.7127018089865785,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000495469910434893,
      "loss": 5.2993,
      "mean_token_accuracy": 0.18139362037181855,
      "num_tokens": 15902920.0,
      "step": 9160
    },
    {
      "entropy": 5.5656671047210695,
      "epoch": 0.713090838358296,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004954643752082392,
      "loss": 5.2935,
      "mean_token_accuracy": 0.1780125841498375,
      "num_tokens": 15911414.0,
      "step": 9165
    },
    {
      "entropy": 5.6202209949493405,
      "epoch": 0.7134798677300136,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004954588366363896,
      "loss": 5.3907,
      "mean_token_accuracy": 0.1711605429649353,
      "num_tokens": 15919855.0,
      "step": 9170
    },
    {
      "entropy": 5.557894134521485,
      "epoch": 0.7138688971017312,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004954532947194279,
      "loss": 5.3023,
      "mean_token_accuracy": 0.17727120369672775,
      "num_tokens": 15928932.0,
      "step": 9175
    },
    {
      "entropy": 5.638629388809204,
      "epoch": 0.7142579264734488,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004954477494574384,
      "loss": 5.3705,
      "mean_token_accuracy": 0.17425462752580642,
      "num_tokens": 15938266.0,
      "step": 9180
    },
    {
      "entropy": 5.692516899108886,
      "epoch": 0.7146469558451664,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004954422008505052,
      "loss": 5.3944,
      "mean_token_accuracy": 0.16764615774154662,
      "num_tokens": 15947100.0,
      "step": 9185
    },
    {
      "entropy": 5.611473178863525,
      "epoch": 0.7150359852168838,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004954366488987125,
      "loss": 5.3783,
      "mean_token_accuracy": 0.1700735718011856,
      "num_tokens": 15955544.0,
      "step": 9190
    },
    {
      "entropy": 5.652182483673096,
      "epoch": 0.7154250145886014,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004954310936021444,
      "loss": 5.4001,
      "mean_token_accuracy": 0.17192965745925903,
      "num_tokens": 15963372.0,
      "step": 9195
    },
    {
      "entropy": 5.644375038146973,
      "epoch": 0.715814043960319,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004954255349608853,
      "loss": 5.3771,
      "mean_token_accuracy": 0.16869236081838607,
      "num_tokens": 15972940.0,
      "step": 9200
    },
    {
      "entropy": 5.633125686645508,
      "epoch": 0.7162030733320366,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004954199729750198,
      "loss": 5.3375,
      "mean_token_accuracy": 0.16978246420621873,
      "num_tokens": 15982238.0,
      "step": 9205
    },
    {
      "entropy": 5.624726867675781,
      "epoch": 0.7165921027037542,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004954144076446318,
      "loss": 5.3454,
      "mean_token_accuracy": 0.17483690232038498,
      "num_tokens": 15990835.0,
      "step": 9210
    },
    {
      "entropy": 5.633648681640625,
      "epoch": 0.7169811320754716,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004954088389698061,
      "loss": 5.4455,
      "mean_token_accuracy": 0.17005229145288467,
      "num_tokens": 16000649.0,
      "step": 9215
    },
    {
      "entropy": 5.612905120849609,
      "epoch": 0.7173701614471892,
      "grad_norm": 1.0625,
      "learning_rate": 0.000495403266950627,
      "loss": 5.3601,
      "mean_token_accuracy": 0.17579613029956817,
      "num_tokens": 16009214.0,
      "step": 9220
    },
    {
      "entropy": 5.648853302001953,
      "epoch": 0.7177591908189068,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004953976915871792,
      "loss": 5.3409,
      "mean_token_accuracy": 0.1749788999557495,
      "num_tokens": 16018173.0,
      "step": 9225
    },
    {
      "entropy": 5.549958801269531,
      "epoch": 0.7181482201906244,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004953921128795472,
      "loss": 5.2362,
      "mean_token_accuracy": 0.18928415328264236,
      "num_tokens": 16026857.0,
      "step": 9230
    },
    {
      "entropy": 5.589247131347657,
      "epoch": 0.718537249562342,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004953865308278156,
      "loss": 5.4026,
      "mean_token_accuracy": 0.16704169660806656,
      "num_tokens": 16035792.0,
      "step": 9235
    },
    {
      "entropy": 5.623420095443725,
      "epoch": 0.7189262789340595,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004953809454320693,
      "loss": 5.3325,
      "mean_token_accuracy": 0.1779796898365021,
      "num_tokens": 16044874.0,
      "step": 9240
    },
    {
      "entropy": 5.614606761932373,
      "epoch": 0.7193153083057771,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004953753566923929,
      "loss": 5.3004,
      "mean_token_accuracy": 0.17517977058887482,
      "num_tokens": 16053705.0,
      "step": 9245
    },
    {
      "entropy": 5.5654683113098145,
      "epoch": 0.7197043376774946,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953697646088712,
      "loss": 5.3101,
      "mean_token_accuracy": 0.1744338884949684,
      "num_tokens": 16062987.0,
      "step": 9250
    },
    {
      "entropy": 5.558124971389771,
      "epoch": 0.7200933670492122,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953641691815891,
      "loss": 5.2853,
      "mean_token_accuracy": 0.1788240686058998,
      "num_tokens": 16071247.0,
      "step": 9255
    },
    {
      "entropy": 5.530323934555054,
      "epoch": 0.7204823964209298,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953585704106315,
      "loss": 5.2264,
      "mean_token_accuracy": 0.1773939922451973,
      "num_tokens": 16080190.0,
      "step": 9260
    },
    {
      "entropy": 5.575647354125977,
      "epoch": 0.7208714257926473,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004953529682960832,
      "loss": 5.3155,
      "mean_token_accuracy": 0.17726240903139115,
      "num_tokens": 16088885.0,
      "step": 9265
    },
    {
      "entropy": 5.561666536331177,
      "epoch": 0.7212604551643649,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004953473628380295,
      "loss": 5.2756,
      "mean_token_accuracy": 0.17821484357118605,
      "num_tokens": 16097202.0,
      "step": 9270
    },
    {
      "entropy": 5.588506174087525,
      "epoch": 0.7216494845360825,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004953417540365553,
      "loss": 5.2431,
      "mean_token_accuracy": 0.18142793625593184,
      "num_tokens": 16105468.0,
      "step": 9275
    },
    {
      "entropy": 5.64068570137024,
      "epoch": 0.7220385139078,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004953361418917455,
      "loss": 5.3979,
      "mean_token_accuracy": 0.1801755577325821,
      "num_tokens": 16114624.0,
      "step": 9280
    },
    {
      "entropy": 5.530179595947265,
      "epoch": 0.7224275432795176,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004953305264036856,
      "loss": 5.2353,
      "mean_token_accuracy": 0.17767134457826614,
      "num_tokens": 16123983.0,
      "step": 9285
    },
    {
      "entropy": 5.638651132583618,
      "epoch": 0.7228165726512352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004953249075724607,
      "loss": 5.3471,
      "mean_token_accuracy": 0.17877227663993836,
      "num_tokens": 16132886.0,
      "step": 9290
    },
    {
      "entropy": 5.554377841949463,
      "epoch": 0.7232056020229527,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004953192853981559,
      "loss": 5.1984,
      "mean_token_accuracy": 0.1873273730278015,
      "num_tokens": 16141526.0,
      "step": 9295
    },
    {
      "entropy": 5.576457214355469,
      "epoch": 0.7235946313946703,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953136598808566,
      "loss": 5.3813,
      "mean_token_accuracy": 0.18493103832006455,
      "num_tokens": 16149949.0,
      "step": 9300
    },
    {
      "entropy": 5.608815240859985,
      "epoch": 0.7239836607663879,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004953080310206481,
      "loss": 5.3449,
      "mean_token_accuracy": 0.17716638147830963,
      "num_tokens": 16158131.0,
      "step": 9305
    },
    {
      "entropy": 5.501528787612915,
      "epoch": 0.7243726901381055,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004953023988176162,
      "loss": 5.2499,
      "mean_token_accuracy": 0.17929585576057433,
      "num_tokens": 16167131.0,
      "step": 9310
    },
    {
      "entropy": 5.6442553997039795,
      "epoch": 0.724761719509823,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004952967632718458,
      "loss": 5.4633,
      "mean_token_accuracy": 0.1731911689043045,
      "num_tokens": 16176198.0,
      "step": 9315
    },
    {
      "entropy": 5.592202043533325,
      "epoch": 0.7251507488815405,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004952911243834227,
      "loss": 5.3427,
      "mean_token_accuracy": 0.17611104547977446,
      "num_tokens": 16184860.0,
      "step": 9320
    },
    {
      "entropy": 5.593503427505493,
      "epoch": 0.7255397782532581,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004952854821524324,
      "loss": 5.328,
      "mean_token_accuracy": 0.17426833510398865,
      "num_tokens": 16193126.0,
      "step": 9325
    },
    {
      "entropy": 5.5762348651885985,
      "epoch": 0.7259288076249757,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004952798365789606,
      "loss": 5.3067,
      "mean_token_accuracy": 0.17548383623361588,
      "num_tokens": 16201763.0,
      "step": 9330
    },
    {
      "entropy": 5.575831174850464,
      "epoch": 0.7263178369966933,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004952741876630928,
      "loss": 5.2582,
      "mean_token_accuracy": 0.18297910690307617,
      "num_tokens": 16210947.0,
      "step": 9335
    },
    {
      "entropy": 5.614086771011353,
      "epoch": 0.7267068663684109,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004952685354049148,
      "loss": 5.3308,
      "mean_token_accuracy": 0.17360346168279647,
      "num_tokens": 16219772.0,
      "step": 9340
    },
    {
      "entropy": 5.637112236022949,
      "epoch": 0.7270958957401283,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004952628798045124,
      "loss": 5.3211,
      "mean_token_accuracy": 0.18236895203590392,
      "num_tokens": 16227759.0,
      "step": 9345
    },
    {
      "entropy": 5.549534940719605,
      "epoch": 0.7274849251118459,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004952572208619714,
      "loss": 5.2177,
      "mean_token_accuracy": 0.17939460426568984,
      "num_tokens": 16236913.0,
      "step": 9350
    },
    {
      "entropy": 5.581374168395996,
      "epoch": 0.7278739544835635,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004952515585773778,
      "loss": 5.3427,
      "mean_token_accuracy": 0.17206512838602067,
      "num_tokens": 16246190.0,
      "step": 9355
    },
    {
      "entropy": 5.5773841381073,
      "epoch": 0.7282629838552811,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004952458929508171,
      "loss": 5.3031,
      "mean_token_accuracy": 0.1751967504620552,
      "num_tokens": 16255476.0,
      "step": 9360
    },
    {
      "entropy": 5.584682369232178,
      "epoch": 0.7286520132269987,
      "grad_norm": 1.0,
      "learning_rate": 0.0004952402239823757,
      "loss": 5.331,
      "mean_token_accuracy": 0.1706559732556343,
      "num_tokens": 16264701.0,
      "step": 9365
    },
    {
      "entropy": 5.441281652450561,
      "epoch": 0.7290410425987162,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004952345516721393,
      "loss": 5.1648,
      "mean_token_accuracy": 0.18849799335002898,
      "num_tokens": 16272999.0,
      "step": 9370
    },
    {
      "entropy": 5.66496958732605,
      "epoch": 0.7294300719704337,
      "grad_norm": 1.0,
      "learning_rate": 0.0004952288760201942,
      "loss": 5.4418,
      "mean_token_accuracy": 0.17025233134627343,
      "num_tokens": 16282537.0,
      "step": 9375
    },
    {
      "entropy": 5.633242988586426,
      "epoch": 0.7298191013421513,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004952231970266265,
      "loss": 5.3001,
      "mean_token_accuracy": 0.17955290973186494,
      "num_tokens": 16291155.0,
      "step": 9380
    },
    {
      "entropy": 5.643191957473755,
      "epoch": 0.7302081307138689,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004952175146915223,
      "loss": 5.3829,
      "mean_token_accuracy": 0.1692135140299797,
      "num_tokens": 16300466.0,
      "step": 9385
    },
    {
      "entropy": 5.596168661117554,
      "epoch": 0.7305971600855865,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004952118290149679,
      "loss": 5.2849,
      "mean_token_accuracy": 0.1738765999674797,
      "num_tokens": 16309250.0,
      "step": 9390
    },
    {
      "entropy": 5.647305631637574,
      "epoch": 0.730986189457304,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004952061399970492,
      "loss": 5.3829,
      "mean_token_accuracy": 0.17449599206447602,
      "num_tokens": 16318056.0,
      "step": 9395
    },
    {
      "entropy": 5.538020944595337,
      "epoch": 0.7313752188290216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000495200447637853,
      "loss": 5.2386,
      "mean_token_accuracy": 0.17811905592679977,
      "num_tokens": 16326455.0,
      "step": 9400
    },
    {
      "entropy": 5.510137224197388,
      "epoch": 0.7317642482007392,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004951947519374655,
      "loss": 5.227,
      "mean_token_accuracy": 0.18160680383443834,
      "num_tokens": 16334584.0,
      "step": 9405
    },
    {
      "entropy": 5.5503466606140135,
      "epoch": 0.7321532775724567,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004951890528959731,
      "loss": 5.2573,
      "mean_token_accuracy": 0.18433501571416855,
      "num_tokens": 16342361.0,
      "step": 9410
    },
    {
      "entropy": 5.635887479782104,
      "epoch": 0.7325423069441743,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004951833505134623,
      "loss": 5.3919,
      "mean_token_accuracy": 0.17290140837430953,
      "num_tokens": 16350414.0,
      "step": 9415
    },
    {
      "entropy": 5.500445413589477,
      "epoch": 0.7329313363158918,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004951776447900196,
      "loss": 5.2269,
      "mean_token_accuracy": 0.18617474734783174,
      "num_tokens": 16359266.0,
      "step": 9420
    },
    {
      "entropy": 5.4902732372283936,
      "epoch": 0.7333203656876094,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004951719357257317,
      "loss": 5.2058,
      "mean_token_accuracy": 0.1826066866517067,
      "num_tokens": 16367558.0,
      "step": 9425
    },
    {
      "entropy": 5.451519298553467,
      "epoch": 0.733709395059327,
      "grad_norm": 1.015625,
      "learning_rate": 0.000495166223320685,
      "loss": 5.1901,
      "mean_token_accuracy": 0.184382164478302,
      "num_tokens": 16375690.0,
      "step": 9430
    },
    {
      "entropy": 5.5959672927856445,
      "epoch": 0.7340984244310446,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004951605075749662,
      "loss": 5.3103,
      "mean_token_accuracy": 0.17972762435674666,
      "num_tokens": 16384157.0,
      "step": 9435
    },
    {
      "entropy": 5.613910102844239,
      "epoch": 0.7344874538027621,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004951547884886623,
      "loss": 5.3572,
      "mean_token_accuracy": 0.18060232400894166,
      "num_tokens": 16392153.0,
      "step": 9440
    },
    {
      "entropy": 5.457367038726806,
      "epoch": 0.7348764831744796,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004951490660618598,
      "loss": 5.1221,
      "mean_token_accuracy": 0.1908814549446106,
      "num_tokens": 16400132.0,
      "step": 9445
    },
    {
      "entropy": 5.553274536132813,
      "epoch": 0.7352655125461972,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004951433402946457,
      "loss": 5.2937,
      "mean_token_accuracy": 0.18400580137968064,
      "num_tokens": 16408297.0,
      "step": 9450
    },
    {
      "entropy": 5.5737401962280275,
      "epoch": 0.7356545419179148,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004951376111871068,
      "loss": 5.3443,
      "mean_token_accuracy": 0.17686122059822082,
      "num_tokens": 16417977.0,
      "step": 9455
    },
    {
      "entropy": 5.58512773513794,
      "epoch": 0.7360435712896324,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004951318787393299,
      "loss": 5.2265,
      "mean_token_accuracy": 0.18320877254009246,
      "num_tokens": 16427156.0,
      "step": 9460
    },
    {
      "entropy": 5.64531135559082,
      "epoch": 0.73643260066135,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004951261429514023,
      "loss": 5.4389,
      "mean_token_accuracy": 0.16900589615106582,
      "num_tokens": 16435845.0,
      "step": 9465
    },
    {
      "entropy": 5.6111023902893065,
      "epoch": 0.7368216300330674,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004951204038234106,
      "loss": 5.3945,
      "mean_token_accuracy": 0.17545305341482162,
      "num_tokens": 16445876.0,
      "step": 9470
    },
    {
      "entropy": 5.601052713394165,
      "epoch": 0.737210659404785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004951146613554424,
      "loss": 5.3357,
      "mean_token_accuracy": 0.18090222775936127,
      "num_tokens": 16454688.0,
      "step": 9475
    },
    {
      "entropy": 5.642931413650513,
      "epoch": 0.7375996887765026,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004951089155475843,
      "loss": 5.4152,
      "mean_token_accuracy": 0.16919267326593398,
      "num_tokens": 16463932.0,
      "step": 9480
    },
    {
      "entropy": 5.616369152069092,
      "epoch": 0.7379887181482202,
      "grad_norm": 0.984375,
      "learning_rate": 0.000495103166399924,
      "loss": 5.3242,
      "mean_token_accuracy": 0.17350419461727143,
      "num_tokens": 16472627.0,
      "step": 9485
    },
    {
      "entropy": 5.644665336608886,
      "epoch": 0.7383777475199378,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004950974139125484,
      "loss": 5.3634,
      "mean_token_accuracy": 0.17660223096609115,
      "num_tokens": 16480819.0,
      "step": 9490
    },
    {
      "entropy": 5.600194644927979,
      "epoch": 0.7387667768916554,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004950916580855448,
      "loss": 5.3286,
      "mean_token_accuracy": 0.17571170181035994,
      "num_tokens": 16489055.0,
      "step": 9495
    },
    {
      "entropy": 5.548560953140258,
      "epoch": 0.7391558062633728,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004950858989190007,
      "loss": 5.3297,
      "mean_token_accuracy": 0.1835804373025894,
      "num_tokens": 16497375.0,
      "step": 9500
    },
    {
      "entropy": 5.493134450912476,
      "epoch": 0.7395448356350904,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004950801364130032,
      "loss": 5.1607,
      "mean_token_accuracy": 0.18572535663843154,
      "num_tokens": 16505183.0,
      "step": 9505
    },
    {
      "entropy": 5.597559452056885,
      "epoch": 0.739933865006808,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004950743705676401,
      "loss": 5.4381,
      "mean_token_accuracy": 0.16902757957577705,
      "num_tokens": 16514018.0,
      "step": 9510
    },
    {
      "entropy": 5.615971994400025,
      "epoch": 0.7403228943785256,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004950686013829987,
      "loss": 5.3076,
      "mean_token_accuracy": 0.18013171702623368,
      "num_tokens": 16522954.0,
      "step": 9515
    },
    {
      "entropy": 5.71082763671875,
      "epoch": 0.7407119237502432,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950628288591665,
      "loss": 5.3547,
      "mean_token_accuracy": 0.17512065768241883,
      "num_tokens": 16530845.0,
      "step": 9520
    },
    {
      "entropy": 5.565211915969849,
      "epoch": 0.7411009531219607,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950570529962311,
      "loss": 5.3313,
      "mean_token_accuracy": 0.1752915546298027,
      "num_tokens": 16540464.0,
      "step": 9525
    },
    {
      "entropy": 5.473846054077148,
      "epoch": 0.7414899824936783,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004950512737942803,
      "loss": 5.2103,
      "mean_token_accuracy": 0.18890926241874695,
      "num_tokens": 16549485.0,
      "step": 9530
    },
    {
      "entropy": 5.6329456806182865,
      "epoch": 0.7418790118653958,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004950454912534015,
      "loss": 5.3656,
      "mean_token_accuracy": 0.17809097170829774,
      "num_tokens": 16557854.0,
      "step": 9535
    },
    {
      "entropy": 5.635037088394165,
      "epoch": 0.7422680412371134,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004950397053736827,
      "loss": 5.3982,
      "mean_token_accuracy": 0.17356866598129272,
      "num_tokens": 16565975.0,
      "step": 9540
    },
    {
      "entropy": 5.603233718872071,
      "epoch": 0.742657070608831,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004950339161552117,
      "loss": 5.366,
      "mean_token_accuracy": 0.17709572613239288,
      "num_tokens": 16575038.0,
      "step": 9545
    },
    {
      "entropy": 5.536765909194946,
      "epoch": 0.7430460999805485,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004950281235980761,
      "loss": 5.2234,
      "mean_token_accuracy": 0.18227089196443558,
      "num_tokens": 16584292.0,
      "step": 9550
    },
    {
      "entropy": 5.583617877960205,
      "epoch": 0.7434351293522661,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000495022327702364,
      "loss": 5.3176,
      "mean_token_accuracy": 0.1783987820148468,
      "num_tokens": 16592331.0,
      "step": 9555
    },
    {
      "entropy": 5.557074785232544,
      "epoch": 0.7438241587239837,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004950165284681631,
      "loss": 5.2694,
      "mean_token_accuracy": 0.17992899864912032,
      "num_tokens": 16601845.0,
      "step": 9560
    },
    {
      "entropy": 5.584936475753784,
      "epoch": 0.7442131880957013,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004950107258955618,
      "loss": 5.3511,
      "mean_token_accuracy": 0.17872835099697112,
      "num_tokens": 16610438.0,
      "step": 9565
    },
    {
      "entropy": 5.540927791595459,
      "epoch": 0.7446022174674188,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004950049199846479,
      "loss": 5.2705,
      "mean_token_accuracy": 0.18069363981485367,
      "num_tokens": 16618892.0,
      "step": 9570
    },
    {
      "entropy": 5.639248418807983,
      "epoch": 0.7449912468391363,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004949991107355095,
      "loss": 5.3096,
      "mean_token_accuracy": 0.17804142832756042,
      "num_tokens": 16627702.0,
      "step": 9575
    },
    {
      "entropy": 5.51157283782959,
      "epoch": 0.7453802762108539,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004949932981482347,
      "loss": 5.2218,
      "mean_token_accuracy": 0.18788692057132722,
      "num_tokens": 16635653.0,
      "step": 9580
    },
    {
      "entropy": 5.5664544105529785,
      "epoch": 0.7457693055825715,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004949874822229118,
      "loss": 5.3531,
      "mean_token_accuracy": 0.18009264767169952,
      "num_tokens": 16643898.0,
      "step": 9585
    },
    {
      "entropy": 5.596207046508789,
      "epoch": 0.7461583349542891,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004949816629596288,
      "loss": 5.2252,
      "mean_token_accuracy": 0.18270762860774994,
      "num_tokens": 16653126.0,
      "step": 9590
    },
    {
      "entropy": 5.656696367263794,
      "epoch": 0.7465473643260067,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004949758403584746,
      "loss": 5.3469,
      "mean_token_accuracy": 0.1753896027803421,
      "num_tokens": 16661561.0,
      "step": 9595
    },
    {
      "entropy": 5.544621753692627,
      "epoch": 0.7469363936977241,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004949700144195368,
      "loss": 5.3028,
      "mean_token_accuracy": 0.17755116224288942,
      "num_tokens": 16669657.0,
      "step": 9600
    },
    {
      "entropy": 5.519941425323486,
      "epoch": 0.7473254230694417,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004949641851429043,
      "loss": 5.3038,
      "mean_token_accuracy": 0.1791791334748268,
      "num_tokens": 16678486.0,
      "step": 9605
    },
    {
      "entropy": 5.595032501220703,
      "epoch": 0.7477144524411593,
      "grad_norm": 1.0,
      "learning_rate": 0.0004949583525286654,
      "loss": 5.2735,
      "mean_token_accuracy": 0.177974633872509,
      "num_tokens": 16686638.0,
      "step": 9610
    },
    {
      "entropy": 5.594528484344482,
      "epoch": 0.7481034818128769,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004949525165769085,
      "loss": 5.2215,
      "mean_token_accuracy": 0.17921450287103652,
      "num_tokens": 16694393.0,
      "step": 9615
    },
    {
      "entropy": 5.568652772903443,
      "epoch": 0.7484925111845945,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004949466772877224,
      "loss": 5.2646,
      "mean_token_accuracy": 0.18511301875114441,
      "num_tokens": 16702613.0,
      "step": 9620
    },
    {
      "entropy": 5.561077070236206,
      "epoch": 0.748881540556312,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004949408346611955,
      "loss": 5.2364,
      "mean_token_accuracy": 0.18199257850646972,
      "num_tokens": 16711210.0,
      "step": 9625
    },
    {
      "entropy": 5.564145612716675,
      "epoch": 0.7492705699280295,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004949349886974165,
      "loss": 5.2972,
      "mean_token_accuracy": 0.1834089621901512,
      "num_tokens": 16719124.0,
      "step": 9630
    },
    {
      "entropy": 5.600499248504638,
      "epoch": 0.7496595992997471,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004949291393964741,
      "loss": 5.2676,
      "mean_token_accuracy": 0.17806679606437684,
      "num_tokens": 16727240.0,
      "step": 9635
    },
    {
      "entropy": 5.576705551147461,
      "epoch": 0.7500486286714647,
      "grad_norm": 1.03125,
      "learning_rate": 0.000494923286758457,
      "loss": 5.38,
      "mean_token_accuracy": 0.17818381488323212,
      "num_tokens": 16736524.0,
      "step": 9640
    },
    {
      "entropy": 5.704974842071533,
      "epoch": 0.7504376580431823,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004949174307834541,
      "loss": 5.4922,
      "mean_token_accuracy": 0.16476022750139235,
      "num_tokens": 16745798.0,
      "step": 9645
    },
    {
      "entropy": 5.589550638198853,
      "epoch": 0.7508266874148998,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004949115714715543,
      "loss": 5.2774,
      "mean_token_accuracy": 0.18031025826931,
      "num_tokens": 16754263.0,
      "step": 9650
    },
    {
      "entropy": 5.509631538391114,
      "epoch": 0.7512157167866174,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004949057088228465,
      "loss": 5.1671,
      "mean_token_accuracy": 0.18788377642631532,
      "num_tokens": 16762610.0,
      "step": 9655
    },
    {
      "entropy": 5.587539911270142,
      "epoch": 0.751604746158335,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004948998428374194,
      "loss": 5.3813,
      "mean_token_accuracy": 0.17164374738931656,
      "num_tokens": 16771436.0,
      "step": 9660
    },
    {
      "entropy": 5.6791328430175785,
      "epoch": 0.7519937755300525,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948939735153622,
      "loss": 5.3954,
      "mean_token_accuracy": 0.17684406042099,
      "num_tokens": 16779647.0,
      "step": 9665
    },
    {
      "entropy": 5.666022825241089,
      "epoch": 0.7523828049017701,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004948881008567641,
      "loss": 5.352,
      "mean_token_accuracy": 0.1790418267250061,
      "num_tokens": 16788116.0,
      "step": 9670
    },
    {
      "entropy": 5.549437808990478,
      "epoch": 0.7527718342734876,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004948822248617139,
      "loss": 5.2878,
      "mean_token_accuracy": 0.18091578632593155,
      "num_tokens": 16796518.0,
      "step": 9675
    },
    {
      "entropy": 5.55680022239685,
      "epoch": 0.7531608636452052,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000494876345530301,
      "loss": 5.2654,
      "mean_token_accuracy": 0.1813149034976959,
      "num_tokens": 16805511.0,
      "step": 9680
    },
    {
      "entropy": 5.583335542678833,
      "epoch": 0.7535498930169228,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004948704628626145,
      "loss": 5.282,
      "mean_token_accuracy": 0.17796677052974702,
      "num_tokens": 16813173.0,
      "step": 9685
    },
    {
      "entropy": 5.627739143371582,
      "epoch": 0.7539389223886404,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004948645768587437,
      "loss": 5.3799,
      "mean_token_accuracy": 0.17010469883680343,
      "num_tokens": 16822193.0,
      "step": 9690
    },
    {
      "entropy": 5.518557786941528,
      "epoch": 0.7543279517603579,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948586875187778,
      "loss": 5.1055,
      "mean_token_accuracy": 0.1896416053175926,
      "num_tokens": 16830421.0,
      "step": 9695
    },
    {
      "entropy": 5.56982102394104,
      "epoch": 0.7547169811320755,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948527948428064,
      "loss": 5.2132,
      "mean_token_accuracy": 0.18295853286981584,
      "num_tokens": 16838988.0,
      "step": 9700
    },
    {
      "entropy": 5.534185743331909,
      "epoch": 0.755106010503793,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004948468988309187,
      "loss": 5.35,
      "mean_token_accuracy": 0.17599449008703233,
      "num_tokens": 16847980.0,
      "step": 9705
    },
    {
      "entropy": 5.496274757385254,
      "epoch": 0.7554950398755106,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004948409994832043,
      "loss": 5.2328,
      "mean_token_accuracy": 0.18318645805120468,
      "num_tokens": 16857263.0,
      "step": 9710
    },
    {
      "entropy": 5.592087888717652,
      "epoch": 0.7558840692472282,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004948350967997526,
      "loss": 5.2375,
      "mean_token_accuracy": 0.1804831087589264,
      "num_tokens": 16865314.0,
      "step": 9715
    },
    {
      "entropy": 5.618679857254028,
      "epoch": 0.7562730986189458,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948291907806532,
      "loss": 5.3287,
      "mean_token_accuracy": 0.18105187863111497,
      "num_tokens": 16875359.0,
      "step": 9720
    },
    {
      "entropy": 5.626615762710571,
      "epoch": 0.7566621279906633,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004948232814259957,
      "loss": 5.4465,
      "mean_token_accuracy": 0.16737451404333115,
      "num_tokens": 16885279.0,
      "step": 9725
    },
    {
      "entropy": 5.67550139427185,
      "epoch": 0.7570511573623808,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004948173687358699,
      "loss": 5.3648,
      "mean_token_accuracy": 0.17636879682540893,
      "num_tokens": 16894463.0,
      "step": 9730
    },
    {
      "entropy": 5.615520334243774,
      "epoch": 0.7574401867340984,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004948114527103652,
      "loss": 5.3019,
      "mean_token_accuracy": 0.16981316953897477,
      "num_tokens": 16903378.0,
      "step": 9735
    },
    {
      "entropy": 5.568920230865478,
      "epoch": 0.757829216105816,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004948055333495717,
      "loss": 5.3493,
      "mean_token_accuracy": 0.17093088179826738,
      "num_tokens": 16912039.0,
      "step": 9740
    },
    {
      "entropy": 5.553165483474731,
      "epoch": 0.7582182454775336,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947996106535791,
      "loss": 5.2949,
      "mean_token_accuracy": 0.17178265303373336,
      "num_tokens": 16920516.0,
      "step": 9745
    },
    {
      "entropy": 5.661890935897827,
      "epoch": 0.7586072748492512,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004947936846224773,
      "loss": 5.3096,
      "mean_token_accuracy": 0.17241132408380508,
      "num_tokens": 16929131.0,
      "step": 9750
    },
    {
      "entropy": 5.539224529266358,
      "epoch": 0.7589963042209686,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004947877552563562,
      "loss": 5.3331,
      "mean_token_accuracy": 0.18203827887773513,
      "num_tokens": 16937352.0,
      "step": 9755
    },
    {
      "entropy": 5.593630933761597,
      "epoch": 0.7593853335926862,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004947818225553056,
      "loss": 5.4071,
      "mean_token_accuracy": 0.17220403701066972,
      "num_tokens": 16946439.0,
      "step": 9760
    },
    {
      "entropy": 5.6424661636352536,
      "epoch": 0.7597743629644038,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004947758865194156,
      "loss": 5.2918,
      "mean_token_accuracy": 0.17695994526147843,
      "num_tokens": 16955402.0,
      "step": 9765
    },
    {
      "entropy": 5.495573997497559,
      "epoch": 0.7601633923361214,
      "grad_norm": 1.0,
      "learning_rate": 0.0004947699471487766,
      "loss": 5.2244,
      "mean_token_accuracy": 0.18328388929367065,
      "num_tokens": 16964299.0,
      "step": 9770
    },
    {
      "entropy": 5.50468430519104,
      "epoch": 0.760552421707839,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004947640044434782,
      "loss": 5.2758,
      "mean_token_accuracy": 0.17400199323892593,
      "num_tokens": 16973483.0,
      "step": 9775
    },
    {
      "entropy": 5.590785169601441,
      "epoch": 0.7609414510795565,
      "grad_norm": 1.0,
      "learning_rate": 0.0004947580584036109,
      "loss": 5.3239,
      "mean_token_accuracy": 0.18279939740896226,
      "num_tokens": 16982234.0,
      "step": 9780
    },
    {
      "entropy": 5.581445789337158,
      "epoch": 0.761330480451274,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004947521090292649,
      "loss": 5.2892,
      "mean_token_accuracy": 0.1782561406493187,
      "num_tokens": 16990669.0,
      "step": 9785
    },
    {
      "entropy": 5.648339176177979,
      "epoch": 0.7617195098229916,
      "grad_norm": 1.0,
      "learning_rate": 0.0004947461563205304,
      "loss": 5.3507,
      "mean_token_accuracy": 0.17734525352716446,
      "num_tokens": 16999998.0,
      "step": 9790
    },
    {
      "entropy": 5.661034297943115,
      "epoch": 0.7621085391947092,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004947402002774977,
      "loss": 5.3529,
      "mean_token_accuracy": 0.1741270527243614,
      "num_tokens": 17008518.0,
      "step": 9795
    },
    {
      "entropy": 5.543142318725586,
      "epoch": 0.7624975685664268,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947342409002572,
      "loss": 5.2809,
      "mean_token_accuracy": 0.17424834668636321,
      "num_tokens": 17017182.0,
      "step": 9800
    },
    {
      "entropy": 5.565796613693237,
      "epoch": 0.7628865979381443,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004947282781888994,
      "loss": 5.2961,
      "mean_token_accuracy": 0.17821226865053177,
      "num_tokens": 17026186.0,
      "step": 9805
    },
    {
      "entropy": 5.473713350296021,
      "epoch": 0.7632756273098619,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004947223121435147,
      "loss": 5.175,
      "mean_token_accuracy": 0.19047538787126542,
      "num_tokens": 17034295.0,
      "step": 9810
    },
    {
      "entropy": 5.44863224029541,
      "epoch": 0.7636646566815795,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004947163427641936,
      "loss": 5.252,
      "mean_token_accuracy": 0.17861233055591583,
      "num_tokens": 17043195.0,
      "step": 9815
    },
    {
      "entropy": 5.518514013290405,
      "epoch": 0.764053686053297,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004947103700510268,
      "loss": 5.1959,
      "mean_token_accuracy": 0.17903678566217424,
      "num_tokens": 17051903.0,
      "step": 9820
    },
    {
      "entropy": 5.539800930023193,
      "epoch": 0.7644427154250146,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004947043940041047,
      "loss": 5.2636,
      "mean_token_accuracy": 0.1832680806517601,
      "num_tokens": 17060495.0,
      "step": 9825
    },
    {
      "entropy": 5.601056957244873,
      "epoch": 0.7648317447967321,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004946984146235183,
      "loss": 5.3227,
      "mean_token_accuracy": 0.17461752742528916,
      "num_tokens": 17069254.0,
      "step": 9830
    },
    {
      "entropy": 5.6530539989471436,
      "epoch": 0.7652207741684497,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004946924319093579,
      "loss": 5.2799,
      "mean_token_accuracy": 0.17788524329662322,
      "num_tokens": 17077323.0,
      "step": 9835
    },
    {
      "entropy": 5.529206657409668,
      "epoch": 0.7656098035401673,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004946864458617148,
      "loss": 5.169,
      "mean_token_accuracy": 0.18858975172042847,
      "num_tokens": 17085377.0,
      "step": 9840
    },
    {
      "entropy": 5.500270652770996,
      "epoch": 0.7659988329118849,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004946804564806793,
      "loss": 5.242,
      "mean_token_accuracy": 0.17937949299812317,
      "num_tokens": 17094288.0,
      "step": 9845
    },
    {
      "entropy": 5.571313285827637,
      "epoch": 0.7663878622836025,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004946744637663427,
      "loss": 5.1776,
      "mean_token_accuracy": 0.18991186916828157,
      "num_tokens": 17101572.0,
      "step": 9850
    },
    {
      "entropy": 5.549397754669189,
      "epoch": 0.7667768916553199,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004946684677187956,
      "loss": 5.315,
      "mean_token_accuracy": 0.1745953306555748,
      "num_tokens": 17110151.0,
      "step": 9855
    },
    {
      "entropy": 5.60110821723938,
      "epoch": 0.7671659210270375,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004946624683381292,
      "loss": 5.3045,
      "mean_token_accuracy": 0.17201686799526214,
      "num_tokens": 17118172.0,
      "step": 9860
    },
    {
      "entropy": 5.522176218032837,
      "epoch": 0.7675549503987551,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004946564656244345,
      "loss": 5.1812,
      "mean_token_accuracy": 0.18621729910373688,
      "num_tokens": 17126855.0,
      "step": 9865
    },
    {
      "entropy": 5.602323389053344,
      "epoch": 0.7679439797704727,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004946504595778026,
      "loss": 5.3748,
      "mean_token_accuracy": 0.17457078397274017,
      "num_tokens": 17135737.0,
      "step": 9870
    },
    {
      "entropy": 5.64451322555542,
      "epoch": 0.7683330091421903,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004946444501983246,
      "loss": 5.3098,
      "mean_token_accuracy": 0.17592848539352418,
      "num_tokens": 17144521.0,
      "step": 9875
    },
    {
      "entropy": 5.6431746006011965,
      "epoch": 0.7687220385139077,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004946384374860916,
      "loss": 5.3946,
      "mean_token_accuracy": 0.1757023736834526,
      "num_tokens": 17153534.0,
      "step": 9880
    },
    {
      "entropy": 5.52467942237854,
      "epoch": 0.7691110678856253,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004946324214411949,
      "loss": 5.2258,
      "mean_token_accuracy": 0.1848434701561928,
      "num_tokens": 17162600.0,
      "step": 9885
    },
    {
      "entropy": 5.569587516784668,
      "epoch": 0.7695000972573429,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004946264020637259,
      "loss": 5.3193,
      "mean_token_accuracy": 0.17557533234357833,
      "num_tokens": 17171373.0,
      "step": 9890
    },
    {
      "entropy": 5.632002305984497,
      "epoch": 0.7698891266290605,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004946203793537757,
      "loss": 5.3351,
      "mean_token_accuracy": 0.17728708386421205,
      "num_tokens": 17180090.0,
      "step": 9895
    },
    {
      "entropy": 5.60519757270813,
      "epoch": 0.7702781560007781,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004946143533114358,
      "loss": 5.2577,
      "mean_token_accuracy": 0.1822723478078842,
      "num_tokens": 17189592.0,
      "step": 9900
    },
    {
      "entropy": 5.519287204742431,
      "epoch": 0.7706671853724957,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004946083239367976,
      "loss": 5.1972,
      "mean_token_accuracy": 0.18773100078105925,
      "num_tokens": 17198287.0,
      "step": 9905
    },
    {
      "entropy": 5.593173360824585,
      "epoch": 0.7710562147442132,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004946022912299527,
      "loss": 5.3013,
      "mean_token_accuracy": 0.17522221207618713,
      "num_tokens": 17206349.0,
      "step": 9910
    },
    {
      "entropy": 5.614915943145752,
      "epoch": 0.7714452441159307,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004945962551909926,
      "loss": 5.4031,
      "mean_token_accuracy": 0.1742765948176384,
      "num_tokens": 17215408.0,
      "step": 9915
    },
    {
      "entropy": 5.684574699401855,
      "epoch": 0.7718342734876483,
      "grad_norm": 1.0,
      "learning_rate": 0.0004945902158200089,
      "loss": 5.3549,
      "mean_token_accuracy": 0.17250179797410964,
      "num_tokens": 17224342.0,
      "step": 9920
    },
    {
      "entropy": 5.59126558303833,
      "epoch": 0.7722233028593659,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004945841731170931,
      "loss": 5.2355,
      "mean_token_accuracy": 0.1765318751335144,
      "num_tokens": 17232532.0,
      "step": 9925
    },
    {
      "entropy": 5.51690354347229,
      "epoch": 0.7726123322310835,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004945781270823369,
      "loss": 5.2437,
      "mean_token_accuracy": 0.18184578269720078,
      "num_tokens": 17240786.0,
      "step": 9930
    },
    {
      "entropy": 5.513026142120362,
      "epoch": 0.773001361602801,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004945720777158323,
      "loss": 5.2295,
      "mean_token_accuracy": 0.18555503934621811,
      "num_tokens": 17248515.0,
      "step": 9935
    },
    {
      "entropy": 5.591573333740234,
      "epoch": 0.7733903909745186,
      "grad_norm": 1.0,
      "learning_rate": 0.0004945660250176708,
      "loss": 5.276,
      "mean_token_accuracy": 0.18310922980308533,
      "num_tokens": 17257098.0,
      "step": 9940
    },
    {
      "entropy": 5.561489295959473,
      "epoch": 0.7737794203462361,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004945599689879443,
      "loss": 5.2754,
      "mean_token_accuracy": 0.17605521529912949,
      "num_tokens": 17266851.0,
      "step": 9945
    },
    {
      "entropy": 5.5478973388671875,
      "epoch": 0.7741684497179537,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004945539096267448,
      "loss": 5.2233,
      "mean_token_accuracy": 0.17532940953969955,
      "num_tokens": 17275568.0,
      "step": 9950
    },
    {
      "entropy": 5.525739049911499,
      "epoch": 0.7745574790896713,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004945478469341642,
      "loss": 5.1988,
      "mean_token_accuracy": 0.18495745062828065,
      "num_tokens": 17284528.0,
      "step": 9955
    },
    {
      "entropy": 5.606883478164673,
      "epoch": 0.7749465084613888,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004945417809102944,
      "loss": 5.3396,
      "mean_token_accuracy": 0.1755354180932045,
      "num_tokens": 17293973.0,
      "step": 9960
    },
    {
      "entropy": 5.524585103988647,
      "epoch": 0.7753355378331064,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004945357115552275,
      "loss": 5.2319,
      "mean_token_accuracy": 0.17984241396188735,
      "num_tokens": 17302321.0,
      "step": 9965
    },
    {
      "entropy": 5.580976581573486,
      "epoch": 0.775724567204824,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945296388690557,
      "loss": 5.3504,
      "mean_token_accuracy": 0.17670010775327682,
      "num_tokens": 17311226.0,
      "step": 9970
    },
    {
      "entropy": 5.609677982330322,
      "epoch": 0.7761135965765416,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000494523562851871,
      "loss": 5.3277,
      "mean_token_accuracy": 0.17759332209825515,
      "num_tokens": 17319818.0,
      "step": 9975
    },
    {
      "entropy": 5.6251226425170895,
      "epoch": 0.7765026259482591,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945174835037655,
      "loss": 5.3474,
      "mean_token_accuracy": 0.17106997072696686,
      "num_tokens": 17329195.0,
      "step": 9980
    },
    {
      "entropy": 5.650672054290771,
      "epoch": 0.7768916553199766,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004945114008248319,
      "loss": 5.4327,
      "mean_token_accuracy": 0.17871665358543395,
      "num_tokens": 17337689.0,
      "step": 9985
    },
    {
      "entropy": 5.582010173797608,
      "epoch": 0.7772806846916942,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004945053148151619,
      "loss": 5.1884,
      "mean_token_accuracy": 0.18419857025146485,
      "num_tokens": 17345467.0,
      "step": 9990
    },
    {
      "entropy": 5.548644399642944,
      "epoch": 0.7776697140634118,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004944992254748482,
      "loss": 5.3342,
      "mean_token_accuracy": 0.1798410654067993,
      "num_tokens": 17353651.0,
      "step": 9995
    },
    {
      "entropy": 5.608665990829468,
      "epoch": 0.7780587434351294,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004944931328039832,
      "loss": 5.3904,
      "mean_token_accuracy": 0.17302385866641998,
      "num_tokens": 17362777.0,
      "step": 10000
    },
    {
      "entropy": 5.664591360092163,
      "epoch": 0.778447772806847,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004944870368026592,
      "loss": 5.3237,
      "mean_token_accuracy": 0.17910516113042832,
      "num_tokens": 17371562.0,
      "step": 10005
    },
    {
      "entropy": 5.633160924911499,
      "epoch": 0.7788368021785644,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004944809374709689,
      "loss": 5.3136,
      "mean_token_accuracy": 0.17074411660432814,
      "num_tokens": 17381185.0,
      "step": 10010
    },
    {
      "entropy": 5.597475528717041,
      "epoch": 0.779225831550282,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004944748348090045,
      "loss": 5.279,
      "mean_token_accuracy": 0.18732508569955825,
      "num_tokens": 17390658.0,
      "step": 10015
    },
    {
      "entropy": 5.5473918437957765,
      "epoch": 0.7796148609219996,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004944687288168589,
      "loss": 5.2168,
      "mean_token_accuracy": 0.17947004586458207,
      "num_tokens": 17399968.0,
      "step": 10020
    },
    {
      "entropy": 5.587859106063843,
      "epoch": 0.7800038902937172,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004944626194946246,
      "loss": 5.3441,
      "mean_token_accuracy": 0.1738036096096039,
      "num_tokens": 17408727.0,
      "step": 10025
    },
    {
      "entropy": 5.583702516555786,
      "epoch": 0.7803929196654348,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004944565068423945,
      "loss": 5.3554,
      "mean_token_accuracy": 0.17960301786661148,
      "num_tokens": 17417517.0,
      "step": 10030
    },
    {
      "entropy": 5.6376337051391605,
      "epoch": 0.7807819490371523,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004944503908602612,
      "loss": 5.3434,
      "mean_token_accuracy": 0.17666764408349991,
      "num_tokens": 17427035.0,
      "step": 10035
    },
    {
      "entropy": 5.641467332839966,
      "epoch": 0.7811709784088698,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004944442715483174,
      "loss": 5.248,
      "mean_token_accuracy": 0.1828450620174408,
      "num_tokens": 17435585.0,
      "step": 10040
    },
    {
      "entropy": 5.551241016387939,
      "epoch": 0.7815600077805874,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004944381489066561,
      "loss": 5.2636,
      "mean_token_accuracy": 0.18091956228017808,
      "num_tokens": 17444317.0,
      "step": 10045
    },
    {
      "entropy": 5.620751953125,
      "epoch": 0.781949037152305,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004944320229353702,
      "loss": 5.4727,
      "mean_token_accuracy": 0.17515989094972612,
      "num_tokens": 17453585.0,
      "step": 10050
    },
    {
      "entropy": 5.586199188232422,
      "epoch": 0.7823380665240226,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004944258936345526,
      "loss": 5.2588,
      "mean_token_accuracy": 0.18479428440332413,
      "num_tokens": 17462233.0,
      "step": 10055
    },
    {
      "entropy": 5.549157810211182,
      "epoch": 0.7827270958957401,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004944197610042963,
      "loss": 5.1583,
      "mean_token_accuracy": 0.1919889599084854,
      "num_tokens": 17470566.0,
      "step": 10060
    },
    {
      "entropy": 5.556268930435181,
      "epoch": 0.7831161252674577,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004944136250446944,
      "loss": 5.2804,
      "mean_token_accuracy": 0.1742836907505989,
      "num_tokens": 17479267.0,
      "step": 10065
    },
    {
      "entropy": 5.4754805088043215,
      "epoch": 0.7835051546391752,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004944074857558399,
      "loss": 5.1722,
      "mean_token_accuracy": 0.18835856914520263,
      "num_tokens": 17488060.0,
      "step": 10070
    },
    {
      "entropy": 5.676930952072143,
      "epoch": 0.7838941840108928,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004944013431378261,
      "loss": 5.4177,
      "mean_token_accuracy": 0.17947075963020326,
      "num_tokens": 17497240.0,
      "step": 10075
    },
    {
      "entropy": 5.641529941558838,
      "epoch": 0.7842832133826104,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004943951971907461,
      "loss": 5.3746,
      "mean_token_accuracy": 0.17811400294303895,
      "num_tokens": 17506017.0,
      "step": 10080
    },
    {
      "entropy": 5.625462532043457,
      "epoch": 0.7846722427543279,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004943890479146933,
      "loss": 5.3175,
      "mean_token_accuracy": 0.178188057243824,
      "num_tokens": 17514764.0,
      "step": 10085
    },
    {
      "entropy": 5.547196388244629,
      "epoch": 0.7850612721260455,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004943828953097607,
      "loss": 5.2942,
      "mean_token_accuracy": 0.1766573965549469,
      "num_tokens": 17524183.0,
      "step": 10090
    },
    {
      "entropy": 5.552852392196655,
      "epoch": 0.7854503014977631,
      "grad_norm": 0.921875,
      "learning_rate": 0.000494376739376042,
      "loss": 5.241,
      "mean_token_accuracy": 0.18242752254009248,
      "num_tokens": 17533426.0,
      "step": 10095
    },
    {
      "entropy": 5.61579270362854,
      "epoch": 0.7858393308694807,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004943705801136304,
      "loss": 5.1873,
      "mean_token_accuracy": 0.1837708830833435,
      "num_tokens": 17541966.0,
      "step": 10100
    },
    {
      "entropy": 5.587601232528686,
      "epoch": 0.7862283602411982,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004943644175226193,
      "loss": 5.2882,
      "mean_token_accuracy": 0.1829829156398773,
      "num_tokens": 17551157.0,
      "step": 10105
    },
    {
      "entropy": 5.52039270401001,
      "epoch": 0.7866173896129158,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004943582516031025,
      "loss": 5.2057,
      "mean_token_accuracy": 0.1859710082411766,
      "num_tokens": 17559999.0,
      "step": 10110
    },
    {
      "entropy": 5.581799268722534,
      "epoch": 0.7870064189846333,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004943520823551732,
      "loss": 5.2531,
      "mean_token_accuracy": 0.18347948640584946,
      "num_tokens": 17568873.0,
      "step": 10115
    },
    {
      "entropy": 5.5961323261260985,
      "epoch": 0.7873954483563509,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004943459097789252,
      "loss": 5.328,
      "mean_token_accuracy": 0.1859396740794182,
      "num_tokens": 17577807.0,
      "step": 10120
    },
    {
      "entropy": 5.599230861663818,
      "epoch": 0.7877844777280685,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004943397338744523,
      "loss": 5.3442,
      "mean_token_accuracy": 0.17688045501708985,
      "num_tokens": 17586646.0,
      "step": 10125
    },
    {
      "entropy": 5.650975227355957,
      "epoch": 0.7881735070997861,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000494333554641848,
      "loss": 5.406,
      "mean_token_accuracy": 0.17036691009998323,
      "num_tokens": 17595458.0,
      "step": 10130
    },
    {
      "entropy": 5.611591577529907,
      "epoch": 0.7885625364715037,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000494327372081206,
      "loss": 5.3205,
      "mean_token_accuracy": 0.178653784096241,
      "num_tokens": 17603997.0,
      "step": 10135
    },
    {
      "entropy": 5.636513948440552,
      "epoch": 0.7889515658432211,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004943211861926204,
      "loss": 5.3409,
      "mean_token_accuracy": 0.17546274811029433,
      "num_tokens": 17612347.0,
      "step": 10140
    },
    {
      "entropy": 5.563936233520508,
      "epoch": 0.7893405952149387,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004943149969761848,
      "loss": 5.1964,
      "mean_token_accuracy": 0.1862559735774994,
      "num_tokens": 17620618.0,
      "step": 10145
    },
    {
      "entropy": 5.495310258865357,
      "epoch": 0.7897296245866563,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004943088044319932,
      "loss": 5.2127,
      "mean_token_accuracy": 0.18083256781101226,
      "num_tokens": 17629164.0,
      "step": 10150
    },
    {
      "entropy": 5.642545223236084,
      "epoch": 0.7901186539583739,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004943026085601395,
      "loss": 5.388,
      "mean_token_accuracy": 0.17380305081605912,
      "num_tokens": 17638454.0,
      "step": 10155
    },
    {
      "entropy": 5.58918867111206,
      "epoch": 0.7905076833300915,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004942964093607178,
      "loss": 5.3049,
      "mean_token_accuracy": 0.17752310484647751,
      "num_tokens": 17647099.0,
      "step": 10160
    },
    {
      "entropy": 5.594116115570069,
      "epoch": 0.790896712701809,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004942902068338222,
      "loss": 5.2704,
      "mean_token_accuracy": 0.18279317766427994,
      "num_tokens": 17655214.0,
      "step": 10165
    },
    {
      "entropy": 5.562254953384399,
      "epoch": 0.7912857420735265,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004942840009795466,
      "loss": 5.236,
      "mean_token_accuracy": 0.1806568518280983,
      "num_tokens": 17663342.0,
      "step": 10170
    },
    {
      "entropy": 5.704637908935547,
      "epoch": 0.7916747714452441,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942777917979855,
      "loss": 5.4433,
      "mean_token_accuracy": 0.16575012654066085,
      "num_tokens": 17671725.0,
      "step": 10175
    },
    {
      "entropy": 5.567132616043091,
      "epoch": 0.7920638008169617,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004942715792892328,
      "loss": 5.3392,
      "mean_token_accuracy": 0.1732524037361145,
      "num_tokens": 17680627.0,
      "step": 10180
    },
    {
      "entropy": 5.618295097351075,
      "epoch": 0.7924528301886793,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000494265363453383,
      "loss": 5.2998,
      "mean_token_accuracy": 0.18326802551746368,
      "num_tokens": 17688392.0,
      "step": 10185
    },
    {
      "entropy": 5.50335283279419,
      "epoch": 0.7928418595603968,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942591442905302,
      "loss": 5.2558,
      "mean_token_accuracy": 0.18379608690738677,
      "num_tokens": 17697546.0,
      "step": 10190
    },
    {
      "entropy": 5.508504199981689,
      "epoch": 0.7932308889321144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004942529218007689,
      "loss": 5.2165,
      "mean_token_accuracy": 0.18102292269468306,
      "num_tokens": 17705555.0,
      "step": 10195
    },
    {
      "entropy": 5.581924962997436,
      "epoch": 0.7936199183038319,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942466959841936,
      "loss": 5.3836,
      "mean_token_accuracy": 0.16984927654266357,
      "num_tokens": 17715652.0,
      "step": 10200
    },
    {
      "entropy": 5.512160062789917,
      "epoch": 0.7940089476755495,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004942404668408985,
      "loss": 5.124,
      "mean_token_accuracy": 0.18311595916748047,
      "num_tokens": 17724111.0,
      "step": 10205
    },
    {
      "entropy": 5.461303615570069,
      "epoch": 0.7943979770472671,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004942342343709783,
      "loss": 5.1396,
      "mean_token_accuracy": 0.18451044708490372,
      "num_tokens": 17732732.0,
      "step": 10210
    },
    {
      "entropy": 5.558248853683471,
      "epoch": 0.7947870064189846,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004942279985745275,
      "loss": 5.2399,
      "mean_token_accuracy": 0.18044923096895218,
      "num_tokens": 17741338.0,
      "step": 10215
    },
    {
      "entropy": 5.6359869003295895,
      "epoch": 0.7951760357907022,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942217594516409,
      "loss": 5.4489,
      "mean_token_accuracy": 0.1734004020690918,
      "num_tokens": 17750651.0,
      "step": 10220
    },
    {
      "entropy": 5.604165887832641,
      "epoch": 0.7955650651624198,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000494215517002413,
      "loss": 5.2573,
      "mean_token_accuracy": 0.18303716629743577,
      "num_tokens": 17759052.0,
      "step": 10225
    },
    {
      "entropy": 5.621202278137207,
      "epoch": 0.7959540945341373,
      "grad_norm": 1.125,
      "learning_rate": 0.0004942092712269384,
      "loss": 5.3162,
      "mean_token_accuracy": 0.1753840208053589,
      "num_tokens": 17767246.0,
      "step": 10230
    },
    {
      "entropy": 5.646518230438232,
      "epoch": 0.7963431239058549,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942030221253122,
      "loss": 5.3416,
      "mean_token_accuracy": 0.17957145124673843,
      "num_tokens": 17777022.0,
      "step": 10235
    },
    {
      "entropy": 5.57313551902771,
      "epoch": 0.7967321532775724,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004941967696976289,
      "loss": 5.3091,
      "mean_token_accuracy": 0.17263340651988984,
      "num_tokens": 17785586.0,
      "step": 10240
    },
    {
      "entropy": 5.538752698898316,
      "epoch": 0.79712118264929,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004941905139439835,
      "loss": 5.3038,
      "mean_token_accuracy": 0.17859912812709808,
      "num_tokens": 17794320.0,
      "step": 10245
    },
    {
      "entropy": 5.632735300064087,
      "epoch": 0.7975102120210076,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004941842548644712,
      "loss": 5.3097,
      "mean_token_accuracy": 0.18384564220905303,
      "num_tokens": 17803384.0,
      "step": 10250
    },
    {
      "entropy": 5.605423831939698,
      "epoch": 0.7978992413927252,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004941779924591864,
      "loss": 5.2807,
      "mean_token_accuracy": 0.17626046687364577,
      "num_tokens": 17811955.0,
      "step": 10255
    },
    {
      "entropy": 5.567848348617554,
      "epoch": 0.7982882707644428,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004941717267282244,
      "loss": 5.3242,
      "mean_token_accuracy": 0.17602663338184357,
      "num_tokens": 17820582.0,
      "step": 10260
    },
    {
      "entropy": 5.643657493591308,
      "epoch": 0.7986773001361602,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004941654576716806,
      "loss": 5.245,
      "mean_token_accuracy": 0.18023714274168015,
      "num_tokens": 17828465.0,
      "step": 10265
    },
    {
      "entropy": 5.5508965969085695,
      "epoch": 0.7990663295078778,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004941591852896495,
      "loss": 5.2445,
      "mean_token_accuracy": 0.18421650677919388,
      "num_tokens": 17836931.0,
      "step": 10270
    },
    {
      "entropy": 5.572287082672119,
      "epoch": 0.7994553588795954,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941529095822268,
      "loss": 5.3278,
      "mean_token_accuracy": 0.17869895994663237,
      "num_tokens": 17845192.0,
      "step": 10275
    },
    {
      "entropy": 5.554533910751343,
      "epoch": 0.799844388251313,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004941466305495074,
      "loss": 5.3059,
      "mean_token_accuracy": 0.17677113115787507,
      "num_tokens": 17854121.0,
      "step": 10280
    },
    {
      "entropy": 5.603392553329468,
      "epoch": 0.8002334176230306,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941403481915867,
      "loss": 5.2917,
      "mean_token_accuracy": 0.17675098776817322,
      "num_tokens": 17862931.0,
      "step": 10285
    },
    {
      "entropy": 5.649144268035888,
      "epoch": 0.800622446994748,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004941340625085601,
      "loss": 5.3176,
      "mean_token_accuracy": 0.17665561735630037,
      "num_tokens": 17871909.0,
      "step": 10290
    },
    {
      "entropy": 5.547114706039428,
      "epoch": 0.8010114763664656,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004941277735005228,
      "loss": 5.231,
      "mean_token_accuracy": 0.18726652562618257,
      "num_tokens": 17880487.0,
      "step": 10295
    },
    {
      "entropy": 5.544336891174316,
      "epoch": 0.8014005057381832,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004941214811675702,
      "loss": 5.301,
      "mean_token_accuracy": 0.17943531274795532,
      "num_tokens": 17888712.0,
      "step": 10300
    },
    {
      "entropy": 5.628103160858155,
      "epoch": 0.8017895351099008,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941151855097982,
      "loss": 5.3364,
      "mean_token_accuracy": 0.16981505006551742,
      "num_tokens": 17898163.0,
      "step": 10305
    },
    {
      "entropy": 5.551942348480225,
      "epoch": 0.8021785644816184,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004941088865273018,
      "loss": 5.227,
      "mean_token_accuracy": 0.18352603912353516,
      "num_tokens": 17906849.0,
      "step": 10310
    },
    {
      "entropy": 5.574260997772217,
      "epoch": 0.802567593853336,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004941025842201769,
      "loss": 5.2549,
      "mean_token_accuracy": 0.1780119240283966,
      "num_tokens": 17915695.0,
      "step": 10315
    },
    {
      "entropy": 5.64654860496521,
      "epoch": 0.8029566232250535,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004940962785885189,
      "loss": 5.3543,
      "mean_token_accuracy": 0.17598017901182175,
      "num_tokens": 17924706.0,
      "step": 10320
    },
    {
      "entropy": 5.561926555633545,
      "epoch": 0.803345652596771,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004940899696324237,
      "loss": 5.3287,
      "mean_token_accuracy": 0.18102977126836778,
      "num_tokens": 17934061.0,
      "step": 10325
    },
    {
      "entropy": 5.536122751235962,
      "epoch": 0.8037346819684886,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004940836573519868,
      "loss": 5.2054,
      "mean_token_accuracy": 0.17839139550924302,
      "num_tokens": 17942580.0,
      "step": 10330
    },
    {
      "entropy": 5.525859594345093,
      "epoch": 0.8041237113402062,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004940773417473043,
      "loss": 5.1004,
      "mean_token_accuracy": 0.18802183270454406,
      "num_tokens": 17951171.0,
      "step": 10335
    },
    {
      "entropy": 5.512113428115844,
      "epoch": 0.8045127407119238,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004940710228184717,
      "loss": 5.1789,
      "mean_token_accuracy": 0.19142742902040483,
      "num_tokens": 17959674.0,
      "step": 10340
    },
    {
      "entropy": 5.538492822647095,
      "epoch": 0.8049017700836413,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000494064700565585,
      "loss": 5.2434,
      "mean_token_accuracy": 0.18216741681098939,
      "num_tokens": 17969036.0,
      "step": 10345
    },
    {
      "entropy": 5.547858619689942,
      "epoch": 0.8052907994553589,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004940583749887403,
      "loss": 5.2602,
      "mean_token_accuracy": 0.18307285904884338,
      "num_tokens": 17977406.0,
      "step": 10350
    },
    {
      "entropy": 5.557126045227051,
      "epoch": 0.8056798288270764,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004940520460880333,
      "loss": 5.2322,
      "mean_token_accuracy": 0.18356891870498657,
      "num_tokens": 17985981.0,
      "step": 10355
    },
    {
      "entropy": 5.637621688842773,
      "epoch": 0.806068858198794,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004940457138635601,
      "loss": 5.4217,
      "mean_token_accuracy": 0.17267650961875916,
      "num_tokens": 17994246.0,
      "step": 10360
    },
    {
      "entropy": 5.581584882736206,
      "epoch": 0.8064578875705116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004940393783154169,
      "loss": 5.2593,
      "mean_token_accuracy": 0.17970990985631943,
      "num_tokens": 18002444.0,
      "step": 10365
    },
    {
      "entropy": 5.532887077331543,
      "epoch": 0.8068469169422291,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004940330394436999,
      "loss": 5.1834,
      "mean_token_accuracy": 0.18363454490900039,
      "num_tokens": 18011076.0,
      "step": 10370
    },
    {
      "entropy": 5.64114203453064,
      "epoch": 0.8072359463139467,
      "grad_norm": 0.96875,
      "learning_rate": 0.000494026697248505,
      "loss": 5.4665,
      "mean_token_accuracy": 0.1663346603512764,
      "num_tokens": 18020232.0,
      "step": 10375
    },
    {
      "entropy": 5.636241626739502,
      "epoch": 0.8076249756856643,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004940203517299287,
      "loss": 5.2743,
      "mean_token_accuracy": 0.1763804465532303,
      "num_tokens": 18029055.0,
      "step": 10380
    },
    {
      "entropy": 5.642540788650512,
      "epoch": 0.8080140050573819,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000494014002888067,
      "loss": 5.4172,
      "mean_token_accuracy": 0.1637917563319206,
      "num_tokens": 18037827.0,
      "step": 10385
    },
    {
      "entropy": 5.536993980407715,
      "epoch": 0.8084030344290994,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004940076507230166,
      "loss": 5.2474,
      "mean_token_accuracy": 0.17653342634439467,
      "num_tokens": 18047017.0,
      "step": 10390
    },
    {
      "entropy": 5.619476079940796,
      "epoch": 0.8087920638008169,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004940012952348735,
      "loss": 5.354,
      "mean_token_accuracy": 0.17621037364006042,
      "num_tokens": 18056256.0,
      "step": 10395
    },
    {
      "entropy": 5.568868350982666,
      "epoch": 0.8091810931725345,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004939949364237345,
      "loss": 5.1129,
      "mean_token_accuracy": 0.19875151962041854,
      "num_tokens": 18064127.0,
      "step": 10400
    },
    {
      "entropy": 5.507281875610351,
      "epoch": 0.8095701225442521,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004939885742896958,
      "loss": 5.2757,
      "mean_token_accuracy": 0.1827508956193924,
      "num_tokens": 18072766.0,
      "step": 10405
    },
    {
      "entropy": 5.5320202827453615,
      "epoch": 0.8099591519159697,
      "grad_norm": 1.046875,
      "learning_rate": 0.000493982208832854,
      "loss": 5.2401,
      "mean_token_accuracy": 0.17939431816339493,
      "num_tokens": 18080685.0,
      "step": 10410
    },
    {
      "entropy": 5.613889408111572,
      "epoch": 0.8103481812876873,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939758400533058,
      "loss": 5.3109,
      "mean_token_accuracy": 0.1760113373398781,
      "num_tokens": 18090026.0,
      "step": 10415
    },
    {
      "entropy": 5.591238641738892,
      "epoch": 0.8107372106594047,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004939694679511478,
      "loss": 5.2274,
      "mean_token_accuracy": 0.18148638010025026,
      "num_tokens": 18098638.0,
      "step": 10420
    },
    {
      "entropy": 5.526708889007568,
      "epoch": 0.8111262400311223,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004939630925264765,
      "loss": 5.2639,
      "mean_token_accuracy": 0.18288784623146057,
      "num_tokens": 18106844.0,
      "step": 10425
    },
    {
      "entropy": 5.581420755386352,
      "epoch": 0.8115152694028399,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493956713779389,
      "loss": 5.2536,
      "mean_token_accuracy": 0.17701498419046402,
      "num_tokens": 18114758.0,
      "step": 10430
    },
    {
      "entropy": 5.5890758514404295,
      "epoch": 0.8119042987745575,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004939503317099817,
      "loss": 5.2254,
      "mean_token_accuracy": 0.18185850977897644,
      "num_tokens": 18122839.0,
      "step": 10435
    },
    {
      "entropy": 5.5715070247650145,
      "epoch": 0.8122933281462751,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004939439463183517,
      "loss": 5.3198,
      "mean_token_accuracy": 0.18264523893594742,
      "num_tokens": 18131709.0,
      "step": 10440
    },
    {
      "entropy": 5.580619525909424,
      "epoch": 0.8126823575179926,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004939375576045958,
      "loss": 5.2965,
      "mean_token_accuracy": 0.17366210371255875,
      "num_tokens": 18140334.0,
      "step": 10445
    },
    {
      "entropy": 5.531055545806884,
      "epoch": 0.8130713868897101,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004939311655688109,
      "loss": 5.293,
      "mean_token_accuracy": 0.1767920583486557,
      "num_tokens": 18148648.0,
      "step": 10450
    },
    {
      "entropy": 5.560915088653564,
      "epoch": 0.8134604162614277,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004939247702110941,
      "loss": 5.296,
      "mean_token_accuracy": 0.18004312962293625,
      "num_tokens": 18156891.0,
      "step": 10455
    },
    {
      "entropy": 5.569944620132446,
      "epoch": 0.8138494456331453,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004939183715315423,
      "loss": 5.2408,
      "mean_token_accuracy": 0.18391807228326798,
      "num_tokens": 18166512.0,
      "step": 10460
    },
    {
      "entropy": 5.591281080245972,
      "epoch": 0.8142384750048629,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004939119695302526,
      "loss": 5.1994,
      "mean_token_accuracy": 0.18591028600931167,
      "num_tokens": 18175064.0,
      "step": 10465
    },
    {
      "entropy": 5.552697563171387,
      "epoch": 0.8146275043765804,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004939055642073224,
      "loss": 5.3331,
      "mean_token_accuracy": 0.17966595739126207,
      "num_tokens": 18184251.0,
      "step": 10470
    },
    {
      "entropy": 5.516888427734375,
      "epoch": 0.815016533748298,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004938991555628484,
      "loss": 5.167,
      "mean_token_accuracy": 0.18957981765270232,
      "num_tokens": 18193274.0,
      "step": 10475
    },
    {
      "entropy": 5.532575511932373,
      "epoch": 0.8154055631200156,
      "grad_norm": 1.0,
      "learning_rate": 0.0004938927435969283,
      "loss": 5.2303,
      "mean_token_accuracy": 0.18930566757917405,
      "num_tokens": 18201878.0,
      "step": 10480
    },
    {
      "entropy": 5.51650972366333,
      "epoch": 0.8157945924917331,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004938863283096592,
      "loss": 5.1984,
      "mean_token_accuracy": 0.18496177345514297,
      "num_tokens": 18210343.0,
      "step": 10485
    },
    {
      "entropy": 5.641515016555786,
      "epoch": 0.8161836218634507,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004938799097011384,
      "loss": 5.3281,
      "mean_token_accuracy": 0.18384619504213334,
      "num_tokens": 18218599.0,
      "step": 10490
    },
    {
      "entropy": 5.550069665908813,
      "epoch": 0.8165726512351682,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004938734877714634,
      "loss": 5.2091,
      "mean_token_accuracy": 0.18761346638202667,
      "num_tokens": 18226384.0,
      "step": 10495
    },
    {
      "entropy": 5.519427156448364,
      "epoch": 0.8169616806068858,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004938670625207317,
      "loss": 5.2014,
      "mean_token_accuracy": 0.1802952140569687,
      "num_tokens": 18234894.0,
      "step": 10500
    },
    {
      "entropy": 5.565952444076538,
      "epoch": 0.8173507099786034,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004938606339490406,
      "loss": 5.2414,
      "mean_token_accuracy": 0.18535373508930206,
      "num_tokens": 18243768.0,
      "step": 10505
    },
    {
      "entropy": 5.584456729888916,
      "epoch": 0.817739739350321,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004938542020564877,
      "loss": 5.2782,
      "mean_token_accuracy": 0.18248629719018936,
      "num_tokens": 18252364.0,
      "step": 10510
    },
    {
      "entropy": 5.53111891746521,
      "epoch": 0.8181287687220385,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004938477668431706,
      "loss": 5.2278,
      "mean_token_accuracy": 0.18297127038240432,
      "num_tokens": 18260997.0,
      "step": 10515
    },
    {
      "entropy": 5.672461938858032,
      "epoch": 0.8185177980937561,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004938413283091871,
      "loss": 5.413,
      "mean_token_accuracy": 0.17652716487646103,
      "num_tokens": 18270171.0,
      "step": 10520
    },
    {
      "entropy": 5.643474149703979,
      "epoch": 0.8189068274654736,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004938348864546347,
      "loss": 5.2559,
      "mean_token_accuracy": 0.18688322007656097,
      "num_tokens": 18278772.0,
      "step": 10525
    },
    {
      "entropy": 5.486976099014282,
      "epoch": 0.8192958568371912,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004938284412796111,
      "loss": 5.0895,
      "mean_token_accuracy": 0.18835799694061278,
      "num_tokens": 18287305.0,
      "step": 10530
    },
    {
      "entropy": 5.470224809646607,
      "epoch": 0.8196848862089088,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004938219927842144,
      "loss": 5.3277,
      "mean_token_accuracy": 0.1742580935359001,
      "num_tokens": 18296532.0,
      "step": 10535
    },
    {
      "entropy": 5.55927472114563,
      "epoch": 0.8200739155806264,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004938155409685422,
      "loss": 5.1859,
      "mean_token_accuracy": 0.18895601928234101,
      "num_tokens": 18304612.0,
      "step": 10540
    },
    {
      "entropy": 5.555217218399048,
      "epoch": 0.820462944952344,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004938090858326923,
      "loss": 5.2545,
      "mean_token_accuracy": 0.18999724090099335,
      "num_tokens": 18312970.0,
      "step": 10545
    },
    {
      "entropy": 5.578555965423584,
      "epoch": 0.8208519743240614,
      "grad_norm": 1.078125,
      "learning_rate": 0.000493802627376763,
      "loss": 5.2793,
      "mean_token_accuracy": 0.181522336602211,
      "num_tokens": 18321073.0,
      "step": 10550
    },
    {
      "entropy": 5.5834558486938475,
      "epoch": 0.821241003695779,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004937961656008518,
      "loss": 5.2667,
      "mean_token_accuracy": 0.18040078282356262,
      "num_tokens": 18329960.0,
      "step": 10555
    },
    {
      "entropy": 5.514997577667236,
      "epoch": 0.8216300330674966,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937897005050573,
      "loss": 5.2308,
      "mean_token_accuracy": 0.18348744064569472,
      "num_tokens": 18337935.0,
      "step": 10560
    },
    {
      "entropy": 5.541813945770263,
      "epoch": 0.8220190624392142,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004937832320894772,
      "loss": 5.2233,
      "mean_token_accuracy": 0.18325609117746353,
      "num_tokens": 18347445.0,
      "step": 10565
    },
    {
      "entropy": 5.586518907546997,
      "epoch": 0.8224080918109318,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004937767603542098,
      "loss": 5.2327,
      "mean_token_accuracy": 0.18263234943151474,
      "num_tokens": 18355817.0,
      "step": 10570
    },
    {
      "entropy": 5.571928024291992,
      "epoch": 0.8227971211826492,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004937702852993534,
      "loss": 5.2282,
      "mean_token_accuracy": 0.18538487404584886,
      "num_tokens": 18363478.0,
      "step": 10575
    },
    {
      "entropy": 5.531855154037475,
      "epoch": 0.8231861505543668,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000493763806925006,
      "loss": 5.2108,
      "mean_token_accuracy": 0.18181927502155304,
      "num_tokens": 18371788.0,
      "step": 10580
    },
    {
      "entropy": 5.53119592666626,
      "epoch": 0.8235751799260844,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004937573252312661,
      "loss": 5.2477,
      "mean_token_accuracy": 0.18195367455482483,
      "num_tokens": 18380361.0,
      "step": 10585
    },
    {
      "entropy": 5.575139045715332,
      "epoch": 0.823964209297802,
      "grad_norm": 1.015625,
      "learning_rate": 0.000493750840218232,
      "loss": 5.3032,
      "mean_token_accuracy": 0.18456340879201888,
      "num_tokens": 18389396.0,
      "step": 10590
    },
    {
      "entropy": 5.549172067642212,
      "epoch": 0.8243532386695196,
      "grad_norm": 1.0,
      "learning_rate": 0.0004937443518860021,
      "loss": 5.3634,
      "mean_token_accuracy": 0.16981100738048555,
      "num_tokens": 18398517.0,
      "step": 10595
    },
    {
      "entropy": 5.645396184921265,
      "epoch": 0.8247422680412371,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004937378602346747,
      "loss": 5.2937,
      "mean_token_accuracy": 0.18099551200866698,
      "num_tokens": 18406620.0,
      "step": 10600
    },
    {
      "entropy": 5.5864983081817625,
      "epoch": 0.8251312974129547,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004937313652643485,
      "loss": 5.3128,
      "mean_token_accuracy": 0.17442767173051835,
      "num_tokens": 18414849.0,
      "step": 10605
    },
    {
      "entropy": 5.521216821670532,
      "epoch": 0.8255203267846722,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000493724866975122,
      "loss": 5.2769,
      "mean_token_accuracy": 0.18221895694732665,
      "num_tokens": 18423675.0,
      "step": 10610
    },
    {
      "entropy": 5.66914439201355,
      "epoch": 0.8259093561563898,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004937183653670937,
      "loss": 5.3275,
      "mean_token_accuracy": 0.17973185926675797,
      "num_tokens": 18432848.0,
      "step": 10615
    },
    {
      "entropy": 5.662630844116211,
      "epoch": 0.8262983855281074,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004937118604403623,
      "loss": 5.3713,
      "mean_token_accuracy": 0.17663578391075135,
      "num_tokens": 18441210.0,
      "step": 10620
    },
    {
      "entropy": 5.59189510345459,
      "epoch": 0.8266874148998249,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004937053521950266,
      "loss": 5.2786,
      "mean_token_accuracy": 0.17604520916938782,
      "num_tokens": 18450084.0,
      "step": 10625
    },
    {
      "entropy": 5.486194515228272,
      "epoch": 0.8270764442715425,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004936988406311854,
      "loss": 5.1548,
      "mean_token_accuracy": 0.18131692260503768,
      "num_tokens": 18458329.0,
      "step": 10630
    },
    {
      "entropy": 5.525796365737915,
      "epoch": 0.8274654736432601,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004936923257489373,
      "loss": 5.2683,
      "mean_token_accuracy": 0.17761918902397156,
      "num_tokens": 18467518.0,
      "step": 10635
    },
    {
      "entropy": 5.520053434371948,
      "epoch": 0.8278545030149776,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004936858075483811,
      "loss": 5.1578,
      "mean_token_accuracy": 0.1974279910326004,
      "num_tokens": 18476051.0,
      "step": 10640
    },
    {
      "entropy": 5.584073686599732,
      "epoch": 0.8282435323866952,
      "grad_norm": 0.96875,
      "learning_rate": 0.000493679286029616,
      "loss": 5.2558,
      "mean_token_accuracy": 0.18164041340351106,
      "num_tokens": 18484492.0,
      "step": 10645
    },
    {
      "entropy": 5.581503963470459,
      "epoch": 0.8286325617584127,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004936727611927407,
      "loss": 5.3281,
      "mean_token_accuracy": 0.1709176108241081,
      "num_tokens": 18493759.0,
      "step": 10650
    },
    {
      "entropy": 5.606781721115112,
      "epoch": 0.8290215911301303,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004936662330378546,
      "loss": 5.1846,
      "mean_token_accuracy": 0.18201837837696075,
      "num_tokens": 18501750.0,
      "step": 10655
    },
    {
      "entropy": 5.490160417556763,
      "epoch": 0.8294106205018479,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004936597015650561,
      "loss": 5.1917,
      "mean_token_accuracy": 0.18296556621789933,
      "num_tokens": 18511003.0,
      "step": 10660
    },
    {
      "entropy": 5.513647127151489,
      "epoch": 0.8297996498735655,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004936531667744448,
      "loss": 5.2433,
      "mean_token_accuracy": 0.18767827302217482,
      "num_tokens": 18519544.0,
      "step": 10665
    },
    {
      "entropy": 5.686276054382324,
      "epoch": 0.8301886792452831,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004936466286661197,
      "loss": 5.3324,
      "mean_token_accuracy": 0.17959838956594468,
      "num_tokens": 18528766.0,
      "step": 10670
    },
    {
      "entropy": 5.601961088180542,
      "epoch": 0.8305777086170005,
      "grad_norm": 0.96875,
      "learning_rate": 0.00049364008724018,
      "loss": 5.3349,
      "mean_token_accuracy": 0.1756493031978607,
      "num_tokens": 18538243.0,
      "step": 10675
    },
    {
      "entropy": 5.592152118682861,
      "epoch": 0.8309667379887181,
      "grad_norm": 1.015625,
      "learning_rate": 0.000493633542496725,
      "loss": 5.3425,
      "mean_token_accuracy": 0.1781254753470421,
      "num_tokens": 18547028.0,
      "step": 10680
    },
    {
      "entropy": 5.618417549133301,
      "epoch": 0.8313557673604357,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004936269944358539,
      "loss": 5.3546,
      "mean_token_accuracy": 0.17658570408821106,
      "num_tokens": 18555821.0,
      "step": 10685
    },
    {
      "entropy": 5.542150449752808,
      "epoch": 0.8317447967321533,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004936204430576664,
      "loss": 5.322,
      "mean_token_accuracy": 0.17941897213459015,
      "num_tokens": 18564213.0,
      "step": 10690
    },
    {
      "entropy": 5.60770435333252,
      "epoch": 0.8321338261038709,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004936138883622614,
      "loss": 5.2899,
      "mean_token_accuracy": 0.17497958540916442,
      "num_tokens": 18573132.0,
      "step": 10695
    },
    {
      "entropy": 5.606252002716064,
      "epoch": 0.8325228554755884,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004936073303497387,
      "loss": 5.1981,
      "mean_token_accuracy": 0.18049157559871673,
      "num_tokens": 18581131.0,
      "step": 10700
    },
    {
      "entropy": 5.564637136459351,
      "epoch": 0.8329118848473059,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004936007690201976,
      "loss": 5.2805,
      "mean_token_accuracy": 0.1784028634428978,
      "num_tokens": 18589602.0,
      "step": 10705
    },
    {
      "entropy": 5.548305988311768,
      "epoch": 0.8333009142190235,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000493594204373738,
      "loss": 5.2647,
      "mean_token_accuracy": 0.1764623209834099,
      "num_tokens": 18598263.0,
      "step": 10710
    },
    {
      "entropy": 5.606611728668213,
      "epoch": 0.8336899435907411,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004935876364104591,
      "loss": 5.1725,
      "mean_token_accuracy": 0.189447121322155,
      "num_tokens": 18606432.0,
      "step": 10715
    },
    {
      "entropy": 5.573548603057861,
      "epoch": 0.8340789729624587,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004935810651304608,
      "loss": 5.3071,
      "mean_token_accuracy": 0.17638261318206788,
      "num_tokens": 18614769.0,
      "step": 10720
    },
    {
      "entropy": 5.607085561752319,
      "epoch": 0.8344680023341763,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004935744905338427,
      "loss": 5.271,
      "mean_token_accuracy": 0.17813154458999633,
      "num_tokens": 18623505.0,
      "step": 10725
    },
    {
      "entropy": 5.621091651916504,
      "epoch": 0.8348570317058938,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004935679126207046,
      "loss": 5.2392,
      "mean_token_accuracy": 0.19103155732154847,
      "num_tokens": 18632072.0,
      "step": 10730
    },
    {
      "entropy": 5.650973272323609,
      "epoch": 0.8352460610776113,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004935613313911463,
      "loss": 5.3168,
      "mean_token_accuracy": 0.17635270208120346,
      "num_tokens": 18640439.0,
      "step": 10735
    },
    {
      "entropy": 5.56662220954895,
      "epoch": 0.8356350904493289,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004935547468452678,
      "loss": 5.3091,
      "mean_token_accuracy": 0.16891513168811798,
      "num_tokens": 18649554.0,
      "step": 10740
    },
    {
      "entropy": 5.607131433486939,
      "epoch": 0.8360241198210465,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004935481589831688,
      "loss": 5.1742,
      "mean_token_accuracy": 0.1841895952820778,
      "num_tokens": 18657918.0,
      "step": 10745
    },
    {
      "entropy": 5.629995822906494,
      "epoch": 0.8364131491927641,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004935415678049492,
      "loss": 5.288,
      "mean_token_accuracy": 0.17434722632169725,
      "num_tokens": 18666735.0,
      "step": 10750
    },
    {
      "entropy": 5.569375228881836,
      "epoch": 0.8368021785644816,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004935349733107094,
      "loss": 5.1825,
      "mean_token_accuracy": 0.1853045254945755,
      "num_tokens": 18674977.0,
      "step": 10755
    },
    {
      "entropy": 5.5684432029724125,
      "epoch": 0.8371912079361992,
      "grad_norm": 1.03125,
      "learning_rate": 0.000493528375500549,
      "loss": 5.3288,
      "mean_token_accuracy": 0.17748333066701888,
      "num_tokens": 18683624.0,
      "step": 10760
    },
    {
      "entropy": 5.524935007095337,
      "epoch": 0.8375802373079168,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004935217743745685,
      "loss": 5.2199,
      "mean_token_accuracy": 0.18397823274135588,
      "num_tokens": 18692183.0,
      "step": 10765
    },
    {
      "entropy": 5.604568290710449,
      "epoch": 0.8379692666796343,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004935151699328677,
      "loss": 5.3085,
      "mean_token_accuracy": 0.17603420168161393,
      "num_tokens": 18700989.0,
      "step": 10770
    },
    {
      "entropy": 5.545925712585449,
      "epoch": 0.8383582960513519,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004935085621755471,
      "loss": 5.1773,
      "mean_token_accuracy": 0.18596030175685882,
      "num_tokens": 18709126.0,
      "step": 10775
    },
    {
      "entropy": 5.587519264221191,
      "epoch": 0.8387473254230694,
      "grad_norm": 1.03125,
      "learning_rate": 0.000493501951102707,
      "loss": 5.2801,
      "mean_token_accuracy": 0.17877295315265657,
      "num_tokens": 18717378.0,
      "step": 10780
    },
    {
      "entropy": 5.650118827819824,
      "epoch": 0.839136354794787,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004934953367144474,
      "loss": 5.3387,
      "mean_token_accuracy": 0.17719743847846986,
      "num_tokens": 18725694.0,
      "step": 10785
    },
    {
      "entropy": 5.557237815856934,
      "epoch": 0.8395253841665046,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004934887190108688,
      "loss": 5.2483,
      "mean_token_accuracy": 0.1862121567130089,
      "num_tokens": 18734737.0,
      "step": 10790
    },
    {
      "entropy": 5.691574382781982,
      "epoch": 0.8399144135382222,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004934820979920719,
      "loss": 5.4604,
      "mean_token_accuracy": 0.1678543671965599,
      "num_tokens": 18743895.0,
      "step": 10795
    },
    {
      "entropy": 5.637724208831787,
      "epoch": 0.8403034429099397,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004934754736581567,
      "loss": 5.3261,
      "mean_token_accuracy": 0.17763975709676744,
      "num_tokens": 18753105.0,
      "step": 10800
    },
    {
      "entropy": 5.481872081756592,
      "epoch": 0.8406924722816572,
      "grad_norm": 1.015625,
      "learning_rate": 0.000493468846009224,
      "loss": 5.186,
      "mean_token_accuracy": 0.18352127373218535,
      "num_tokens": 18761593.0,
      "step": 10805
    },
    {
      "entropy": 5.628566932678223,
      "epoch": 0.8410815016533748,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004934622150453742,
      "loss": 5.2388,
      "mean_token_accuracy": 0.18627728074789046,
      "num_tokens": 18770105.0,
      "step": 10810
    },
    {
      "entropy": 5.605173635482788,
      "epoch": 0.8414705310250924,
      "grad_norm": 0.953125,
      "learning_rate": 0.000493455580766708,
      "loss": 5.2866,
      "mean_token_accuracy": 0.17801288664340972,
      "num_tokens": 18778429.0,
      "step": 10815
    },
    {
      "entropy": 5.651268720626831,
      "epoch": 0.84185956039681,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004934489431733262,
      "loss": 5.327,
      "mean_token_accuracy": 0.17554858028888704,
      "num_tokens": 18787072.0,
      "step": 10820
    },
    {
      "entropy": 5.529681539535522,
      "epoch": 0.8422485897685276,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004934423022653293,
      "loss": 5.1998,
      "mean_token_accuracy": 0.18770936578512193,
      "num_tokens": 18795628.0,
      "step": 10825
    },
    {
      "entropy": 5.560863208770752,
      "epoch": 0.842637619140245,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004934356580428182,
      "loss": 5.2882,
      "mean_token_accuracy": 0.17767442017793655,
      "num_tokens": 18804313.0,
      "step": 10830
    },
    {
      "entropy": 5.613304376602173,
      "epoch": 0.8430266485119626,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004934290105058937,
      "loss": 5.3149,
      "mean_token_accuracy": 0.1774200201034546,
      "num_tokens": 18813170.0,
      "step": 10835
    },
    {
      "entropy": 5.5797830581665036,
      "epoch": 0.8434156778836802,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004934223596546565,
      "loss": 5.2632,
      "mean_token_accuracy": 0.1828976258635521,
      "num_tokens": 18823009.0,
      "step": 10840
    },
    {
      "entropy": 5.589580345153808,
      "epoch": 0.8438047072553978,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004934157054892077,
      "loss": 5.3109,
      "mean_token_accuracy": 0.17641228437423706,
      "num_tokens": 18831272.0,
      "step": 10845
    },
    {
      "entropy": 5.562094211578369,
      "epoch": 0.8441937366271154,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004934090480096482,
      "loss": 5.299,
      "mean_token_accuracy": 0.17283533960580827,
      "num_tokens": 18839481.0,
      "step": 10850
    },
    {
      "entropy": 5.531780529022217,
      "epoch": 0.8445827659988329,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004934023872160791,
      "loss": 5.26,
      "mean_token_accuracy": 0.18713913410902022,
      "num_tokens": 18848586.0,
      "step": 10855
    },
    {
      "entropy": 5.568729019165039,
      "epoch": 0.8449717953705504,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004933957231086014,
      "loss": 5.1588,
      "mean_token_accuracy": 0.18601678311824799,
      "num_tokens": 18856948.0,
      "step": 10860
    },
    {
      "entropy": 5.581716871261596,
      "epoch": 0.845360824742268,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004933890556873161,
      "loss": 5.2363,
      "mean_token_accuracy": 0.1828387498855591,
      "num_tokens": 18864741.0,
      "step": 10865
    },
    {
      "entropy": 5.566734266281128,
      "epoch": 0.8457498541139856,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004933823849523246,
      "loss": 5.2831,
      "mean_token_accuracy": 0.17859045118093492,
      "num_tokens": 18873373.0,
      "step": 10870
    },
    {
      "entropy": 5.621180152893066,
      "epoch": 0.8461388834857032,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004933757109037279,
      "loss": 5.1734,
      "mean_token_accuracy": 0.18435244262218475,
      "num_tokens": 18881590.0,
      "step": 10875
    },
    {
      "entropy": 5.598472595214844,
      "epoch": 0.8465279128574207,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004933690335416274,
      "loss": 5.2838,
      "mean_token_accuracy": 0.1795947551727295,
      "num_tokens": 18889838.0,
      "step": 10880
    },
    {
      "entropy": 5.509574937820434,
      "epoch": 0.8469169422291383,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004933623528661245,
      "loss": 5.1859,
      "mean_token_accuracy": 0.18743556141853332,
      "num_tokens": 18898577.0,
      "step": 10885
    },
    {
      "entropy": 5.6272741794586185,
      "epoch": 0.8473059716008559,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004933556688773203,
      "loss": 5.3269,
      "mean_token_accuracy": 0.1822746753692627,
      "num_tokens": 18907309.0,
      "step": 10890
    },
    {
      "entropy": 5.579664325714111,
      "epoch": 0.8476950009725734,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004933489815753165,
      "loss": 5.259,
      "mean_token_accuracy": 0.1802939161658287,
      "num_tokens": 18915956.0,
      "step": 10895
    },
    {
      "entropy": 5.562767219543457,
      "epoch": 0.848084030344291,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004933422909602143,
      "loss": 5.3698,
      "mean_token_accuracy": 0.17607285976409912,
      "num_tokens": 18924603.0,
      "step": 10900
    },
    {
      "entropy": 5.6034938335418705,
      "epoch": 0.8484730597160085,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004933355970321155,
      "loss": 5.2233,
      "mean_token_accuracy": 0.19376989603042602,
      "num_tokens": 18933231.0,
      "step": 10905
    },
    {
      "entropy": 5.643422651290893,
      "epoch": 0.8488620890877261,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004933288997911215,
      "loss": 5.3461,
      "mean_token_accuracy": 0.17191843837499618,
      "num_tokens": 18942172.0,
      "step": 10910
    },
    {
      "entropy": 5.601671886444092,
      "epoch": 0.8492511184594437,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004933221992373338,
      "loss": 5.3014,
      "mean_token_accuracy": 0.18420382738113403,
      "num_tokens": 18951440.0,
      "step": 10915
    },
    {
      "entropy": 5.595506048202514,
      "epoch": 0.8496401478311613,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004933154953708544,
      "loss": 5.2545,
      "mean_token_accuracy": 0.17833930402994155,
      "num_tokens": 18959797.0,
      "step": 10920
    },
    {
      "entropy": 5.569685554504394,
      "epoch": 0.8500291772028788,
      "grad_norm": 1.0,
      "learning_rate": 0.0004933087881917848,
      "loss": 5.2557,
      "mean_token_accuracy": 0.18182741552591325,
      "num_tokens": 18968811.0,
      "step": 10925
    },
    {
      "entropy": 5.647049331665039,
      "epoch": 0.8504182065745964,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004933020777002268,
      "loss": 5.3944,
      "mean_token_accuracy": 0.1689011871814728,
      "num_tokens": 18977467.0,
      "step": 10930
    },
    {
      "entropy": 5.63600492477417,
      "epoch": 0.8508072359463139,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004932953638962823,
      "loss": 5.2903,
      "mean_token_accuracy": 0.17715788632631302,
      "num_tokens": 18985760.0,
      "step": 10935
    },
    {
      "entropy": 5.675756120681763,
      "epoch": 0.8511962653180315,
      "grad_norm": 1.0625,
      "learning_rate": 0.000493288646780053,
      "loss": 5.3285,
      "mean_token_accuracy": 0.17073256522417068,
      "num_tokens": 18993812.0,
      "step": 10940
    },
    {
      "entropy": 5.579043006896972,
      "epoch": 0.8515852946897491,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004932819263516409,
      "loss": 5.3169,
      "mean_token_accuracy": 0.17365503907203675,
      "num_tokens": 19003056.0,
      "step": 10945
    },
    {
      "entropy": 5.640437602996826,
      "epoch": 0.8519743240614667,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004932752026111481,
      "loss": 5.326,
      "mean_token_accuracy": 0.18457741886377335,
      "num_tokens": 19011533.0,
      "step": 10950
    },
    {
      "entropy": 5.589516830444336,
      "epoch": 0.8523633534331843,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004932684755586765,
      "loss": 5.2553,
      "mean_token_accuracy": 0.18523511439561843,
      "num_tokens": 19020773.0,
      "step": 10955
    },
    {
      "entropy": 5.556193828582764,
      "epoch": 0.8527523828049017,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004932617451943282,
      "loss": 5.2692,
      "mean_token_accuracy": 0.1786767065525055,
      "num_tokens": 19029809.0,
      "step": 10960
    },
    {
      "entropy": 5.588405799865723,
      "epoch": 0.8531414121766193,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004932550115182053,
      "loss": 5.1919,
      "mean_token_accuracy": 0.18426750153303145,
      "num_tokens": 19038875.0,
      "step": 10965
    },
    {
      "entropy": 5.668352508544922,
      "epoch": 0.8535304415483369,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049324827453041,
      "loss": 5.2954,
      "mean_token_accuracy": 0.1859284058213234,
      "num_tokens": 19047273.0,
      "step": 10970
    },
    {
      "entropy": 5.5029138088226315,
      "epoch": 0.8539194709200545,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004932415342310446,
      "loss": 5.226,
      "mean_token_accuracy": 0.18578009456396102,
      "num_tokens": 19056309.0,
      "step": 10975
    },
    {
      "entropy": 5.576099920272827,
      "epoch": 0.8543085002917721,
      "grad_norm": 1.0,
      "learning_rate": 0.0004932347906202111,
      "loss": 5.1929,
      "mean_token_accuracy": 0.18371533304452897,
      "num_tokens": 19064487.0,
      "step": 10980
    },
    {
      "entropy": 5.629020071029663,
      "epoch": 0.8546975296634896,
      "grad_norm": 1.03125,
      "learning_rate": 0.000493228043698012,
      "loss": 5.4229,
      "mean_token_accuracy": 0.1786932975053787,
      "num_tokens": 19073431.0,
      "step": 10985
    },
    {
      "entropy": 5.579223012924194,
      "epoch": 0.8550865590352071,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004932212934645498,
      "loss": 5.2955,
      "mean_token_accuracy": 0.1799983873963356,
      "num_tokens": 19081648.0,
      "step": 10990
    },
    {
      "entropy": 5.494553661346435,
      "epoch": 0.8554755884069247,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004932145399199268,
      "loss": 5.2156,
      "mean_token_accuracy": 0.18520181775093078,
      "num_tokens": 19090581.0,
      "step": 10995
    },
    {
      "entropy": 5.652668523788452,
      "epoch": 0.8558646177786423,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004932077830642455,
      "loss": 5.291,
      "mean_token_accuracy": 0.17706617563962937,
      "num_tokens": 19098924.0,
      "step": 11000
    },
    {
      "entropy": 5.634372806549072,
      "epoch": 0.8562536471503599,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004932010228976084,
      "loss": 5.2951,
      "mean_token_accuracy": 0.18166774958372117,
      "num_tokens": 19108267.0,
      "step": 11005
    },
    {
      "entropy": 5.644163608551025,
      "epoch": 0.8566426765220774,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000493194259420118,
      "loss": 5.3173,
      "mean_token_accuracy": 0.1759129285812378,
      "num_tokens": 19117158.0,
      "step": 11010
    },
    {
      "entropy": 5.562337493896484,
      "epoch": 0.857031705893795,
      "grad_norm": 0.9375,
      "learning_rate": 0.000493187492631877,
      "loss": 5.2836,
      "mean_token_accuracy": 0.18314954042434692,
      "num_tokens": 19126013.0,
      "step": 11015
    },
    {
      "entropy": 5.554191589355469,
      "epoch": 0.8574207352655125,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004931807225329882,
      "loss": 5.2311,
      "mean_token_accuracy": 0.18219724297523499,
      "num_tokens": 19134655.0,
      "step": 11020
    },
    {
      "entropy": 5.543011665344238,
      "epoch": 0.8578097646372301,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004931739491235541,
      "loss": 5.1314,
      "mean_token_accuracy": 0.18265994936227797,
      "num_tokens": 19143311.0,
      "step": 11025
    },
    {
      "entropy": 5.607172536849975,
      "epoch": 0.8581987940089477,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004931671724036777,
      "loss": 5.2656,
      "mean_token_accuracy": 0.18656713962554933,
      "num_tokens": 19152422.0,
      "step": 11030
    },
    {
      "entropy": 5.608309650421143,
      "epoch": 0.8585878233806652,
      "grad_norm": 1.0,
      "learning_rate": 0.0004931603923734616,
      "loss": 5.3731,
      "mean_token_accuracy": 0.17477262318134307,
      "num_tokens": 19161510.0,
      "step": 11035
    },
    {
      "entropy": 5.52988109588623,
      "epoch": 0.8589768527523828,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004931536090330088,
      "loss": 5.1933,
      "mean_token_accuracy": 0.18769892752170564,
      "num_tokens": 19170097.0,
      "step": 11040
    },
    {
      "entropy": 5.5798032760620115,
      "epoch": 0.8593658821241004,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004931468223824222,
      "loss": 5.2359,
      "mean_token_accuracy": 0.18533996045589446,
      "num_tokens": 19179131.0,
      "step": 11045
    },
    {
      "entropy": 5.510780191421508,
      "epoch": 0.859754911495818,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004931400324218048,
      "loss": 5.1521,
      "mean_token_accuracy": 0.18891196697950363,
      "num_tokens": 19187289.0,
      "step": 11050
    },
    {
      "entropy": 5.5318817615509035,
      "epoch": 0.8601439408675355,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004931332391512597,
      "loss": 5.2588,
      "mean_token_accuracy": 0.18268831372261046,
      "num_tokens": 19195746.0,
      "step": 11055
    },
    {
      "entropy": 5.657726907730103,
      "epoch": 0.860532970239253,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004931264425708897,
      "loss": 5.2772,
      "mean_token_accuracy": 0.17957644909620285,
      "num_tokens": 19204683.0,
      "step": 11060
    },
    {
      "entropy": 5.565079832077027,
      "epoch": 0.8609219996109706,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004931196426807983,
      "loss": 5.1942,
      "mean_token_accuracy": 0.1739495277404785,
      "num_tokens": 19213405.0,
      "step": 11065
    },
    {
      "entropy": 5.558668422698974,
      "epoch": 0.8613110289826882,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004931128394810884,
      "loss": 5.2932,
      "mean_token_accuracy": 0.1813378795981407,
      "num_tokens": 19222717.0,
      "step": 11070
    },
    {
      "entropy": 5.656969976425171,
      "epoch": 0.8617000583544058,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004931060329718634,
      "loss": 5.2941,
      "mean_token_accuracy": 0.17425547689199447,
      "num_tokens": 19230688.0,
      "step": 11075
    },
    {
      "entropy": 5.613022518157959,
      "epoch": 0.8620890877261234,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004930992231532265,
      "loss": 5.3514,
      "mean_token_accuracy": 0.1688617140054703,
      "num_tokens": 19239192.0,
      "step": 11080
    },
    {
      "entropy": 5.621093082427978,
      "epoch": 0.8624781170978408,
      "grad_norm": 1.03125,
      "learning_rate": 0.000493092410025281,
      "loss": 5.3322,
      "mean_token_accuracy": 0.17372597455978395,
      "num_tokens": 19247903.0,
      "step": 11085
    },
    {
      "entropy": 5.603265428543091,
      "epoch": 0.8628671464695584,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004930855935881302,
      "loss": 5.2803,
      "mean_token_accuracy": 0.17990875244140625,
      "num_tokens": 19256127.0,
      "step": 11090
    },
    {
      "entropy": 5.596434783935547,
      "epoch": 0.863256175841276,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004930787738418777,
      "loss": 5.2475,
      "mean_token_accuracy": 0.18377451449632645,
      "num_tokens": 19264742.0,
      "step": 11095
    },
    {
      "entropy": 5.573207521438599,
      "epoch": 0.8636452052129936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004930719507866269,
      "loss": 5.213,
      "mean_token_accuracy": 0.18178284168243408,
      "num_tokens": 19273618.0,
      "step": 11100
    },
    {
      "entropy": 5.607818603515625,
      "epoch": 0.8640342345847112,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004930651244224814,
      "loss": 5.2712,
      "mean_token_accuracy": 0.18519938588142396,
      "num_tokens": 19281792.0,
      "step": 11105
    },
    {
      "entropy": 5.511165904998779,
      "epoch": 0.8644232639564287,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004930582947495448,
      "loss": 5.2364,
      "mean_token_accuracy": 0.17781448662281035,
      "num_tokens": 19289738.0,
      "step": 11110
    },
    {
      "entropy": 5.619218063354492,
      "epoch": 0.8648122933281462,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004930514617679206,
      "loss": 5.225,
      "mean_token_accuracy": 0.18328847289085387,
      "num_tokens": 19298711.0,
      "step": 11115
    },
    {
      "entropy": 5.670982360839844,
      "epoch": 0.8652013226998638,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004930446254777125,
      "loss": 5.3287,
      "mean_token_accuracy": 0.18125866651535033,
      "num_tokens": 19307435.0,
      "step": 11120
    },
    {
      "entropy": 5.551834154129028,
      "epoch": 0.8655903520715814,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004930377858790242,
      "loss": 5.2042,
      "mean_token_accuracy": 0.18154563158750534,
      "num_tokens": 19315327.0,
      "step": 11125
    },
    {
      "entropy": 5.545620679855347,
      "epoch": 0.865979381443299,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004930309429719595,
      "loss": 5.1551,
      "mean_token_accuracy": 0.1891895890235901,
      "num_tokens": 19323261.0,
      "step": 11130
    },
    {
      "entropy": 5.4682965755462645,
      "epoch": 0.8663684108150166,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004930240967566224,
      "loss": 5.1283,
      "mean_token_accuracy": 0.18640220314264297,
      "num_tokens": 19332297.0,
      "step": 11135
    },
    {
      "entropy": 5.541049814224243,
      "epoch": 0.8667574401867341,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004930172472331167,
      "loss": 5.1663,
      "mean_token_accuracy": 0.18291076570749282,
      "num_tokens": 19340848.0,
      "step": 11140
    },
    {
      "entropy": 5.5286298274993895,
      "epoch": 0.8671464695584516,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004930103944015463,
      "loss": 5.2089,
      "mean_token_accuracy": 0.18289046585559846,
      "num_tokens": 19349072.0,
      "step": 11145
    },
    {
      "entropy": 5.621784114837647,
      "epoch": 0.8675354989301692,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004930035382620149,
      "loss": 5.2659,
      "mean_token_accuracy": 0.18448480367660522,
      "num_tokens": 19357723.0,
      "step": 11150
    },
    {
      "entropy": 5.515363550186157,
      "epoch": 0.8679245283018868,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000492996678814627,
      "loss": 5.2027,
      "mean_token_accuracy": 0.1899164915084839,
      "num_tokens": 19365925.0,
      "step": 11155
    },
    {
      "entropy": 5.50625205039978,
      "epoch": 0.8683135576736044,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004929898160594865,
      "loss": 5.2557,
      "mean_token_accuracy": 0.1823277398943901,
      "num_tokens": 19374190.0,
      "step": 11160
    },
    {
      "entropy": 5.544011402130127,
      "epoch": 0.8687025870453219,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004929829499966974,
      "loss": 5.244,
      "mean_token_accuracy": 0.18081277012825012,
      "num_tokens": 19382375.0,
      "step": 11165
    },
    {
      "entropy": 5.645644044876098,
      "epoch": 0.8690916164170395,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004929760806263641,
      "loss": 5.2952,
      "mean_token_accuracy": 0.18304144740104675,
      "num_tokens": 19390191.0,
      "step": 11170
    },
    {
      "entropy": 5.537923336029053,
      "epoch": 0.8694806457887571,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004929692079485906,
      "loss": 5.0954,
      "mean_token_accuracy": 0.19233231097459794,
      "num_tokens": 19399401.0,
      "step": 11175
    },
    {
      "entropy": 5.514873600006103,
      "epoch": 0.8698696751604746,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004929623319634814,
      "loss": 5.1988,
      "mean_token_accuracy": 0.1855992019176483,
      "num_tokens": 19407956.0,
      "step": 11180
    },
    {
      "entropy": 5.560956764221191,
      "epoch": 0.8702587045321922,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004929554526711407,
      "loss": 5.3478,
      "mean_token_accuracy": 0.180048431456089,
      "num_tokens": 19416432.0,
      "step": 11185
    },
    {
      "entropy": 5.546083641052246,
      "epoch": 0.8706477339039097,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004929485700716729,
      "loss": 5.2248,
      "mean_token_accuracy": 0.18299429714679719,
      "num_tokens": 19425208.0,
      "step": 11190
    },
    {
      "entropy": 5.644758796691894,
      "epoch": 0.8710367632756273,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004929416841651824,
      "loss": 5.3298,
      "mean_token_accuracy": 0.18089791685342788,
      "num_tokens": 19434169.0,
      "step": 11195
    },
    {
      "entropy": 5.550636720657349,
      "epoch": 0.8714257926473449,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004929347949517739,
      "loss": 5.1011,
      "mean_token_accuracy": 0.19305706769227982,
      "num_tokens": 19443484.0,
      "step": 11200
    },
    {
      "entropy": 5.513714361190796,
      "epoch": 0.8718148220190625,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004929279024315516,
      "loss": 5.2667,
      "mean_token_accuracy": 0.18305621594190596,
      "num_tokens": 19452193.0,
      "step": 11205
    },
    {
      "entropy": 5.679654121398926,
      "epoch": 0.87220385139078,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004929210066046204,
      "loss": 5.3846,
      "mean_token_accuracy": 0.17571600824594497,
      "num_tokens": 19460684.0,
      "step": 11210
    },
    {
      "entropy": 5.62515287399292,
      "epoch": 0.8725928807624975,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004929141074710847,
      "loss": 5.2333,
      "mean_token_accuracy": 0.18159908354282378,
      "num_tokens": 19469206.0,
      "step": 11215
    },
    {
      "entropy": 5.55255913734436,
      "epoch": 0.8729819101342151,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004929072050310491,
      "loss": 5.2785,
      "mean_token_accuracy": 0.17789066433906556,
      "num_tokens": 19477514.0,
      "step": 11220
    },
    {
      "entropy": 5.546413421630859,
      "epoch": 0.8733709395059327,
      "grad_norm": 1.125,
      "learning_rate": 0.0004929002992846188,
      "loss": 5.3287,
      "mean_token_accuracy": 0.17711522430181503,
      "num_tokens": 19486792.0,
      "step": 11225
    },
    {
      "entropy": 5.614936256408692,
      "epoch": 0.8737599688776503,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004928933902318981,
      "loss": 5.2662,
      "mean_token_accuracy": 0.1872934803366661,
      "num_tokens": 19495708.0,
      "step": 11230
    },
    {
      "entropy": 5.656894969940185,
      "epoch": 0.8741489982493679,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492886477872992,
      "loss": 5.4256,
      "mean_token_accuracy": 0.1650377556681633,
      "num_tokens": 19505063.0,
      "step": 11235
    },
    {
      "entropy": 5.657841396331787,
      "epoch": 0.8745380276210853,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004928795622080054,
      "loss": 5.3524,
      "mean_token_accuracy": 0.17346434891223908,
      "num_tokens": 19514641.0,
      "step": 11240
    },
    {
      "entropy": 5.57841305732727,
      "epoch": 0.8749270569928029,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004928726432370434,
      "loss": 5.2196,
      "mean_token_accuracy": 0.18648359775543213,
      "num_tokens": 19523136.0,
      "step": 11245
    },
    {
      "entropy": 5.529555797576904,
      "epoch": 0.8753160863645205,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004928657209602107,
      "loss": 5.1199,
      "mean_token_accuracy": 0.1928300753235817,
      "num_tokens": 19532258.0,
      "step": 11250
    },
    {
      "entropy": 5.586338853836059,
      "epoch": 0.8757051157362381,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004928587953776125,
      "loss": 5.3626,
      "mean_token_accuracy": 0.17475418597459794,
      "num_tokens": 19541235.0,
      "step": 11255
    },
    {
      "entropy": 5.591464805603027,
      "epoch": 0.8760941451079557,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004928518664893538,
      "loss": 5.2969,
      "mean_token_accuracy": 0.17780354842543603,
      "num_tokens": 19550259.0,
      "step": 11260
    },
    {
      "entropy": 5.645087766647339,
      "epoch": 0.8764831744796732,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004928449342955397,
      "loss": 5.2585,
      "mean_token_accuracy": 0.18799372911453247,
      "num_tokens": 19558500.0,
      "step": 11265
    },
    {
      "entropy": 5.588253593444824,
      "epoch": 0.8768722038513908,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004928379987962756,
      "loss": 5.2056,
      "mean_token_accuracy": 0.18315169215202332,
      "num_tokens": 19566562.0,
      "step": 11270
    },
    {
      "entropy": 5.54454984664917,
      "epoch": 0.8772612332231083,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004928310599916666,
      "loss": 5.23,
      "mean_token_accuracy": 0.1823637768626213,
      "num_tokens": 19575219.0,
      "step": 11275
    },
    {
      "entropy": 5.516556358337402,
      "epoch": 0.8776502625948259,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004928241178818178,
      "loss": 5.1663,
      "mean_token_accuracy": 0.18708829283714296,
      "num_tokens": 19584311.0,
      "step": 11280
    },
    {
      "entropy": 5.5517336368560795,
      "epoch": 0.8780392919665435,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004928171724668349,
      "loss": 5.3106,
      "mean_token_accuracy": 0.1739381805062294,
      "num_tokens": 19593802.0,
      "step": 11285
    },
    {
      "entropy": 5.596087551116943,
      "epoch": 0.878428321338261,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004928102237468229,
      "loss": 5.2712,
      "mean_token_accuracy": 0.18105054050683975,
      "num_tokens": 19603046.0,
      "step": 11290
    },
    {
      "entropy": 5.492772626876831,
      "epoch": 0.8788173507099786,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004928032717218876,
      "loss": 5.154,
      "mean_token_accuracy": 0.1845415100455284,
      "num_tokens": 19611609.0,
      "step": 11295
    },
    {
      "entropy": 5.5506798267364506,
      "epoch": 0.8792063800816962,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004927963163921343,
      "loss": 5.183,
      "mean_token_accuracy": 0.18141555041074753,
      "num_tokens": 19619380.0,
      "step": 11300
    },
    {
      "entropy": 5.6594970703125,
      "epoch": 0.8795954094534137,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004927893577576685,
      "loss": 5.3483,
      "mean_token_accuracy": 0.17350445538759232,
      "num_tokens": 19628336.0,
      "step": 11305
    },
    {
      "entropy": 5.58415036201477,
      "epoch": 0.8799844388251313,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004927823958185959,
      "loss": 5.3135,
      "mean_token_accuracy": 0.18502842932939528,
      "num_tokens": 19637888.0,
      "step": 11310
    },
    {
      "entropy": 5.542030572891235,
      "epoch": 0.8803734681968488,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492775430575022,
      "loss": 5.2315,
      "mean_token_accuracy": 0.17773975431919098,
      "num_tokens": 19646123.0,
      "step": 11315
    },
    {
      "entropy": 5.550410032272339,
      "epoch": 0.8807624975685664,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004927684620270527,
      "loss": 5.1987,
      "mean_token_accuracy": 0.18863066732883454,
      "num_tokens": 19655264.0,
      "step": 11320
    },
    {
      "entropy": 5.672732877731323,
      "epoch": 0.881151526940284,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004927614901747935,
      "loss": 5.3354,
      "mean_token_accuracy": 0.17010691314935683,
      "num_tokens": 19663594.0,
      "step": 11325
    },
    {
      "entropy": 5.629295063018799,
      "epoch": 0.8815405563120016,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004927545150183503,
      "loss": 5.2882,
      "mean_token_accuracy": 0.17685989290475845,
      "num_tokens": 19671515.0,
      "step": 11330
    },
    {
      "entropy": 5.561790657043457,
      "epoch": 0.8819295856837192,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000492747536557829,
      "loss": 5.1631,
      "mean_token_accuracy": 0.18699969053268434,
      "num_tokens": 19680288.0,
      "step": 11335
    },
    {
      "entropy": 5.577661752700806,
      "epoch": 0.8823186150554367,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004927405547933353,
      "loss": 5.1824,
      "mean_token_accuracy": 0.18766162991523744,
      "num_tokens": 19688717.0,
      "step": 11340
    },
    {
      "entropy": 5.499565696716308,
      "epoch": 0.8827076444271542,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004927335697249753,
      "loss": 5.1276,
      "mean_token_accuracy": 0.19210835099220275,
      "num_tokens": 19697412.0,
      "step": 11345
    },
    {
      "entropy": 5.540387344360352,
      "epoch": 0.8830966737988718,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004927265813528549,
      "loss": 5.2759,
      "mean_token_accuracy": 0.17797500044107437,
      "num_tokens": 19706127.0,
      "step": 11350
    },
    {
      "entropy": 5.658319091796875,
      "epoch": 0.8834857031705894,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004927195896770804,
      "loss": 5.3498,
      "mean_token_accuracy": 0.17664047032594682,
      "num_tokens": 19715178.0,
      "step": 11355
    },
    {
      "entropy": 5.6233717918396,
      "epoch": 0.883874732542307,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004927125946977574,
      "loss": 5.2247,
      "mean_token_accuracy": 0.18304519653320311,
      "num_tokens": 19723814.0,
      "step": 11360
    },
    {
      "entropy": 5.64864411354065,
      "epoch": 0.8842637619140246,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004927055964149923,
      "loss": 5.3606,
      "mean_token_accuracy": 0.17506654262542726,
      "num_tokens": 19732153.0,
      "step": 11365
    },
    {
      "entropy": 5.611151599884034,
      "epoch": 0.884652791285742,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004926985948288914,
      "loss": 5.2671,
      "mean_token_accuracy": 0.18468063473701476,
      "num_tokens": 19740251.0,
      "step": 11370
    },
    {
      "entropy": 5.501591968536377,
      "epoch": 0.8850418206574596,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004926915899395608,
      "loss": 5.1897,
      "mean_token_accuracy": 0.19013190865516663,
      "num_tokens": 19748695.0,
      "step": 11375
    },
    {
      "entropy": 5.552240657806396,
      "epoch": 0.8854308500291772,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004926845817471068,
      "loss": 5.2068,
      "mean_token_accuracy": 0.17810254842042922,
      "num_tokens": 19757674.0,
      "step": 11380
    },
    {
      "entropy": 5.509834671020508,
      "epoch": 0.8858198794008948,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004926775702516358,
      "loss": 5.0912,
      "mean_token_accuracy": 0.18988971710205077,
      "num_tokens": 19765555.0,
      "step": 11385
    },
    {
      "entropy": 5.489891624450683,
      "epoch": 0.8862089087726124,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004926705554532541,
      "loss": 5.2398,
      "mean_token_accuracy": 0.18355209976434708,
      "num_tokens": 19773986.0,
      "step": 11390
    },
    {
      "entropy": 5.598408555984497,
      "epoch": 0.8865979381443299,
      "grad_norm": 1.125,
      "learning_rate": 0.0004926635373520682,
      "loss": 5.2618,
      "mean_token_accuracy": 0.185488785803318,
      "num_tokens": 19782308.0,
      "step": 11395
    },
    {
      "entropy": 5.5852643013000485,
      "epoch": 0.8869869675160474,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004926565159481845,
      "loss": 5.2103,
      "mean_token_accuracy": 0.17813374549150468,
      "num_tokens": 19790447.0,
      "step": 11400
    },
    {
      "entropy": 5.586507368087768,
      "epoch": 0.887375996887765,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004926494912417097,
      "loss": 5.3312,
      "mean_token_accuracy": 0.1808769389986992,
      "num_tokens": 19798889.0,
      "step": 11405
    },
    {
      "entropy": 5.540809535980225,
      "epoch": 0.8877650262594826,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004926424632327503,
      "loss": 5.1603,
      "mean_token_accuracy": 0.18644642382860183,
      "num_tokens": 19807192.0,
      "step": 11410
    },
    {
      "entropy": 5.607935762405395,
      "epoch": 0.8881540556312002,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004926354319214129,
      "loss": 5.2886,
      "mean_token_accuracy": 0.18072779178619386,
      "num_tokens": 19815864.0,
      "step": 11415
    },
    {
      "entropy": 5.609080839157104,
      "epoch": 0.8885430850029177,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004926283973078041,
      "loss": 5.2193,
      "mean_token_accuracy": 0.17907162606716157,
      "num_tokens": 19823218.0,
      "step": 11420
    },
    {
      "entropy": 5.563449764251709,
      "epoch": 0.8889321143746353,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004926213593920309,
      "loss": 5.2678,
      "mean_token_accuracy": 0.18048640489578247,
      "num_tokens": 19832098.0,
      "step": 11425
    },
    {
      "entropy": 5.553020858764649,
      "epoch": 0.8893211437463528,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004926143181742,
      "loss": 5.1955,
      "mean_token_accuracy": 0.18601744025945663,
      "num_tokens": 19840180.0,
      "step": 11430
    },
    {
      "entropy": 5.574113178253174,
      "epoch": 0.8897101731180704,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004926072736544181,
      "loss": 5.2475,
      "mean_token_accuracy": 0.17808691263198853,
      "num_tokens": 19849016.0,
      "step": 11435
    },
    {
      "entropy": 5.4969141483306885,
      "epoch": 0.890099202489788,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004926002258327922,
      "loss": 5.1901,
      "mean_token_accuracy": 0.18305204212665557,
      "num_tokens": 19858044.0,
      "step": 11440
    },
    {
      "entropy": 5.507494163513184,
      "epoch": 0.8904882318615055,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004925931747094292,
      "loss": 5.1242,
      "mean_token_accuracy": 0.1915612429380417,
      "num_tokens": 19867557.0,
      "step": 11445
    },
    {
      "entropy": 5.559364366531372,
      "epoch": 0.8908772612332231,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004925861202844361,
      "loss": 5.2385,
      "mean_token_accuracy": 0.18321475982666016,
      "num_tokens": 19876009.0,
      "step": 11450
    },
    {
      "entropy": 5.592780590057373,
      "epoch": 0.8912662906049407,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049257906255792,
      "loss": 5.3242,
      "mean_token_accuracy": 0.177417029440403,
      "num_tokens": 19885085.0,
      "step": 11455
    },
    {
      "entropy": 5.660267639160156,
      "epoch": 0.8916553199766583,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492572001529988,
      "loss": 5.2678,
      "mean_token_accuracy": 0.1806852012872696,
      "num_tokens": 19893877.0,
      "step": 11460
    },
    {
      "entropy": 5.664147233963012,
      "epoch": 0.8920443493483758,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004925649372007469,
      "loss": 5.3886,
      "mean_token_accuracy": 0.17875148504972457,
      "num_tokens": 19902480.0,
      "step": 11465
    },
    {
      "entropy": 5.562815952301025,
      "epoch": 0.8924333787200933,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004925578695703045,
      "loss": 5.1605,
      "mean_token_accuracy": 0.1883439913392067,
      "num_tokens": 19910981.0,
      "step": 11470
    },
    {
      "entropy": 5.591741371154785,
      "epoch": 0.8928224080918109,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004925507986387676,
      "loss": 5.2847,
      "mean_token_accuracy": 0.1760623872280121,
      "num_tokens": 19920055.0,
      "step": 11475
    },
    {
      "entropy": 5.62313141822815,
      "epoch": 0.8932114374635285,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004925437244062436,
      "loss": 5.3377,
      "mean_token_accuracy": 0.18255451917648316,
      "num_tokens": 19928010.0,
      "step": 11480
    },
    {
      "entropy": 5.62455415725708,
      "epoch": 0.8936004668352461,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004925366468728397,
      "loss": 5.2213,
      "mean_token_accuracy": 0.1811926007270813,
      "num_tokens": 19935992.0,
      "step": 11485
    },
    {
      "entropy": 5.6355571269989015,
      "epoch": 0.8939894962069637,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004925295660386635,
      "loss": 5.3433,
      "mean_token_accuracy": 0.17380679547786712,
      "num_tokens": 19944522.0,
      "step": 11490
    },
    {
      "entropy": 5.688939332962036,
      "epoch": 0.8943785255786811,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004925224819038224,
      "loss": 5.2745,
      "mean_token_accuracy": 0.18633388429880143,
      "num_tokens": 19952991.0,
      "step": 11495
    },
    {
      "entropy": 5.544963312149048,
      "epoch": 0.8947675549503987,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004925153944684239,
      "loss": 5.2916,
      "mean_token_accuracy": 0.1772780016064644,
      "num_tokens": 19962301.0,
      "step": 11500
    },
    {
      "entropy": 5.602243041992187,
      "epoch": 0.8951565843221163,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004925083037325754,
      "loss": 5.3624,
      "mean_token_accuracy": 0.1738313317298889,
      "num_tokens": 19971159.0,
      "step": 11505
    },
    {
      "entropy": 5.633229351043701,
      "epoch": 0.8955456136938339,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004925012096963847,
      "loss": 5.2399,
      "mean_token_accuracy": 0.17955626994371415,
      "num_tokens": 19979608.0,
      "step": 11510
    },
    {
      "entropy": 5.585269641876221,
      "epoch": 0.8959346430655515,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004924941123599593,
      "loss": 5.2266,
      "mean_token_accuracy": 0.18074027001857756,
      "num_tokens": 19988348.0,
      "step": 11515
    },
    {
      "entropy": 5.546251201629639,
      "epoch": 0.896323672437269,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924870117234069,
      "loss": 5.2439,
      "mean_token_accuracy": 0.17923223376274108,
      "num_tokens": 19996970.0,
      "step": 11520
    },
    {
      "entropy": 5.660845375061035,
      "epoch": 0.8967127018089865,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004924799077868353,
      "loss": 5.3695,
      "mean_token_accuracy": 0.18099111020565034,
      "num_tokens": 20005297.0,
      "step": 11525
    },
    {
      "entropy": 5.6322602272033695,
      "epoch": 0.8971017311807041,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924728005503522,
      "loss": 5.2969,
      "mean_token_accuracy": 0.1728695034980774,
      "num_tokens": 20013971.0,
      "step": 11530
    },
    {
      "entropy": 5.596248197555542,
      "epoch": 0.8974907605524217,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004924656900140655,
      "loss": 5.2611,
      "mean_token_accuracy": 0.17985946238040923,
      "num_tokens": 20022536.0,
      "step": 11535
    },
    {
      "entropy": 5.545647048950196,
      "epoch": 0.8978797899241393,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924585761780831,
      "loss": 5.184,
      "mean_token_accuracy": 0.1842767894268036,
      "num_tokens": 20031676.0,
      "step": 11540
    },
    {
      "entropy": 5.617026472091675,
      "epoch": 0.8982688192958569,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000492451459042513,
      "loss": 5.2615,
      "mean_token_accuracy": 0.17981941103935242,
      "num_tokens": 20039783.0,
      "step": 11545
    },
    {
      "entropy": 5.565978336334228,
      "epoch": 0.8986578486675744,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492444338607463,
      "loss": 5.3078,
      "mean_token_accuracy": 0.17649320214986802,
      "num_tokens": 20047859.0,
      "step": 11550
    },
    {
      "entropy": 5.628445768356324,
      "epoch": 0.899046878039292,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004924372148730413,
      "loss": 5.3464,
      "mean_token_accuracy": 0.17112823724746704,
      "num_tokens": 20056711.0,
      "step": 11555
    },
    {
      "entropy": 5.5603772640228275,
      "epoch": 0.8994359074110095,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000492430087839356,
      "loss": 5.1817,
      "mean_token_accuracy": 0.18270686715841294,
      "num_tokens": 20065715.0,
      "step": 11560
    },
    {
      "entropy": 5.559036779403686,
      "epoch": 0.8998249367827271,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004924229575065152,
      "loss": 5.2834,
      "mean_token_accuracy": 0.18429706245660782,
      "num_tokens": 20073815.0,
      "step": 11565
    },
    {
      "entropy": 5.615834140777588,
      "epoch": 0.9002139661544447,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000492415823874627,
      "loss": 5.2647,
      "mean_token_accuracy": 0.18063076734542846,
      "num_tokens": 20082331.0,
      "step": 11570
    },
    {
      "entropy": 5.588377904891968,
      "epoch": 0.9006029955261622,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004924086869437998,
      "loss": 5.336,
      "mean_token_accuracy": 0.180070897936821,
      "num_tokens": 20091095.0,
      "step": 11575
    },
    {
      "entropy": 5.589910840988159,
      "epoch": 0.9009920248978798,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004924015467141417,
      "loss": 5.2478,
      "mean_token_accuracy": 0.18325827717781068,
      "num_tokens": 20100111.0,
      "step": 11580
    },
    {
      "entropy": 5.6593669891357425,
      "epoch": 0.9013810542695974,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004923944031857613,
      "loss": 5.3149,
      "mean_token_accuracy": 0.183930803835392,
      "num_tokens": 20108542.0,
      "step": 11585
    },
    {
      "entropy": 5.56063814163208,
      "epoch": 0.901770083641315,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004923872563587668,
      "loss": 5.2359,
      "mean_token_accuracy": 0.18408215492963792,
      "num_tokens": 20117831.0,
      "step": 11590
    },
    {
      "entropy": 5.555573034286499,
      "epoch": 0.9021591130130325,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004923801062332666,
      "loss": 5.1851,
      "mean_token_accuracy": 0.18191388994455338,
      "num_tokens": 20127322.0,
      "step": 11595
    },
    {
      "entropy": 5.639308929443359,
      "epoch": 0.90254814238475,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923729528093694,
      "loss": 5.2383,
      "mean_token_accuracy": 0.18221660554409028,
      "num_tokens": 20135772.0,
      "step": 11600
    },
    {
      "entropy": 5.572954702377319,
      "epoch": 0.9029371717564676,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004923657960871836,
      "loss": 5.2528,
      "mean_token_accuracy": 0.18276721090078354,
      "num_tokens": 20144110.0,
      "step": 11605
    },
    {
      "entropy": 5.613551473617553,
      "epoch": 0.9033262011281852,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004923586360668178,
      "loss": 5.3595,
      "mean_token_accuracy": 0.17548990547657012,
      "num_tokens": 20153391.0,
      "step": 11610
    },
    {
      "entropy": 5.531225395202637,
      "epoch": 0.9037152304999028,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004923514727483807,
      "loss": 5.1895,
      "mean_token_accuracy": 0.1855894073843956,
      "num_tokens": 20161877.0,
      "step": 11615
    },
    {
      "entropy": 5.497713804244995,
      "epoch": 0.9041042598716204,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492344306131981,
      "loss": 5.1516,
      "mean_token_accuracy": 0.18799446523189545,
      "num_tokens": 20169922.0,
      "step": 11620
    },
    {
      "entropy": 5.488319158554077,
      "epoch": 0.9044932892433378,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004923371362177272,
      "loss": 5.2081,
      "mean_token_accuracy": 0.18522389382123947,
      "num_tokens": 20177780.0,
      "step": 11625
    },
    {
      "entropy": 5.560370588302613,
      "epoch": 0.9048823186150554,
      "grad_norm": 1.0,
      "learning_rate": 0.0004923299630057284,
      "loss": 5.1973,
      "mean_token_accuracy": 0.18629174381494523,
      "num_tokens": 20186303.0,
      "step": 11630
    },
    {
      "entropy": 5.5183638572692875,
      "epoch": 0.905271347986773,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004923227864960934,
      "loss": 5.1837,
      "mean_token_accuracy": 0.18678800463676454,
      "num_tokens": 20194042.0,
      "step": 11635
    },
    {
      "entropy": 5.456225490570068,
      "epoch": 0.9056603773584906,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492315606688931,
      "loss": 5.1875,
      "mean_token_accuracy": 0.1878139778971672,
      "num_tokens": 20202929.0,
      "step": 11640
    },
    {
      "entropy": 5.558633136749267,
      "epoch": 0.9060494067302082,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004923084235843501,
      "loss": 5.2404,
      "mean_token_accuracy": 0.17828413993120193,
      "num_tokens": 20210864.0,
      "step": 11645
    },
    {
      "entropy": 5.628408622741699,
      "epoch": 0.9064384361019256,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004923012371824598,
      "loss": 5.2621,
      "mean_token_accuracy": 0.18141427040100097,
      "num_tokens": 20219390.0,
      "step": 11650
    },
    {
      "entropy": 5.574519968032837,
      "epoch": 0.9068274654736432,
      "grad_norm": 1.078125,
      "learning_rate": 0.000492294047483369,
      "loss": 5.2039,
      "mean_token_accuracy": 0.19042831808328628,
      "num_tokens": 20227241.0,
      "step": 11655
    },
    {
      "entropy": 5.537701225280761,
      "epoch": 0.9072164948453608,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004922868544871869,
      "loss": 5.357,
      "mean_token_accuracy": 0.176102514564991,
      "num_tokens": 20236001.0,
      "step": 11660
    },
    {
      "entropy": 5.62885308265686,
      "epoch": 0.9076055242170784,
      "grad_norm": 1.0,
      "learning_rate": 0.0004922796581940227,
      "loss": 5.2372,
      "mean_token_accuracy": 0.18050403892993927,
      "num_tokens": 20244316.0,
      "step": 11665
    },
    {
      "entropy": 5.531038522720337,
      "epoch": 0.907994553588796,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004922724586039855,
      "loss": 5.2143,
      "mean_token_accuracy": 0.1854308769106865,
      "num_tokens": 20252693.0,
      "step": 11670
    },
    {
      "entropy": 5.507435750961304,
      "epoch": 0.9083835829605135,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004922652557171846,
      "loss": 5.2566,
      "mean_token_accuracy": 0.1850678652524948,
      "num_tokens": 20262252.0,
      "step": 11675
    },
    {
      "entropy": 5.552498912811279,
      "epoch": 0.9087726123322311,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004922580495337292,
      "loss": 5.0966,
      "mean_token_accuracy": 0.1937366932630539,
      "num_tokens": 20270755.0,
      "step": 11680
    },
    {
      "entropy": 5.562260913848877,
      "epoch": 0.9091616417039486,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004922508400537288,
      "loss": 5.1804,
      "mean_token_accuracy": 0.18096505105495453,
      "num_tokens": 20278739.0,
      "step": 11685
    },
    {
      "entropy": 5.556400442123413,
      "epoch": 0.9095506710756662,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004922436272772926,
      "loss": 5.3292,
      "mean_token_accuracy": 0.1719320610165596,
      "num_tokens": 20288160.0,
      "step": 11690
    },
    {
      "entropy": 5.566889762878418,
      "epoch": 0.9099397004473838,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004922364112045301,
      "loss": 5.2625,
      "mean_token_accuracy": 0.18204254359006883,
      "num_tokens": 20296730.0,
      "step": 11695
    },
    {
      "entropy": 5.574968862533569,
      "epoch": 0.9103287298191013,
      "grad_norm": 1.0,
      "learning_rate": 0.000492229191835551,
      "loss": 5.2072,
      "mean_token_accuracy": 0.18102122992277145,
      "num_tokens": 20305941.0,
      "step": 11700
    },
    {
      "entropy": 5.555197858810425,
      "epoch": 0.9107177591908189,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004922219691704645,
      "loss": 5.2278,
      "mean_token_accuracy": 0.18434746861457824,
      "num_tokens": 20314081.0,
      "step": 11705
    },
    {
      "entropy": 5.618513488769532,
      "epoch": 0.9111067885625365,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004922147432093805,
      "loss": 5.2905,
      "mean_token_accuracy": 0.17522201985120772,
      "num_tokens": 20323004.0,
      "step": 11710
    },
    {
      "entropy": 5.584374189376831,
      "epoch": 0.911495817934254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004922075139524083,
      "loss": 5.1905,
      "mean_token_accuracy": 0.18178245425224304,
      "num_tokens": 20331122.0,
      "step": 11715
    },
    {
      "entropy": 5.551940536499023,
      "epoch": 0.9118848473059716,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492200281399658,
      "loss": 5.2733,
      "mean_token_accuracy": 0.17392054200172424,
      "num_tokens": 20339602.0,
      "step": 11720
    },
    {
      "entropy": 5.56371488571167,
      "epoch": 0.9122738766776891,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000492193045551239,
      "loss": 5.1372,
      "mean_token_accuracy": 0.1840537041425705,
      "num_tokens": 20348431.0,
      "step": 11725
    },
    {
      "entropy": 5.523512554168701,
      "epoch": 0.9126629060494067,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004921858064072612,
      "loss": 5.1864,
      "mean_token_accuracy": 0.1874503806233406,
      "num_tokens": 20357235.0,
      "step": 11730
    },
    {
      "entropy": 5.556927728652954,
      "epoch": 0.9130519354211243,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004921785639678347,
      "loss": 5.1602,
      "mean_token_accuracy": 0.1839824616909027,
      "num_tokens": 20366169.0,
      "step": 11735
    },
    {
      "entropy": 5.5867105484008786,
      "epoch": 0.9134409647928419,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000492171318233069,
      "loss": 5.2393,
      "mean_token_accuracy": 0.18073418438434602,
      "num_tokens": 20375213.0,
      "step": 11740
    },
    {
      "entropy": 5.527808046340942,
      "epoch": 0.9138299941645595,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004921640692030742,
      "loss": 5.2407,
      "mean_token_accuracy": 0.17796168327331544,
      "num_tokens": 20383992.0,
      "step": 11745
    },
    {
      "entropy": 5.578649473190308,
      "epoch": 0.914219023536277,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004921568168779603,
      "loss": 5.2133,
      "mean_token_accuracy": 0.1853245973587036,
      "num_tokens": 20392318.0,
      "step": 11750
    },
    {
      "entropy": 5.641834449768067,
      "epoch": 0.9146080529079945,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004921495612578374,
      "loss": 5.3316,
      "mean_token_accuracy": 0.1788714274764061,
      "num_tokens": 20401556.0,
      "step": 11755
    },
    {
      "entropy": 5.55460295677185,
      "epoch": 0.9149970822797121,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004921423023428156,
      "loss": 5.2237,
      "mean_token_accuracy": 0.18192702680826187,
      "num_tokens": 20410480.0,
      "step": 11760
    },
    {
      "entropy": 5.535108041763306,
      "epoch": 0.9153861116514297,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004921350401330049,
      "loss": 5.175,
      "mean_token_accuracy": 0.18815360516309737,
      "num_tokens": 20419748.0,
      "step": 11765
    },
    {
      "entropy": 5.574000644683838,
      "epoch": 0.9157751410231473,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004921277746285155,
      "loss": 5.2429,
      "mean_token_accuracy": 0.18199095427989959,
      "num_tokens": 20429552.0,
      "step": 11770
    },
    {
      "entropy": 5.713226652145385,
      "epoch": 0.9161641703948649,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004921205058294579,
      "loss": 5.3085,
      "mean_token_accuracy": 0.1789688929915428,
      "num_tokens": 20438476.0,
      "step": 11775
    },
    {
      "entropy": 5.5220348834991455,
      "epoch": 0.9165531997665823,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492113233735942,
      "loss": 5.1148,
      "mean_token_accuracy": 0.18550708293914794,
      "num_tokens": 20447035.0,
      "step": 11780
    },
    {
      "entropy": 5.5483856201171875,
      "epoch": 0.9169422291382999,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004921059583480785,
      "loss": 5.2711,
      "mean_token_accuracy": 0.18236800283193588,
      "num_tokens": 20456200.0,
      "step": 11785
    },
    {
      "entropy": 5.585720109939575,
      "epoch": 0.9173312585100175,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004920986796659775,
      "loss": 5.26,
      "mean_token_accuracy": 0.1862987145781517,
      "num_tokens": 20465444.0,
      "step": 11790
    },
    {
      "entropy": 5.536608648300171,
      "epoch": 0.9177202878817351,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004920913976897498,
      "loss": 5.1567,
      "mean_token_accuracy": 0.1878991261124611,
      "num_tokens": 20473851.0,
      "step": 11795
    },
    {
      "entropy": 5.518877983093262,
      "epoch": 0.9181093172534527,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004920841124195055,
      "loss": 5.2428,
      "mean_token_accuracy": 0.18687252402305604,
      "num_tokens": 20482232.0,
      "step": 11800
    },
    {
      "entropy": 5.495103645324707,
      "epoch": 0.9184983466251702,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004920768238553554,
      "loss": 5.1363,
      "mean_token_accuracy": 0.19422158151865004,
      "num_tokens": 20490498.0,
      "step": 11805
    },
    {
      "entropy": 5.533331823348999,
      "epoch": 0.9188873759968877,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004920695319974099,
      "loss": 5.2071,
      "mean_token_accuracy": 0.18390069752931595,
      "num_tokens": 20499125.0,
      "step": 11810
    },
    {
      "entropy": 5.585284805297851,
      "epoch": 0.9192764053686053,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004920622368457798,
      "loss": 5.3503,
      "mean_token_accuracy": 0.18015943318605424,
      "num_tokens": 20507665.0,
      "step": 11815
    },
    {
      "entropy": 5.595728492736816,
      "epoch": 0.9196654347403229,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004920549384005759,
      "loss": 5.2787,
      "mean_token_accuracy": 0.18034791201353073,
      "num_tokens": 20516509.0,
      "step": 11820
    },
    {
      "entropy": 5.597612285614014,
      "epoch": 0.9200544641120405,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004920476366619086,
      "loss": 5.3042,
      "mean_token_accuracy": 0.1835848107933998,
      "num_tokens": 20524830.0,
      "step": 11825
    },
    {
      "entropy": 5.581564140319824,
      "epoch": 0.920443493483758,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492040331629889,
      "loss": 5.2012,
      "mean_token_accuracy": 0.17886543720960618,
      "num_tokens": 20533746.0,
      "step": 11830
    },
    {
      "entropy": 5.539874839782715,
      "epoch": 0.9208325228554756,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004920330233046277,
      "loss": 5.1779,
      "mean_token_accuracy": 0.1898143023252487,
      "num_tokens": 20542350.0,
      "step": 11835
    },
    {
      "entropy": 5.5576598167419435,
      "epoch": 0.9212215522271932,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004920257116862359,
      "loss": 5.3021,
      "mean_token_accuracy": 0.18487888127565383,
      "num_tokens": 20551670.0,
      "step": 11840
    },
    {
      "entropy": 5.608811140060425,
      "epoch": 0.9216105815989107,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004920183967748241,
      "loss": 5.2466,
      "mean_token_accuracy": 0.17925604432821274,
      "num_tokens": 20560692.0,
      "step": 11845
    },
    {
      "entropy": 5.589293432235718,
      "epoch": 0.9219996109706283,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004920110785705037,
      "loss": 5.1998,
      "mean_token_accuracy": 0.18300125300884246,
      "num_tokens": 20569262.0,
      "step": 11850
    },
    {
      "entropy": 5.537121105194092,
      "epoch": 0.9223886403423458,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004920037570733857,
      "loss": 5.1988,
      "mean_token_accuracy": 0.1828903526067734,
      "num_tokens": 20577850.0,
      "step": 11855
    },
    {
      "entropy": 5.579100322723389,
      "epoch": 0.9227776697140634,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004919964322835809,
      "loss": 5.2813,
      "mean_token_accuracy": 0.17564146369695663,
      "num_tokens": 20587021.0,
      "step": 11860
    },
    {
      "entropy": 5.5887792110443115,
      "epoch": 0.923166699085781,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004919891042012006,
      "loss": 5.2035,
      "mean_token_accuracy": 0.1853077918291092,
      "num_tokens": 20594631.0,
      "step": 11865
    },
    {
      "entropy": 5.519110488891601,
      "epoch": 0.9235557284574986,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004919817728263562,
      "loss": 5.1377,
      "mean_token_accuracy": 0.18740917146205902,
      "num_tokens": 20602765.0,
      "step": 11870
    },
    {
      "entropy": 5.637236833572388,
      "epoch": 0.9239447578292161,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004919744381591586,
      "loss": 5.2648,
      "mean_token_accuracy": 0.19036677181720735,
      "num_tokens": 20611350.0,
      "step": 11875
    },
    {
      "entropy": 5.563019943237305,
      "epoch": 0.9243337872009336,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004919671001997193,
      "loss": 5.192,
      "mean_token_accuracy": 0.18654754161834716,
      "num_tokens": 20620201.0,
      "step": 11880
    },
    {
      "entropy": 5.603507232666016,
      "epoch": 0.9247228165726512,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004919597589481497,
      "loss": 5.2478,
      "mean_token_accuracy": 0.1864937126636505,
      "num_tokens": 20627870.0,
      "step": 11885
    },
    {
      "entropy": 5.5766901016235355,
      "epoch": 0.9251118459443688,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000491952414404561,
      "loss": 5.2431,
      "mean_token_accuracy": 0.18415390998125075,
      "num_tokens": 20635894.0,
      "step": 11890
    },
    {
      "entropy": 5.59851861000061,
      "epoch": 0.9255008753160864,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004919450665690646,
      "loss": 5.2941,
      "mean_token_accuracy": 0.18335267305374145,
      "num_tokens": 20644091.0,
      "step": 11895
    },
    {
      "entropy": 5.58082594871521,
      "epoch": 0.925889904687804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004919377154417724,
      "loss": 5.1567,
      "mean_token_accuracy": 0.18717770129442216,
      "num_tokens": 20651994.0,
      "step": 11900
    },
    {
      "entropy": 5.537593746185303,
      "epoch": 0.9262789340595214,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004919303610227954,
      "loss": 5.1806,
      "mean_token_accuracy": 0.18851915895938873,
      "num_tokens": 20660192.0,
      "step": 11905
    },
    {
      "entropy": 5.65830979347229,
      "epoch": 0.926667963431239,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004919230033122457,
      "loss": 5.2585,
      "mean_token_accuracy": 0.1791141539812088,
      "num_tokens": 20668275.0,
      "step": 11910
    },
    {
      "entropy": 5.521208429336548,
      "epoch": 0.9270569928029566,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004919156423102345,
      "loss": 5.2391,
      "mean_token_accuracy": 0.18416546583175658,
      "num_tokens": 20676936.0,
      "step": 11915
    },
    {
      "entropy": 5.591800546646118,
      "epoch": 0.9274460221746742,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004919082780168736,
      "loss": 5.3476,
      "mean_token_accuracy": 0.17247433364391326,
      "num_tokens": 20685600.0,
      "step": 11920
    },
    {
      "entropy": 5.603094625473022,
      "epoch": 0.9278350515463918,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004919009104322751,
      "loss": 5.2913,
      "mean_token_accuracy": 0.18120404779911042,
      "num_tokens": 20693714.0,
      "step": 11925
    },
    {
      "entropy": 5.575480985641479,
      "epoch": 0.9282240809181093,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004918935395565503,
      "loss": 5.2368,
      "mean_token_accuracy": 0.1812037318944931,
      "num_tokens": 20702529.0,
      "step": 11930
    },
    {
      "entropy": 5.541878509521484,
      "epoch": 0.9286131102898268,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004918861653898113,
      "loss": 5.1647,
      "mean_token_accuracy": 0.1846123978495598,
      "num_tokens": 20711775.0,
      "step": 11935
    },
    {
      "entropy": 5.582861375808716,
      "epoch": 0.9290021396615444,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004918787879321701,
      "loss": 5.2248,
      "mean_token_accuracy": 0.18018672317266465,
      "num_tokens": 20720653.0,
      "step": 11940
    },
    {
      "entropy": 5.6048431396484375,
      "epoch": 0.929391169033262,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004918714071837384,
      "loss": 5.2494,
      "mean_token_accuracy": 0.18456124812364577,
      "num_tokens": 20729208.0,
      "step": 11945
    },
    {
      "entropy": 5.589124155044556,
      "epoch": 0.9297801984049796,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004918640231446282,
      "loss": 5.3006,
      "mean_token_accuracy": 0.1827417090535164,
      "num_tokens": 20738670.0,
      "step": 11950
    },
    {
      "entropy": 5.562734603881836,
      "epoch": 0.9301692277766972,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004918566358149517,
      "loss": 5.3508,
      "mean_token_accuracy": 0.17917992323637008,
      "num_tokens": 20748315.0,
      "step": 11955
    },
    {
      "entropy": 5.656414747238159,
      "epoch": 0.9305582571484147,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000491849245194821,
      "loss": 5.3902,
      "mean_token_accuracy": 0.17061794102191924,
      "num_tokens": 20757623.0,
      "step": 11960
    },
    {
      "entropy": 5.608323287963867,
      "epoch": 0.9309472865201323,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004918418512843479,
      "loss": 5.2138,
      "mean_token_accuracy": 0.18692825585603715,
      "num_tokens": 20765817.0,
      "step": 11965
    },
    {
      "entropy": 5.521482181549072,
      "epoch": 0.9313363158918498,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004918344540836451,
      "loss": 5.1088,
      "mean_token_accuracy": 0.19574368596076966,
      "num_tokens": 20773931.0,
      "step": 11970
    },
    {
      "entropy": 5.625129652023316,
      "epoch": 0.9317253452635674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004918270535928244,
      "loss": 5.2492,
      "mean_token_accuracy": 0.17974198013544082,
      "num_tokens": 20782652.0,
      "step": 11975
    },
    {
      "entropy": 5.567449998855591,
      "epoch": 0.932114374635285,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004918196498119984,
      "loss": 5.146,
      "mean_token_accuracy": 0.19232137799263,
      "num_tokens": 20791293.0,
      "step": 11980
    },
    {
      "entropy": 5.533898782730103,
      "epoch": 0.9325034040070025,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004918122427412793,
      "loss": 5.2505,
      "mean_token_accuracy": 0.17886458784341813,
      "num_tokens": 20800317.0,
      "step": 11985
    },
    {
      "entropy": 5.54930272102356,
      "epoch": 0.9328924333787201,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004918048323807795,
      "loss": 5.2269,
      "mean_token_accuracy": 0.19041681736707688,
      "num_tokens": 20809231.0,
      "step": 11990
    },
    {
      "entropy": 5.571619272232056,
      "epoch": 0.9332814627504377,
      "grad_norm": 1.0,
      "learning_rate": 0.0004917974187306114,
      "loss": 5.1505,
      "mean_token_accuracy": 0.1843850404024124,
      "num_tokens": 20818327.0,
      "step": 11995
    },
    {
      "entropy": 5.614779090881347,
      "epoch": 0.9336704921221552,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004917900017908875,
      "loss": 5.3149,
      "mean_token_accuracy": 0.17424997091293334,
      "num_tokens": 20826995.0,
      "step": 12000
    },
    {
      "epoch": 0.9336704921221552,
      "eval_entropy": 5.3652247088361635,
      "eval_loss": 5.260458469390869,
      "eval_mean_token_accuracy": 0.19026692330234893,
      "eval_num_tokens": 20826995.0,
      "eval_runtime": 21.5957,
      "eval_samples_per_second": 1924.368,
      "eval_steps_per_second": 240.558,
      "step": 12000
    },
    {
      "entropy": 5.476382541656494,
      "epoch": 0.9340595214938728,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004917825815617205,
      "loss": 5.0362,
      "mean_token_accuracy": 0.1938027784228325,
      "num_tokens": 20835641.0,
      "step": 12005
    },
    {
      "entropy": 5.586126899719238,
      "epoch": 0.9344485508655903,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004917751580432229,
      "loss": 5.1696,
      "mean_token_accuracy": 0.18758552074432372,
      "num_tokens": 20844728.0,
      "step": 12010
    },
    {
      "entropy": 5.511921501159668,
      "epoch": 0.9348375802373079,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004917677312355072,
      "loss": 5.1688,
      "mean_token_accuracy": 0.18927164226770402,
      "num_tokens": 20853013.0,
      "step": 12015
    },
    {
      "entropy": 5.629630088806152,
      "epoch": 0.9352266096090255,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004917603011386863,
      "loss": 5.24,
      "mean_token_accuracy": 0.18145138621330262,
      "num_tokens": 20861592.0,
      "step": 12020
    },
    {
      "entropy": 5.552721214294434,
      "epoch": 0.9356156389807431,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004917528677528727,
      "loss": 5.2046,
      "mean_token_accuracy": 0.18107561618089676,
      "num_tokens": 20869645.0,
      "step": 12025
    },
    {
      "entropy": 5.590920114517212,
      "epoch": 0.9360046683524607,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004917454310781795,
      "loss": 5.2526,
      "mean_token_accuracy": 0.18239778727293016,
      "num_tokens": 20878154.0,
      "step": 12030
    },
    {
      "entropy": 5.593302011489868,
      "epoch": 0.9363936977241781,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004917379911147193,
      "loss": 5.1964,
      "mean_token_accuracy": 0.19307635277509688,
      "num_tokens": 20886260.0,
      "step": 12035
    },
    {
      "entropy": 5.5298034191131595,
      "epoch": 0.9367827270958957,
      "grad_norm": 1.0,
      "learning_rate": 0.0004917305478626049,
      "loss": 5.1609,
      "mean_token_accuracy": 0.1862980455160141,
      "num_tokens": 20895688.0,
      "step": 12040
    },
    {
      "entropy": 5.5384777069091795,
      "epoch": 0.9371717564676133,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004917231013219495,
      "loss": 5.2756,
      "mean_token_accuracy": 0.1831613928079605,
      "num_tokens": 20903904.0,
      "step": 12045
    },
    {
      "entropy": 5.602621603012085,
      "epoch": 0.9375607858393309,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000491715651492866,
      "loss": 5.2673,
      "mean_token_accuracy": 0.18519030660390853,
      "num_tokens": 20913411.0,
      "step": 12050
    },
    {
      "entropy": 5.581089019775391,
      "epoch": 0.9379498152110485,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004917081983754675,
      "loss": 5.2303,
      "mean_token_accuracy": 0.18019794374704362,
      "num_tokens": 20922088.0,
      "step": 12055
    },
    {
      "entropy": 5.4662697315216064,
      "epoch": 0.938338844582766,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004917007419698669,
      "loss": 5.2267,
      "mean_token_accuracy": 0.18475015014410018,
      "num_tokens": 20930879.0,
      "step": 12060
    },
    {
      "entropy": 5.519509220123291,
      "epoch": 0.9387278739544835,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004916932822761776,
      "loss": 5.192,
      "mean_token_accuracy": 0.18773053586483002,
      "num_tokens": 20939092.0,
      "step": 12065
    },
    {
      "entropy": 5.567389726638794,
      "epoch": 0.9391169033262011,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004916858192945126,
      "loss": 5.2424,
      "mean_token_accuracy": 0.1826593279838562,
      "num_tokens": 20947599.0,
      "step": 12070
    },
    {
      "entropy": 5.617564725875854,
      "epoch": 0.9395059326979187,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004916783530249852,
      "loss": 5.297,
      "mean_token_accuracy": 0.17814621776342393,
      "num_tokens": 20955725.0,
      "step": 12075
    },
    {
      "entropy": 5.498813343048096,
      "epoch": 0.9398949620696363,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004916708834677086,
      "loss": 5.105,
      "mean_token_accuracy": 0.19145222306251525,
      "num_tokens": 20965040.0,
      "step": 12080
    },
    {
      "entropy": 5.581015491485596,
      "epoch": 0.9402839914413538,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004916634106227962,
      "loss": 5.2505,
      "mean_token_accuracy": 0.1820598751306534,
      "num_tokens": 20974204.0,
      "step": 12085
    },
    {
      "entropy": 5.6329758644104,
      "epoch": 0.9406730208130714,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004916559344903614,
      "loss": 5.2536,
      "mean_token_accuracy": 0.18456069827079774,
      "num_tokens": 20982109.0,
      "step": 12090
    },
    {
      "entropy": 5.579155397415161,
      "epoch": 0.941062050184789,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004916484550705176,
      "loss": 5.2732,
      "mean_token_accuracy": 0.1756502389907837,
      "num_tokens": 20990375.0,
      "step": 12095
    },
    {
      "entropy": 5.571983337402344,
      "epoch": 0.9414510795565065,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004916409723633785,
      "loss": 5.2959,
      "mean_token_accuracy": 0.18447652012109755,
      "num_tokens": 20998535.0,
      "step": 12100
    },
    {
      "entropy": 5.517458200454712,
      "epoch": 0.9418401089282241,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004916334863690573,
      "loss": 5.1476,
      "mean_token_accuracy": 0.18308708369731902,
      "num_tokens": 21006597.0,
      "step": 12105
    },
    {
      "entropy": 5.572100687026977,
      "epoch": 0.9422291382999416,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004916259970876678,
      "loss": 5.2134,
      "mean_token_accuracy": 0.18781745731830596,
      "num_tokens": 21015498.0,
      "step": 12110
    },
    {
      "entropy": 5.6254716396331785,
      "epoch": 0.9426181676716592,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004916185045193237,
      "loss": 5.2534,
      "mean_token_accuracy": 0.17619127333164214,
      "num_tokens": 21024090.0,
      "step": 12115
    },
    {
      "entropy": 5.594352388381958,
      "epoch": 0.9430071970433768,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004916110086641384,
      "loss": 5.2696,
      "mean_token_accuracy": 0.17645178884267806,
      "num_tokens": 21032928.0,
      "step": 12120
    },
    {
      "entropy": 5.620367193222046,
      "epoch": 0.9433962264150944,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004916035095222259,
      "loss": 5.3366,
      "mean_token_accuracy": 0.17488097101449968,
      "num_tokens": 21041467.0,
      "step": 12125
    },
    {
      "entropy": 5.601778411865235,
      "epoch": 0.9437852557868119,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004915960070936999,
      "loss": 5.2584,
      "mean_token_accuracy": 0.1897648349404335,
      "num_tokens": 21049325.0,
      "step": 12130
    },
    {
      "entropy": 5.568003797531128,
      "epoch": 0.9441742851585294,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004915885013786742,
      "loss": 5.3149,
      "mean_token_accuracy": 0.17808953821659088,
      "num_tokens": 21057911.0,
      "step": 12135
    },
    {
      "entropy": 5.611033773422241,
      "epoch": 0.944563314530247,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004915809923772628,
      "loss": 5.298,
      "mean_token_accuracy": 0.17284227013587952,
      "num_tokens": 21066974.0,
      "step": 12140
    },
    {
      "entropy": 5.602640533447266,
      "epoch": 0.9449523439019646,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004915734800895796,
      "loss": 5.2294,
      "mean_token_accuracy": 0.1814958468079567,
      "num_tokens": 21075591.0,
      "step": 12145
    },
    {
      "entropy": 5.572117328643799,
      "epoch": 0.9453413732736822,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004915659645157383,
      "loss": 5.2703,
      "mean_token_accuracy": 0.18109863251447678,
      "num_tokens": 21084496.0,
      "step": 12150
    },
    {
      "entropy": 5.6210774898529055,
      "epoch": 0.9457304026453998,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004915584456558535,
      "loss": 5.1573,
      "mean_token_accuracy": 0.18336624652147293,
      "num_tokens": 21092811.0,
      "step": 12155
    },
    {
      "entropy": 5.612228155136108,
      "epoch": 0.9461194320171173,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004915509235100388,
      "loss": 5.2221,
      "mean_token_accuracy": 0.18512363731861115,
      "num_tokens": 21101449.0,
      "step": 12160
    },
    {
      "entropy": 5.67351598739624,
      "epoch": 0.9465084613888348,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004915433980784086,
      "loss": 5.3351,
      "mean_token_accuracy": 0.17369821220636367,
      "num_tokens": 21109476.0,
      "step": 12165
    },
    {
      "entropy": 5.556552505493164,
      "epoch": 0.9468974907605524,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004915358693610769,
      "loss": 5.2253,
      "mean_token_accuracy": 0.18115261644124986,
      "num_tokens": 21118431.0,
      "step": 12170
    },
    {
      "entropy": 5.540670347213745,
      "epoch": 0.94728652013227,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004915283373581581,
      "loss": 5.2133,
      "mean_token_accuracy": 0.18355928510427474,
      "num_tokens": 21127390.0,
      "step": 12175
    },
    {
      "entropy": 5.659769105911255,
      "epoch": 0.9476755495039876,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004915208020697664,
      "loss": 5.3076,
      "mean_token_accuracy": 0.17318570911884307,
      "num_tokens": 21136323.0,
      "step": 12180
    },
    {
      "entropy": 5.5947404384613035,
      "epoch": 0.9480645788757052,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004915132634960162,
      "loss": 5.2529,
      "mean_token_accuracy": 0.1843479171395302,
      "num_tokens": 21145228.0,
      "step": 12185
    },
    {
      "entropy": 5.575587368011474,
      "epoch": 0.9484536082474226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004915057216370218,
      "loss": 5.1696,
      "mean_token_accuracy": 0.18563294559717178,
      "num_tokens": 21153288.0,
      "step": 12190
    },
    {
      "entropy": 5.552964353561402,
      "epoch": 0.9488426376191402,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004914981764928977,
      "loss": 5.215,
      "mean_token_accuracy": 0.1865030899643898,
      "num_tokens": 21161802.0,
      "step": 12195
    },
    {
      "entropy": 5.504632616043091,
      "epoch": 0.9492316669908578,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004914906280637584,
      "loss": 5.1629,
      "mean_token_accuracy": 0.18965061008930206,
      "num_tokens": 21170018.0,
      "step": 12200
    },
    {
      "entropy": 5.659635448455811,
      "epoch": 0.9496206963625754,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004914830763497183,
      "loss": 5.3583,
      "mean_token_accuracy": 0.16727759540081025,
      "num_tokens": 21179128.0,
      "step": 12205
    },
    {
      "entropy": 5.635401105880737,
      "epoch": 0.950009725734293,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004914755213508922,
      "loss": 5.2064,
      "mean_token_accuracy": 0.18544783741235732,
      "num_tokens": 21188138.0,
      "step": 12210
    },
    {
      "entropy": 5.600381088256836,
      "epoch": 0.9503987551060105,
      "grad_norm": 1.125,
      "learning_rate": 0.0004914679630673945,
      "loss": 5.2836,
      "mean_token_accuracy": 0.18301821649074554,
      "num_tokens": 21196465.0,
      "step": 12215
    },
    {
      "entropy": 5.557601499557495,
      "epoch": 0.950787784477728,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004914604014993401,
      "loss": 5.1828,
      "mean_token_accuracy": 0.1813039004802704,
      "num_tokens": 21205309.0,
      "step": 12220
    },
    {
      "entropy": 5.536256980895996,
      "epoch": 0.9511768138494456,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004914528366468436,
      "loss": 5.148,
      "mean_token_accuracy": 0.19110174030065535,
      "num_tokens": 21214216.0,
      "step": 12225
    },
    {
      "entropy": 5.52139630317688,
      "epoch": 0.9515658432211632,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004914452685100199,
      "loss": 5.1673,
      "mean_token_accuracy": 0.18793029487133026,
      "num_tokens": 21222246.0,
      "step": 12230
    },
    {
      "entropy": 5.547867727279663,
      "epoch": 0.9519548725928808,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004914376970889836,
      "loss": 5.2197,
      "mean_token_accuracy": 0.18555752485990523,
      "num_tokens": 21230696.0,
      "step": 12235
    },
    {
      "entropy": 5.6180586338043215,
      "epoch": 0.9523439019645983,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004914301223838497,
      "loss": 5.3321,
      "mean_token_accuracy": 0.18029067367315293,
      "num_tokens": 21239909.0,
      "step": 12240
    },
    {
      "entropy": 5.555740070343018,
      "epoch": 0.9527329313363159,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004914225443947334,
      "loss": 5.1897,
      "mean_token_accuracy": 0.1832535222172737,
      "num_tokens": 21248124.0,
      "step": 12245
    },
    {
      "entropy": 5.688931512832641,
      "epoch": 0.9531219607080335,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004914149631217494,
      "loss": 5.3857,
      "mean_token_accuracy": 0.1684218630194664,
      "num_tokens": 21257042.0,
      "step": 12250
    },
    {
      "entropy": 5.659134578704834,
      "epoch": 0.953510990079751,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004914073785650127,
      "loss": 5.319,
      "mean_token_accuracy": 0.17946300953626632,
      "num_tokens": 21265627.0,
      "step": 12255
    },
    {
      "entropy": 5.622348594665527,
      "epoch": 0.9539000194514686,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004913997907246385,
      "loss": 5.2302,
      "mean_token_accuracy": 0.1794123485684395,
      "num_tokens": 21273688.0,
      "step": 12260
    },
    {
      "entropy": 5.573261499404907,
      "epoch": 0.9542890488231861,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000491392199600742,
      "loss": 5.2167,
      "mean_token_accuracy": 0.1829856112599373,
      "num_tokens": 21283432.0,
      "step": 12265
    },
    {
      "entropy": 5.601358890533447,
      "epoch": 0.9546780781949037,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004913846051934382,
      "loss": 5.257,
      "mean_token_accuracy": 0.1796272873878479,
      "num_tokens": 21292003.0,
      "step": 12270
    },
    {
      "entropy": 5.541041088104248,
      "epoch": 0.9550671075666213,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004913770075028425,
      "loss": 5.1712,
      "mean_token_accuracy": 0.18643339425325395,
      "num_tokens": 21299914.0,
      "step": 12275
    },
    {
      "entropy": 5.59332242012024,
      "epoch": 0.9554561369383389,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004913694065290701,
      "loss": 5.3817,
      "mean_token_accuracy": 0.16797806769609452,
      "num_tokens": 21309398.0,
      "step": 12280
    },
    {
      "entropy": 5.588516473770142,
      "epoch": 0.9558451663100564,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004913618022722363,
      "loss": 5.2111,
      "mean_token_accuracy": 0.18153220415115356,
      "num_tokens": 21317828.0,
      "step": 12285
    },
    {
      "entropy": 5.65195107460022,
      "epoch": 0.9562341956817739,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004913541947324566,
      "loss": 5.3281,
      "mean_token_accuracy": 0.1786283001303673,
      "num_tokens": 21326266.0,
      "step": 12290
    },
    {
      "entropy": 5.663617181777954,
      "epoch": 0.9566232250534915,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004913465839098463,
      "loss": 5.2539,
      "mean_token_accuracy": 0.17708837538957595,
      "num_tokens": 21335476.0,
      "step": 12295
    },
    {
      "entropy": 5.6138804912567135,
      "epoch": 0.9570122544252091,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004913389698045209,
      "loss": 5.187,
      "mean_token_accuracy": 0.18348243534564973,
      "num_tokens": 21344045.0,
      "step": 12300
    },
    {
      "entropy": 5.578237009048462,
      "epoch": 0.9574012837969267,
      "grad_norm": 1.078125,
      "learning_rate": 0.000491331352416596,
      "loss": 5.2092,
      "mean_token_accuracy": 0.1864698812365532,
      "num_tokens": 21352309.0,
      "step": 12305
    },
    {
      "entropy": 5.512121343612671,
      "epoch": 0.9577903131686443,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004913237317461872,
      "loss": 5.2582,
      "mean_token_accuracy": 0.18393941223621368,
      "num_tokens": 21362164.0,
      "step": 12310
    },
    {
      "entropy": 5.687423753738403,
      "epoch": 0.9581793425403617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049131610779341,
      "loss": 5.218,
      "mean_token_accuracy": 0.18826501369476317,
      "num_tokens": 21372817.0,
      "step": 12315
    },
    {
      "entropy": 5.599769735336304,
      "epoch": 0.9585683719120793,
      "grad_norm": 1.125,
      "learning_rate": 0.0004913084805583803,
      "loss": 5.2089,
      "mean_token_accuracy": 0.18398718237876893,
      "num_tokens": 21381035.0,
      "step": 12320
    },
    {
      "entropy": 5.522907781600952,
      "epoch": 0.9589574012837969,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004913008500412136,
      "loss": 5.1759,
      "mean_token_accuracy": 0.19183382838964463,
      "num_tokens": 21390074.0,
      "step": 12325
    },
    {
      "entropy": 5.576920461654663,
      "epoch": 0.9593464306555145,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004912932162420259,
      "loss": 5.2284,
      "mean_token_accuracy": 0.17680705487728118,
      "num_tokens": 21398726.0,
      "step": 12330
    },
    {
      "entropy": 5.612211561203003,
      "epoch": 0.9597354600272321,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004912855791609327,
      "loss": 5.2216,
      "mean_token_accuracy": 0.17984077483415603,
      "num_tokens": 21407225.0,
      "step": 12335
    },
    {
      "entropy": 5.5571564674377445,
      "epoch": 0.9601244893989496,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004912779387980502,
      "loss": 5.2014,
      "mean_token_accuracy": 0.18145482540130614,
      "num_tokens": 21416746.0,
      "step": 12340
    },
    {
      "entropy": 5.579716920852661,
      "epoch": 0.9605135187706672,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004912702951534942,
      "loss": 5.2703,
      "mean_token_accuracy": 0.17398476898670195,
      "num_tokens": 21425621.0,
      "step": 12345
    },
    {
      "entropy": 5.5823915004730225,
      "epoch": 0.9609025481423847,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004912626482273809,
      "loss": 5.1622,
      "mean_token_accuracy": 0.19778977781534196,
      "num_tokens": 21433674.0,
      "step": 12350
    },
    {
      "entropy": 5.59567699432373,
      "epoch": 0.9612915775141023,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004912549980198258,
      "loss": 5.263,
      "mean_token_accuracy": 0.18211924731731416,
      "num_tokens": 21442285.0,
      "step": 12355
    },
    {
      "entropy": 5.591646671295166,
      "epoch": 0.9616806068858199,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004912473445309455,
      "loss": 5.2198,
      "mean_token_accuracy": 0.1867017537355423,
      "num_tokens": 21450345.0,
      "step": 12360
    },
    {
      "entropy": 5.490157270431519,
      "epoch": 0.9620696362575375,
      "grad_norm": 1.09375,
      "learning_rate": 0.000491239687760856,
      "loss": 5.1355,
      "mean_token_accuracy": 0.19054339826107025,
      "num_tokens": 21458618.0,
      "step": 12365
    },
    {
      "entropy": 5.485005903244018,
      "epoch": 0.962458665629255,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004912320277096732,
      "loss": 5.1701,
      "mean_token_accuracy": 0.18405425697565078,
      "num_tokens": 21467410.0,
      "step": 12370
    },
    {
      "entropy": 5.5545494556427,
      "epoch": 0.9628476950009726,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004912243643775137,
      "loss": 5.2239,
      "mean_token_accuracy": 0.17915583848953248,
      "num_tokens": 21475454.0,
      "step": 12375
    },
    {
      "entropy": 5.539155912399292,
      "epoch": 0.9632367243726901,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004912166977644936,
      "loss": 5.1399,
      "mean_token_accuracy": 0.18764171302318572,
      "num_tokens": 21483431.0,
      "step": 12380
    },
    {
      "entropy": 5.560909032821655,
      "epoch": 0.9636257537444077,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004912090278707293,
      "loss": 5.2157,
      "mean_token_accuracy": 0.18055916875600814,
      "num_tokens": 21491199.0,
      "step": 12385
    },
    {
      "entropy": 5.563940858840942,
      "epoch": 0.9640147831161253,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000491201354696337,
      "loss": 5.3014,
      "mean_token_accuracy": 0.17994246184825896,
      "num_tokens": 21500339.0,
      "step": 12390
    },
    {
      "entropy": 5.576026535034179,
      "epoch": 0.9644038124878428,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004911936782414334,
      "loss": 5.2521,
      "mean_token_accuracy": 0.18026733994483948,
      "num_tokens": 21508945.0,
      "step": 12395
    },
    {
      "entropy": 5.604353952407837,
      "epoch": 0.9647928418595604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004911859985061348,
      "loss": 5.214,
      "mean_token_accuracy": 0.1814992293715477,
      "num_tokens": 21517649.0,
      "step": 12400
    },
    {
      "entropy": 5.590505218505859,
      "epoch": 0.965181871231278,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004911783154905577,
      "loss": 5.2124,
      "mean_token_accuracy": 0.1816068634390831,
      "num_tokens": 21526650.0,
      "step": 12405
    },
    {
      "entropy": 5.612040615081787,
      "epoch": 0.9655709006029956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004911706291948188,
      "loss": 5.2318,
      "mean_token_accuracy": 0.18331041932106018,
      "num_tokens": 21535401.0,
      "step": 12410
    },
    {
      "entropy": 5.647918748855591,
      "epoch": 0.9659599299747131,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004911629396190348,
      "loss": 5.2607,
      "mean_token_accuracy": 0.17844109684228898,
      "num_tokens": 21543973.0,
      "step": 12415
    },
    {
      "entropy": 5.522848224639892,
      "epoch": 0.9663489593464306,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004911552467633221,
      "loss": 5.2032,
      "mean_token_accuracy": 0.18064285814762115,
      "num_tokens": 21552875.0,
      "step": 12420
    },
    {
      "entropy": 5.484639310836792,
      "epoch": 0.9667379887181482,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004911475506277976,
      "loss": 5.1251,
      "mean_token_accuracy": 0.1943623512983322,
      "num_tokens": 21561469.0,
      "step": 12425
    },
    {
      "entropy": 5.5896384716033936,
      "epoch": 0.9671270180898658,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000491139851212578,
      "loss": 5.1998,
      "mean_token_accuracy": 0.19158778935670853,
      "num_tokens": 21569523.0,
      "step": 12430
    },
    {
      "entropy": 5.535619258880615,
      "epoch": 0.9675160474615834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004911321485177802,
      "loss": 5.1592,
      "mean_token_accuracy": 0.18858758807182313,
      "num_tokens": 21577700.0,
      "step": 12435
    },
    {
      "entropy": 5.593519830703736,
      "epoch": 0.967905076833301,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004911244425435212,
      "loss": 5.2029,
      "mean_token_accuracy": 0.18343724608421325,
      "num_tokens": 21586233.0,
      "step": 12440
    },
    {
      "entropy": 5.572824001312256,
      "epoch": 0.9682941062050184,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004911167332899176,
      "loss": 5.2521,
      "mean_token_accuracy": 0.18704710602760316,
      "num_tokens": 21595330.0,
      "step": 12445
    },
    {
      "entropy": 5.528388977050781,
      "epoch": 0.968683135576736,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004911090207570867,
      "loss": 5.1758,
      "mean_token_accuracy": 0.19266084283590318,
      "num_tokens": 21604522.0,
      "step": 12450
    },
    {
      "entropy": 5.607219982147217,
      "epoch": 0.9690721649484536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004911013049451453,
      "loss": 5.1593,
      "mean_token_accuracy": 0.18465932756662368,
      "num_tokens": 21613599.0,
      "step": 12455
    },
    {
      "entropy": 5.546594190597534,
      "epoch": 0.9694611943201712,
      "grad_norm": 1.0,
      "learning_rate": 0.0004910935858542106,
      "loss": 5.2375,
      "mean_token_accuracy": 0.17949807196855544,
      "num_tokens": 21622587.0,
      "step": 12460
    },
    {
      "entropy": 5.554617023468017,
      "epoch": 0.9698502236918888,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004910858634843997,
      "loss": 5.2538,
      "mean_token_accuracy": 0.17767165154218673,
      "num_tokens": 21630782.0,
      "step": 12465
    },
    {
      "entropy": 5.584745407104492,
      "epoch": 0.9702392530636063,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004910781378358297,
      "loss": 5.1233,
      "mean_token_accuracy": 0.191708105802536,
      "num_tokens": 21639354.0,
      "step": 12470
    },
    {
      "entropy": 5.5527793884277346,
      "epoch": 0.9706282824353238,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004910704089086178,
      "loss": 5.2856,
      "mean_token_accuracy": 0.18243273645639418,
      "num_tokens": 21648292.0,
      "step": 12475
    },
    {
      "entropy": 5.603460311889648,
      "epoch": 0.9710173118070414,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004910626767028815,
      "loss": 5.2205,
      "mean_token_accuracy": 0.19073559641838073,
      "num_tokens": 21657382.0,
      "step": 12480
    },
    {
      "entropy": 5.598748397827149,
      "epoch": 0.971406341178759,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004910549412187379,
      "loss": 5.2136,
      "mean_token_accuracy": 0.17809648513793946,
      "num_tokens": 21666327.0,
      "step": 12485
    },
    {
      "entropy": 5.567792081832886,
      "epoch": 0.9717953705504766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004910472024563045,
      "loss": 5.2357,
      "mean_token_accuracy": 0.18883391767740249,
      "num_tokens": 21674882.0,
      "step": 12490
    },
    {
      "entropy": 5.679027414321899,
      "epoch": 0.9721843999221941,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004910394604156987,
      "loss": 5.3274,
      "mean_token_accuracy": 0.1758040115237236,
      "num_tokens": 21683255.0,
      "step": 12495
    },
    {
      "entropy": 5.648699760437012,
      "epoch": 0.9725734292939117,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004910317150970379,
      "loss": 5.1908,
      "mean_token_accuracy": 0.1852326661348343,
      "num_tokens": 21692274.0,
      "step": 12500
    },
    {
      "entropy": 5.4729467868804935,
      "epoch": 0.9729624586656292,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004910239665004397,
      "loss": 5.1105,
      "mean_token_accuracy": 0.19836493879556655,
      "num_tokens": 21700767.0,
      "step": 12505
    },
    {
      "entropy": 5.53968825340271,
      "epoch": 0.9733514880373468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004910162146260216,
      "loss": 5.2643,
      "mean_token_accuracy": 0.18542049676179886,
      "num_tokens": 21709430.0,
      "step": 12510
    },
    {
      "entropy": 5.546899795532227,
      "epoch": 0.9737405174090644,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004910084594739013,
      "loss": 5.1657,
      "mean_token_accuracy": 0.1877752274274826,
      "num_tokens": 21717978.0,
      "step": 12515
    },
    {
      "entropy": 5.43008942604065,
      "epoch": 0.9741295467807819,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004910007010441964,
      "loss": 5.0367,
      "mean_token_accuracy": 0.19908907860517502,
      "num_tokens": 21726608.0,
      "step": 12520
    },
    {
      "entropy": 5.491342258453369,
      "epoch": 0.9745185761524995,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004909929393370248,
      "loss": 5.1987,
      "mean_token_accuracy": 0.18284936547279357,
      "num_tokens": 21735402.0,
      "step": 12525
    },
    {
      "entropy": 5.523431444168091,
      "epoch": 0.9749076055242171,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004909851743525041,
      "loss": 5.1633,
      "mean_token_accuracy": 0.18822280317544937,
      "num_tokens": 21744620.0,
      "step": 12530
    },
    {
      "entropy": 5.5951985836029055,
      "epoch": 0.9752966348959347,
      "grad_norm": 1.09375,
      "learning_rate": 0.000490977406090752,
      "loss": 5.2426,
      "mean_token_accuracy": 0.1834847956895828,
      "num_tokens": 21753270.0,
      "step": 12535
    },
    {
      "entropy": 5.623544311523437,
      "epoch": 0.9756856642676522,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004909696345518867,
      "loss": 5.2587,
      "mean_token_accuracy": 0.1800254002213478,
      "num_tokens": 21761477.0,
      "step": 12540
    },
    {
      "entropy": 5.48270354270935,
      "epoch": 0.9760746936393697,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004909618597360258,
      "loss": 5.1723,
      "mean_token_accuracy": 0.1867610439658165,
      "num_tokens": 21769925.0,
      "step": 12545
    },
    {
      "entropy": 5.573631429672242,
      "epoch": 0.9764637230110873,
      "grad_norm": 1.0,
      "learning_rate": 0.0004909540816432876,
      "loss": 5.2373,
      "mean_token_accuracy": 0.1894602969288826,
      "num_tokens": 21778850.0,
      "step": 12550
    },
    {
      "entropy": 5.593955755233765,
      "epoch": 0.9768527523828049,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004909463002737897,
      "loss": 5.2941,
      "mean_token_accuracy": 0.17965374290943145,
      "num_tokens": 21787408.0,
      "step": 12555
    },
    {
      "entropy": 5.635477209091187,
      "epoch": 0.9772417817545225,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004909385156276506,
      "loss": 5.2508,
      "mean_token_accuracy": 0.1876464918255806,
      "num_tokens": 21795967.0,
      "step": 12560
    },
    {
      "entropy": 5.592546081542968,
      "epoch": 0.9776308111262401,
      "grad_norm": 1.0,
      "learning_rate": 0.0004909307277049881,
      "loss": 5.2039,
      "mean_token_accuracy": 0.1853122130036354,
      "num_tokens": 21804472.0,
      "step": 12565
    },
    {
      "entropy": 5.489376401901245,
      "epoch": 0.9780198404979576,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004909229365059205,
      "loss": 5.2057,
      "mean_token_accuracy": 0.18564027696847915,
      "num_tokens": 21813233.0,
      "step": 12570
    },
    {
      "entropy": 5.490174007415772,
      "epoch": 0.9784088698696751,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004909151420305661,
      "loss": 5.0931,
      "mean_token_accuracy": 0.19649251252412797,
      "num_tokens": 21821413.0,
      "step": 12575
    },
    {
      "entropy": 5.563841390609741,
      "epoch": 0.9787978992413927,
      "grad_norm": 1.125,
      "learning_rate": 0.000490907344279043,
      "loss": 5.1751,
      "mean_token_accuracy": 0.1877642899751663,
      "num_tokens": 21829222.0,
      "step": 12580
    },
    {
      "entropy": 5.5319915294647215,
      "epoch": 0.9791869286131103,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004908995432514698,
      "loss": 5.194,
      "mean_token_accuracy": 0.1828892335295677,
      "num_tokens": 21837033.0,
      "step": 12585
    },
    {
      "entropy": 5.637091732025146,
      "epoch": 0.9795759579848279,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004908917389479645,
      "loss": 5.2439,
      "mean_token_accuracy": 0.18043948113918304,
      "num_tokens": 21845464.0,
      "step": 12590
    },
    {
      "entropy": 5.51136360168457,
      "epoch": 0.9799649873565455,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004908839313686456,
      "loss": 5.1854,
      "mean_token_accuracy": 0.18375110179185866,
      "num_tokens": 21853497.0,
      "step": 12595
    },
    {
      "entropy": 5.593607711791992,
      "epoch": 0.980354016728263,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004908761205136319,
      "loss": 5.2689,
      "mean_token_accuracy": 0.18147289454936982,
      "num_tokens": 21861710.0,
      "step": 12600
    },
    {
      "entropy": 5.645310544967652,
      "epoch": 0.9807430460999805,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004908683063830414,
      "loss": 5.1995,
      "mean_token_accuracy": 0.17797971218824388,
      "num_tokens": 21870472.0,
      "step": 12605
    },
    {
      "entropy": 5.6198813915252686,
      "epoch": 0.9811320754716981,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004908604889769932,
      "loss": 5.2964,
      "mean_token_accuracy": 0.17758426517248155,
      "num_tokens": 21878287.0,
      "step": 12610
    },
    {
      "entropy": 5.634679317474365,
      "epoch": 0.9815211048434157,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004908526682956054,
      "loss": 5.3133,
      "mean_token_accuracy": 0.17233989685773848,
      "num_tokens": 21887587.0,
      "step": 12615
    },
    {
      "entropy": 5.562415647506714,
      "epoch": 0.9819101342151333,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004908448443389972,
      "loss": 5.1428,
      "mean_token_accuracy": 0.19184593260288238,
      "num_tokens": 21895923.0,
      "step": 12620
    },
    {
      "entropy": 5.562763929367065,
      "epoch": 0.9822991635868508,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004908370171072869,
      "loss": 5.18,
      "mean_token_accuracy": 0.19212259501218795,
      "num_tokens": 21904439.0,
      "step": 12625
    },
    {
      "entropy": 5.540858364105224,
      "epoch": 0.9826881929585684,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004908291866005933,
      "loss": 5.1488,
      "mean_token_accuracy": 0.18386462330818176,
      "num_tokens": 21912473.0,
      "step": 12630
    },
    {
      "entropy": 5.486727380752564,
      "epoch": 0.9830772223302859,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004908213528190355,
      "loss": 5.2317,
      "mean_token_accuracy": 0.17889027893543244,
      "num_tokens": 21921631.0,
      "step": 12635
    },
    {
      "entropy": 5.492412996292114,
      "epoch": 0.9834662517020035,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004908135157627321,
      "loss": 5.1311,
      "mean_token_accuracy": 0.19090985506772995,
      "num_tokens": 21929984.0,
      "step": 12640
    },
    {
      "entropy": 5.536947298049927,
      "epoch": 0.9838552810737211,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004908056754318022,
      "loss": 5.0891,
      "mean_token_accuracy": 0.19027605503797532,
      "num_tokens": 21937349.0,
      "step": 12645
    },
    {
      "entropy": 5.503980445861816,
      "epoch": 0.9842443104454386,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004907978318263646,
      "loss": 5.2024,
      "mean_token_accuracy": 0.1863625630736351,
      "num_tokens": 21946176.0,
      "step": 12650
    },
    {
      "entropy": 5.584465789794922,
      "epoch": 0.9846333398171562,
      "grad_norm": 1.125,
      "learning_rate": 0.0004907899849465383,
      "loss": 5.1879,
      "mean_token_accuracy": 0.18762608766555786,
      "num_tokens": 21954645.0,
      "step": 12655
    },
    {
      "entropy": 5.578148937225341,
      "epoch": 0.9850223691888738,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004907821347924424,
      "loss": 5.2262,
      "mean_token_accuracy": 0.17646457105875016,
      "num_tokens": 21963290.0,
      "step": 12660
    },
    {
      "entropy": 5.574510526657105,
      "epoch": 0.9854113985605913,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004907742813641963,
      "loss": 5.242,
      "mean_token_accuracy": 0.18453686386346818,
      "num_tokens": 21971005.0,
      "step": 12665
    },
    {
      "entropy": 5.538446092605591,
      "epoch": 0.9858004279323089,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004907664246619187,
      "loss": 5.1549,
      "mean_token_accuracy": 0.1910643011331558,
      "num_tokens": 21980067.0,
      "step": 12670
    },
    {
      "entropy": 5.566548156738281,
      "epoch": 0.9861894573040264,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004907585646857293,
      "loss": 5.1899,
      "mean_token_accuracy": 0.18357451856136323,
      "num_tokens": 21988511.0,
      "step": 12675
    },
    {
      "entropy": 5.535651922225952,
      "epoch": 0.986578486675744,
      "grad_norm": 1.0,
      "learning_rate": 0.0004907507014357467,
      "loss": 5.1687,
      "mean_token_accuracy": 0.18767756074666977,
      "num_tokens": 21997074.0,
      "step": 12680
    },
    {
      "entropy": 5.59835262298584,
      "epoch": 0.9869675160474616,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004907428349120909,
      "loss": 5.2995,
      "mean_token_accuracy": 0.1809012398123741,
      "num_tokens": 22005412.0,
      "step": 12685
    },
    {
      "entropy": 5.521641683578491,
      "epoch": 0.9873565454191792,
      "grad_norm": 1.0,
      "learning_rate": 0.0004907349651148809,
      "loss": 5.1009,
      "mean_token_accuracy": 0.19311340302228927,
      "num_tokens": 22013588.0,
      "step": 12690
    },
    {
      "entropy": 5.560098314285279,
      "epoch": 0.9877455747908968,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004907270920442361,
      "loss": 5.2467,
      "mean_token_accuracy": 0.18230018764734268,
      "num_tokens": 22021669.0,
      "step": 12695
    },
    {
      "entropy": 5.50910496711731,
      "epoch": 0.9881346041626142,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004907192157002762,
      "loss": 5.1106,
      "mean_token_accuracy": 0.19061779975891113,
      "num_tokens": 22030295.0,
      "step": 12700
    },
    {
      "entropy": 5.589793014526367,
      "epoch": 0.9885236335343318,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004907113360831204,
      "loss": 5.2625,
      "mean_token_accuracy": 0.18287634551525117,
      "num_tokens": 22038924.0,
      "step": 12705
    },
    {
      "entropy": 5.5686742782592775,
      "epoch": 0.9889126629060494,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004907034531928886,
      "loss": 5.1742,
      "mean_token_accuracy": 0.19149626344442366,
      "num_tokens": 22047586.0,
      "step": 12710
    },
    {
      "entropy": 5.523001956939697,
      "epoch": 0.989301692277767,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004906955670297001,
      "loss": 5.1547,
      "mean_token_accuracy": 0.1865146890282631,
      "num_tokens": 22055695.0,
      "step": 12715
    },
    {
      "entropy": 5.614638900756836,
      "epoch": 0.9896907216494846,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004906876775936746,
      "loss": 5.3082,
      "mean_token_accuracy": 0.17966843992471696,
      "num_tokens": 22064472.0,
      "step": 12720
    },
    {
      "entropy": 5.631753921508789,
      "epoch": 0.990079751021202,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004906797848849321,
      "loss": 5.2373,
      "mean_token_accuracy": 0.1752263605594635,
      "num_tokens": 22072726.0,
      "step": 12725
    },
    {
      "entropy": 5.581534147262573,
      "epoch": 0.9904687803929196,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000490671888903592,
      "loss": 5.1329,
      "mean_token_accuracy": 0.18759203255176543,
      "num_tokens": 22081074.0,
      "step": 12730
    },
    {
      "entropy": 5.500444126129151,
      "epoch": 0.9908578097646372,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004906639896497744,
      "loss": 5.1214,
      "mean_token_accuracy": 0.19771044552326203,
      "num_tokens": 22089924.0,
      "step": 12735
    },
    {
      "entropy": 5.588295030593872,
      "epoch": 0.9912468391363548,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004906560871235988,
      "loss": 5.1378,
      "mean_token_accuracy": 0.18912340700626373,
      "num_tokens": 22099088.0,
      "step": 12740
    },
    {
      "entropy": 5.651731348037719,
      "epoch": 0.9916358685080724,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004906481813251854,
      "loss": 5.2843,
      "mean_token_accuracy": 0.1810019001364708,
      "num_tokens": 22108344.0,
      "step": 12745
    },
    {
      "entropy": 5.563349723815918,
      "epoch": 0.9920248978797899,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004906402722546542,
      "loss": 5.197,
      "mean_token_accuracy": 0.18123072236776352,
      "num_tokens": 22116476.0,
      "step": 12750
    },
    {
      "entropy": 5.621095132827759,
      "epoch": 0.9924139272515075,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000490632359912125,
      "loss": 5.2968,
      "mean_token_accuracy": 0.1742379203438759,
      "num_tokens": 22125445.0,
      "step": 12755
    },
    {
      "entropy": 5.602992916107178,
      "epoch": 0.992802956623225,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000490624444297718,
      "loss": 5.3181,
      "mean_token_accuracy": 0.18033766895532607,
      "num_tokens": 22134886.0,
      "step": 12760
    },
    {
      "entropy": 5.525603151321411,
      "epoch": 0.9931919859949426,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004906165254115533,
      "loss": 5.1782,
      "mean_token_accuracy": 0.18814245462417603,
      "num_tokens": 22144521.0,
      "step": 12765
    },
    {
      "entropy": 5.496958541870117,
      "epoch": 0.9935810153666602,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000490608603253751,
      "loss": 5.1939,
      "mean_token_accuracy": 0.1895970657467842,
      "num_tokens": 22152754.0,
      "step": 12770
    },
    {
      "entropy": 5.548004865646362,
      "epoch": 0.9939700447383778,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004906006778244312,
      "loss": 5.1773,
      "mean_token_accuracy": 0.1882224589586258,
      "num_tokens": 22161711.0,
      "step": 12775
    },
    {
      "entropy": 5.605820941925049,
      "epoch": 0.9943590741100953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004905927491237145,
      "loss": 5.3014,
      "mean_token_accuracy": 0.18251292556524276,
      "num_tokens": 22169762.0,
      "step": 12780
    },
    {
      "entropy": 5.568092966079712,
      "epoch": 0.9947481034818129,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004905848171517208,
      "loss": 5.1474,
      "mean_token_accuracy": 0.18988931626081468,
      "num_tokens": 22178415.0,
      "step": 12785
    },
    {
      "entropy": 5.499771595001221,
      "epoch": 0.9951371328535304,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004905768819085706,
      "loss": 5.0934,
      "mean_token_accuracy": 0.19471519887447358,
      "num_tokens": 22187643.0,
      "step": 12790
    },
    {
      "entropy": 5.4495015144348145,
      "epoch": 0.995526162225248,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004905689433943846,
      "loss": 5.1344,
      "mean_token_accuracy": 0.189376999437809,
      "num_tokens": 22196355.0,
      "step": 12795
    },
    {
      "entropy": 5.576473808288574,
      "epoch": 0.9959151915969656,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004905610016092828,
      "loss": 5.2524,
      "mean_token_accuracy": 0.18064917623996735,
      "num_tokens": 22204648.0,
      "step": 12800
    },
    {
      "entropy": 5.555432415008545,
      "epoch": 0.9963042209686831,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004905530565533859,
      "loss": 5.2268,
      "mean_token_accuracy": 0.18231253921985627,
      "num_tokens": 22212811.0,
      "step": 12805
    },
    {
      "entropy": 5.482769107818603,
      "epoch": 0.9966932503404007,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004905451082268146,
      "loss": 5.1944,
      "mean_token_accuracy": 0.18336111754179002,
      "num_tokens": 22220611.0,
      "step": 12810
    },
    {
      "entropy": 5.578828048706055,
      "epoch": 0.9970822797121183,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004905371566296891,
      "loss": 5.2084,
      "mean_token_accuracy": 0.18011167496442795,
      "num_tokens": 22229606.0,
      "step": 12815
    },
    {
      "entropy": 5.642096424102784,
      "epoch": 0.9974713090838359,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004905292017621305,
      "loss": 5.1853,
      "mean_token_accuracy": 0.18634316176176072,
      "num_tokens": 22238157.0,
      "step": 12820
    },
    {
      "entropy": 5.547455549240112,
      "epoch": 0.9978603384555534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004905212436242593,
      "loss": 5.2221,
      "mean_token_accuracy": 0.18357642143964767,
      "num_tokens": 22246696.0,
      "step": 12825
    },
    {
      "entropy": 5.544768905639648,
      "epoch": 0.9982493678272709,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004905132822161962,
      "loss": 5.2526,
      "mean_token_accuracy": 0.18186190873384475,
      "num_tokens": 22255028.0,
      "step": 12830
    },
    {
      "entropy": 5.6687843799591064,
      "epoch": 0.9986383971989885,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004905053175380621,
      "loss": 5.3011,
      "mean_token_accuracy": 0.18592547625303268,
      "num_tokens": 22263198.0,
      "step": 12835
    },
    {
      "entropy": 5.623014688491821,
      "epoch": 0.9990274265707061,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004904973495899778,
      "loss": 5.2992,
      "mean_token_accuracy": 0.18507118821144103,
      "num_tokens": 22273780.0,
      "step": 12840
    },
    {
      "entropy": 5.659826850891113,
      "epoch": 0.9994164559424237,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004904893783720642,
      "loss": 5.3248,
      "mean_token_accuracy": 0.17507429122924806,
      "num_tokens": 22281545.0,
      "step": 12845
    },
    {
      "entropy": 5.550436544418335,
      "epoch": 0.9998054853141413,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004904814038844424,
      "loss": 5.2104,
      "mean_token_accuracy": 0.17978432923555374,
      "num_tokens": 22290426.0,
      "step": 12850
    },
    {
      "entropy": 5.602494610680474,
      "epoch": 1.000155611748687,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004904734261272332,
      "loss": 5.2511,
      "mean_token_accuracy": 0.1884353756904602,
      "num_tokens": 22297475.0,
      "step": 12855
    },
    {
      "entropy": 5.497782230377197,
      "epoch": 1.0005446411204046,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004904654451005576,
      "loss": 4.9866,
      "mean_token_accuracy": 0.1947847932577133,
      "num_tokens": 22305784.0,
      "step": 12860
    },
    {
      "entropy": 5.526057243347168,
      "epoch": 1.0009336704921221,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004904574608045369,
      "loss": 5.0917,
      "mean_token_accuracy": 0.18853476494550706,
      "num_tokens": 22313951.0,
      "step": 12865
    },
    {
      "entropy": 5.63850769996643,
      "epoch": 1.0013226998638398,
      "grad_norm": 1.125,
      "learning_rate": 0.0004904494732392923,
      "loss": 5.1972,
      "mean_token_accuracy": 0.18424230217933654,
      "num_tokens": 22322509.0,
      "step": 12870
    },
    {
      "entropy": 5.545155191421509,
      "epoch": 1.0017117292355573,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004904414824049448,
      "loss": 5.1513,
      "mean_token_accuracy": 0.18539026528596877,
      "num_tokens": 22331475.0,
      "step": 12875
    },
    {
      "entropy": 5.533708000183106,
      "epoch": 1.0021007586072748,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004904334883016156,
      "loss": 5.0775,
      "mean_token_accuracy": 0.19584077447652817,
      "num_tokens": 22341114.0,
      "step": 12880
    },
    {
      "entropy": 5.581373596191407,
      "epoch": 1.0024897879789925,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004904254909294261,
      "loss": 5.1495,
      "mean_token_accuracy": 0.190461228787899,
      "num_tokens": 22349199.0,
      "step": 12885
    },
    {
      "entropy": 5.566561937332153,
      "epoch": 1.00287881735071,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004904174902884978,
      "loss": 5.121,
      "mean_token_accuracy": 0.18924966603517532,
      "num_tokens": 22358485.0,
      "step": 12890
    },
    {
      "entropy": 5.634268808364868,
      "epoch": 1.0032678467224276,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004904094863789519,
      "loss": 5.1754,
      "mean_token_accuracy": 0.1821126490831375,
      "num_tokens": 22366936.0,
      "step": 12895
    },
    {
      "entropy": 5.652461814880371,
      "epoch": 1.0036568760941451,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049040147920091,
      "loss": 5.1534,
      "mean_token_accuracy": 0.18067814260721207,
      "num_tokens": 22376007.0,
      "step": 12900
    },
    {
      "entropy": 5.512387371063232,
      "epoch": 1.0040459054658626,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004903934687544933,
      "loss": 5.0686,
      "mean_token_accuracy": 0.2005569815635681,
      "num_tokens": 22384747.0,
      "step": 12905
    },
    {
      "entropy": 5.56094217300415,
      "epoch": 1.0044349348375803,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004903854550398237,
      "loss": 5.0965,
      "mean_token_accuracy": 0.19653735160827637,
      "num_tokens": 22393730.0,
      "step": 12910
    },
    {
      "entropy": 5.604393053054809,
      "epoch": 1.0048239642092978,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004903774380570226,
      "loss": 5.1229,
      "mean_token_accuracy": 0.18444818258285522,
      "num_tokens": 22402121.0,
      "step": 12915
    },
    {
      "entropy": 5.559228849411011,
      "epoch": 1.0052129935810155,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004903694178062117,
      "loss": 5.1351,
      "mean_token_accuracy": 0.18738230615854262,
      "num_tokens": 22410943.0,
      "step": 12920
    },
    {
      "entropy": 5.606460475921631,
      "epoch": 1.005602022952733,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004903613942875126,
      "loss": 5.2004,
      "mean_token_accuracy": 0.18455881029367446,
      "num_tokens": 22420137.0,
      "step": 12925
    },
    {
      "entropy": 5.575601530075073,
      "epoch": 1.0059910523244504,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004903533675010472,
      "loss": 5.1336,
      "mean_token_accuracy": 0.18333719819784164,
      "num_tokens": 22429031.0,
      "step": 12930
    },
    {
      "entropy": 5.526545190811158,
      "epoch": 1.006380081696168,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004903453374469373,
      "loss": 5.1805,
      "mean_token_accuracy": 0.18613078743219375,
      "num_tokens": 22438735.0,
      "step": 12935
    },
    {
      "entropy": 5.603658199310303,
      "epoch": 1.0067691110678856,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004903373041253045,
      "loss": 5.172,
      "mean_token_accuracy": 0.18485343158245088,
      "num_tokens": 22447374.0,
      "step": 12940
    },
    {
      "entropy": 5.6252185821533205,
      "epoch": 1.0071581404396033,
      "grad_norm": 1.0,
      "learning_rate": 0.0004903292675362709,
      "loss": 5.108,
      "mean_token_accuracy": 0.18936037570238112,
      "num_tokens": 22455542.0,
      "step": 12945
    },
    {
      "entropy": 5.505251741409301,
      "epoch": 1.0075471698113208,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004903212276799584,
      "loss": 5.1622,
      "mean_token_accuracy": 0.18490004986524583,
      "num_tokens": 22464591.0,
      "step": 12950
    },
    {
      "entropy": 5.559835195541382,
      "epoch": 1.0079361991830382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004903131845564889,
      "loss": 5.0936,
      "mean_token_accuracy": 0.19423397928476333,
      "num_tokens": 22472315.0,
      "step": 12955
    },
    {
      "entropy": 5.580622291564941,
      "epoch": 1.008325228554756,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004903051381659847,
      "loss": 5.2402,
      "mean_token_accuracy": 0.17842978835105897,
      "num_tokens": 22480662.0,
      "step": 12960
    },
    {
      "entropy": 5.4951142311096195,
      "epoch": 1.0087142579264734,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004902970885085677,
      "loss": 5.1334,
      "mean_token_accuracy": 0.18665957897901536,
      "num_tokens": 22489600.0,
      "step": 12965
    },
    {
      "entropy": 5.615678405761718,
      "epoch": 1.009103287298191,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049028903558436,
      "loss": 5.1742,
      "mean_token_accuracy": 0.1867367744445801,
      "num_tokens": 22498114.0,
      "step": 12970
    },
    {
      "entropy": 5.534137105941772,
      "epoch": 1.0094923166699086,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004902809793934838,
      "loss": 5.009,
      "mean_token_accuracy": 0.20099946856498718,
      "num_tokens": 22506349.0,
      "step": 12975
    },
    {
      "entropy": 5.538618516921997,
      "epoch": 1.009881346041626,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004902729199360615,
      "loss": 5.0936,
      "mean_token_accuracy": 0.1934603363275528,
      "num_tokens": 22514878.0,
      "step": 12980
    },
    {
      "entropy": 5.550827550888061,
      "epoch": 1.0102703754133437,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004902648572122153,
      "loss": 5.1364,
      "mean_token_accuracy": 0.17829016596078873,
      "num_tokens": 22522992.0,
      "step": 12985
    },
    {
      "entropy": 5.64747986793518,
      "epoch": 1.0106594047850612,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004902567912220674,
      "loss": 5.1871,
      "mean_token_accuracy": 0.18084194958209993,
      "num_tokens": 22531407.0,
      "step": 12990
    },
    {
      "entropy": 5.508750486373901,
      "epoch": 1.011048434156779,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004902487219657404,
      "loss": 5.0794,
      "mean_token_accuracy": 0.18966627568006517,
      "num_tokens": 22540255.0,
      "step": 12995
    },
    {
      "entropy": 5.533417701721191,
      "epoch": 1.0114374635284964,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004902406494433566,
      "loss": 5.1183,
      "mean_token_accuracy": 0.18966990560293198,
      "num_tokens": 22549139.0,
      "step": 13000
    },
    {
      "entropy": 5.5960956573486325,
      "epoch": 1.0118264929002139,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004902325736550386,
      "loss": 5.1184,
      "mean_token_accuracy": 0.1921576127409935,
      "num_tokens": 22557746.0,
      "step": 13005
    },
    {
      "entropy": 5.578761196136474,
      "epoch": 1.0122155222719316,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004902244946009088,
      "loss": 5.1094,
      "mean_token_accuracy": 0.18693674057722093,
      "num_tokens": 22566695.0,
      "step": 13010
    },
    {
      "entropy": 5.564864826202393,
      "epoch": 1.012604551643649,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004902164122810899,
      "loss": 5.1119,
      "mean_token_accuracy": 0.1888263612985611,
      "num_tokens": 22575196.0,
      "step": 13015
    },
    {
      "entropy": 5.566193246841431,
      "epoch": 1.0129935810153667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004902083266957045,
      "loss": 5.1133,
      "mean_token_accuracy": 0.1898096725344658,
      "num_tokens": 22583520.0,
      "step": 13020
    },
    {
      "entropy": 5.558519744873047,
      "epoch": 1.0133826103870842,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004902002378448753,
      "loss": 5.1734,
      "mean_token_accuracy": 0.18409187942743302,
      "num_tokens": 22591896.0,
      "step": 13025
    },
    {
      "entropy": 5.55157585144043,
      "epoch": 1.0137716397588017,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000490192145728725,
      "loss": 5.0948,
      "mean_token_accuracy": 0.19964988231658937,
      "num_tokens": 22600097.0,
      "step": 13030
    },
    {
      "entropy": 5.540172433853149,
      "epoch": 1.0141606691305194,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004901840503473764,
      "loss": 5.0375,
      "mean_token_accuracy": 0.18559068590402603,
      "num_tokens": 22607956.0,
      "step": 13035
    },
    {
      "entropy": 5.540137910842896,
      "epoch": 1.0145496985022369,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004901759517009522,
      "loss": 5.1338,
      "mean_token_accuracy": 0.17898587733507157,
      "num_tokens": 22616549.0,
      "step": 13040
    },
    {
      "entropy": 5.55298810005188,
      "epoch": 1.0149387278739546,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004901678497895755,
      "loss": 5.0563,
      "mean_token_accuracy": 0.1862951323390007,
      "num_tokens": 22625138.0,
      "step": 13045
    },
    {
      "entropy": 5.566691541671753,
      "epoch": 1.015327757245672,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004901597446133692,
      "loss": 5.1496,
      "mean_token_accuracy": 0.18422176837921142,
      "num_tokens": 22633808.0,
      "step": 13050
    },
    {
      "entropy": 5.528288269042969,
      "epoch": 1.0157167866173895,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004901516361724564,
      "loss": 5.0905,
      "mean_token_accuracy": 0.19693392962217332,
      "num_tokens": 22642517.0,
      "step": 13055
    },
    {
      "entropy": 5.581601953506469,
      "epoch": 1.0161058159891072,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004901435244669597,
      "loss": 5.1511,
      "mean_token_accuracy": 0.1824340358376503,
      "num_tokens": 22651822.0,
      "step": 13060
    },
    {
      "entropy": 5.583770418167115,
      "epoch": 1.0164948453608247,
      "grad_norm": 1.0,
      "learning_rate": 0.0004901354094970025,
      "loss": 5.1236,
      "mean_token_accuracy": 0.19076863974332808,
      "num_tokens": 22660763.0,
      "step": 13065
    },
    {
      "entropy": 5.521487760543823,
      "epoch": 1.0168838747325424,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004901272912627081,
      "loss": 5.1253,
      "mean_token_accuracy": 0.18654322177171706,
      "num_tokens": 22669449.0,
      "step": 13070
    },
    {
      "entropy": 5.5992106914520265,
      "epoch": 1.0172729041042599,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004901191697641992,
      "loss": 5.1501,
      "mean_token_accuracy": 0.18977264761924745,
      "num_tokens": 22678212.0,
      "step": 13075
    },
    {
      "entropy": 5.553779935836792,
      "epoch": 1.0176619334759773,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004901110450015994,
      "loss": 5.0905,
      "mean_token_accuracy": 0.19180919378995895,
      "num_tokens": 22686509.0,
      "step": 13080
    },
    {
      "entropy": 5.5005475044250485,
      "epoch": 1.018050962847695,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004901029169750319,
      "loss": 5.1146,
      "mean_token_accuracy": 0.18717952966690063,
      "num_tokens": 22695473.0,
      "step": 13085
    },
    {
      "entropy": 5.57834529876709,
      "epoch": 1.0184399922194125,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004900947856846201,
      "loss": 5.1808,
      "mean_token_accuracy": 0.18466604948043824,
      "num_tokens": 22704567.0,
      "step": 13090
    },
    {
      "entropy": 5.540963172912598,
      "epoch": 1.0188290215911302,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000490086651130487,
      "loss": 5.0544,
      "mean_token_accuracy": 0.19342739433050155,
      "num_tokens": 22713981.0,
      "step": 13095
    },
    {
      "entropy": 5.477065086364746,
      "epoch": 1.0192180509628477,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004900785133127566,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20526853054761887,
      "num_tokens": 22722855.0,
      "step": 13100
    },
    {
      "entropy": 5.523031425476074,
      "epoch": 1.0196070803345652,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000490070372231552,
      "loss": 5.1184,
      "mean_token_accuracy": 0.18693454563617706,
      "num_tokens": 22732454.0,
      "step": 13105
    },
    {
      "entropy": 5.524135732650757,
      "epoch": 1.0199961097062828,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004900622278869968,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20482657998800277,
      "num_tokens": 22741331.0,
      "step": 13110
    },
    {
      "entropy": 5.584689283370972,
      "epoch": 1.0203851390780003,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004900540802792146,
      "loss": 5.1816,
      "mean_token_accuracy": 0.1842080235481262,
      "num_tokens": 22749983.0,
      "step": 13115
    },
    {
      "entropy": 5.501961898803711,
      "epoch": 1.020774168449718,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000490045929408329,
      "loss": 5.1246,
      "mean_token_accuracy": 0.19006228744983672,
      "num_tokens": 22759112.0,
      "step": 13120
    },
    {
      "entropy": 5.560851001739502,
      "epoch": 1.0211631978214355,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900377752744637,
      "loss": 5.1706,
      "mean_token_accuracy": 0.18927655816078187,
      "num_tokens": 22768456.0,
      "step": 13125
    },
    {
      "entropy": 5.56893401145935,
      "epoch": 1.021552227193153,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900296178777426,
      "loss": 5.0771,
      "mean_token_accuracy": 0.19508728981018067,
      "num_tokens": 22777450.0,
      "step": 13130
    },
    {
      "entropy": 5.584453201293945,
      "epoch": 1.0219412565648707,
      "grad_norm": 1.09375,
      "learning_rate": 0.000490021457218289,
      "loss": 5.1347,
      "mean_token_accuracy": 0.19422568678855895,
      "num_tokens": 22785704.0,
      "step": 13135
    },
    {
      "entropy": 5.484196329116822,
      "epoch": 1.0223302859365881,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900132932962272,
      "loss": 5.1152,
      "mean_token_accuracy": 0.18883706331253053,
      "num_tokens": 22794649.0,
      "step": 13140
    },
    {
      "entropy": 5.600627803802491,
      "epoch": 1.0227193153083058,
      "grad_norm": 1.046875,
      "learning_rate": 0.000490005126111681,
      "loss": 5.2152,
      "mean_token_accuracy": 0.1855715736746788,
      "num_tokens": 22803501.0,
      "step": 13145
    },
    {
      "entropy": 5.671587705612183,
      "epoch": 1.0231083446800233,
      "grad_norm": 0.984375,
      "learning_rate": 0.000489996955664774,
      "loss": 5.1421,
      "mean_token_accuracy": 0.18789013475179672,
      "num_tokens": 22811779.0,
      "step": 13150
    },
    {
      "entropy": 5.528184700012207,
      "epoch": 1.023497374051741,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004899887819556306,
      "loss": 5.1439,
      "mean_token_accuracy": 0.18700369745492934,
      "num_tokens": 22821108.0,
      "step": 13155
    },
    {
      "entropy": 5.5858643531799315,
      "epoch": 1.0238864034234585,
      "grad_norm": 1.125,
      "learning_rate": 0.0004899806049843745,
      "loss": 5.1645,
      "mean_token_accuracy": 0.1873817428946495,
      "num_tokens": 22829369.0,
      "step": 13160
    },
    {
      "entropy": 5.567682218551636,
      "epoch": 1.024275432795176,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004899724247511299,
      "loss": 5.1211,
      "mean_token_accuracy": 0.1896756872534752,
      "num_tokens": 22838441.0,
      "step": 13165
    },
    {
      "entropy": 5.580206680297851,
      "epoch": 1.0246644621668937,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000489964241256021,
      "loss": 5.0986,
      "mean_token_accuracy": 0.18903386741876602,
      "num_tokens": 22847587.0,
      "step": 13170
    },
    {
      "entropy": 5.494615125656128,
      "epoch": 1.0250534915386111,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004899560544991718,
      "loss": 5.0652,
      "mean_token_accuracy": 0.19684161096811295,
      "num_tokens": 22856051.0,
      "step": 13175
    },
    {
      "entropy": 5.556175231933594,
      "epoch": 1.0254425209103288,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004899478644807068,
      "loss": 5.1392,
      "mean_token_accuracy": 0.18752205967903138,
      "num_tokens": 22864822.0,
      "step": 13180
    },
    {
      "entropy": 5.568244171142578,
      "epoch": 1.0258315502820463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004899396712007498,
      "loss": 5.111,
      "mean_token_accuracy": 0.18785069435834884,
      "num_tokens": 22873199.0,
      "step": 13185
    },
    {
      "entropy": 5.540246438980103,
      "epoch": 1.0262205796537638,
      "grad_norm": 1.125,
      "learning_rate": 0.0004899314746594256,
      "loss": 5.1798,
      "mean_token_accuracy": 0.18196630626916885,
      "num_tokens": 22881184.0,
      "step": 13190
    },
    {
      "entropy": 5.590057373046875,
      "epoch": 1.0266096090254815,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004899232748568584,
      "loss": 5.0867,
      "mean_token_accuracy": 0.19181957691907883,
      "num_tokens": 22890087.0,
      "step": 13195
    },
    {
      "entropy": 5.570810317993164,
      "epoch": 1.026998638397199,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004899150717931724,
      "loss": 5.1042,
      "mean_token_accuracy": 0.19140806347131728,
      "num_tokens": 22899013.0,
      "step": 13200
    },
    {
      "entropy": 5.522170639038086,
      "epoch": 1.0273876677689167,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004899068654684924,
      "loss": 5.2531,
      "mean_token_accuracy": 0.18083969503641129,
      "num_tokens": 22908740.0,
      "step": 13205
    },
    {
      "entropy": 5.623792028427124,
      "epoch": 1.0277766971406341,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004898986558829428,
      "loss": 5.1725,
      "mean_token_accuracy": 0.18605524599552153,
      "num_tokens": 22917502.0,
      "step": 13210
    },
    {
      "entropy": 5.571014165878296,
      "epoch": 1.0281657265123516,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004898904430366479,
      "loss": 5.0847,
      "mean_token_accuracy": 0.18972016572952272,
      "num_tokens": 22927076.0,
      "step": 13215
    },
    {
      "entropy": 5.582257509231567,
      "epoch": 1.0285547558840693,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004898822269297328,
      "loss": 5.1514,
      "mean_token_accuracy": 0.1842966914176941,
      "num_tokens": 22936111.0,
      "step": 13220
    },
    {
      "entropy": 5.503234243392944,
      "epoch": 1.0289437852557868,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004898740075623218,
      "loss": 5.0257,
      "mean_token_accuracy": 0.1970045030117035,
      "num_tokens": 22944752.0,
      "step": 13225
    },
    {
      "entropy": 5.614857482910156,
      "epoch": 1.0293328146275045,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004898657849345399,
      "loss": 5.2587,
      "mean_token_accuracy": 0.18242369294166566,
      "num_tokens": 22952971.0,
      "step": 13230
    },
    {
      "entropy": 5.598722982406616,
      "epoch": 1.029721843999222,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004898575590465116,
      "loss": 5.1625,
      "mean_token_accuracy": 0.19350814819335938,
      "num_tokens": 22961257.0,
      "step": 13235
    },
    {
      "entropy": 5.54152193069458,
      "epoch": 1.0301108733709394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004898493298983619,
      "loss": 5.2078,
      "mean_token_accuracy": 0.1872486114501953,
      "num_tokens": 22970659.0,
      "step": 13240
    },
    {
      "entropy": 5.5529930114746096,
      "epoch": 1.0304999027426571,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004898410974902155,
      "loss": 5.1463,
      "mean_token_accuracy": 0.19037771075963975,
      "num_tokens": 22978830.0,
      "step": 13245
    },
    {
      "entropy": 5.623561668395996,
      "epoch": 1.0308889321143746,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004898328618221974,
      "loss": 5.116,
      "mean_token_accuracy": 0.19342898428440095,
      "num_tokens": 22987100.0,
      "step": 13250
    },
    {
      "entropy": 5.573216152191162,
      "epoch": 1.0312779614860923,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004898246228944327,
      "loss": 5.1436,
      "mean_token_accuracy": 0.187310791015625,
      "num_tokens": 22995749.0,
      "step": 13255
    },
    {
      "entropy": 5.646274137496948,
      "epoch": 1.0316669908578098,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004898163807070461,
      "loss": 5.1989,
      "mean_token_accuracy": 0.19106748849153518,
      "num_tokens": 23004410.0,
      "step": 13260
    },
    {
      "entropy": 5.5814062595367435,
      "epoch": 1.0320560202295272,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004898081352601629,
      "loss": 5.0764,
      "mean_token_accuracy": 0.18742465525865554,
      "num_tokens": 23012935.0,
      "step": 13265
    },
    {
      "entropy": 5.515347146987915,
      "epoch": 1.032445049601245,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004897998865539082,
      "loss": 5.0627,
      "mean_token_accuracy": 0.19089292585849763,
      "num_tokens": 23021769.0,
      "step": 13270
    },
    {
      "entropy": 5.5167755603790285,
      "epoch": 1.0328340789729624,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004897916345884069,
      "loss": 5.0689,
      "mean_token_accuracy": 0.18745699375867844,
      "num_tokens": 23030915.0,
      "step": 13275
    },
    {
      "entropy": 5.565987968444825,
      "epoch": 1.0332231083446801,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004897833793637847,
      "loss": 5.184,
      "mean_token_accuracy": 0.18107736855745316,
      "num_tokens": 23039422.0,
      "step": 13280
    },
    {
      "entropy": 5.5884381294250485,
      "epoch": 1.0336121377163976,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004897751208801665,
      "loss": 5.2361,
      "mean_token_accuracy": 0.17725082188844682,
      "num_tokens": 23048485.0,
      "step": 13285
    },
    {
      "entropy": 5.524944734573364,
      "epoch": 1.034001167088115,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004897668591376777,
      "loss": 5.1014,
      "mean_token_accuracy": 0.19463111460208893,
      "num_tokens": 23057506.0,
      "step": 13290
    },
    {
      "entropy": 5.522286653518677,
      "epoch": 1.0343901964598328,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004897585941364436,
      "loss": 4.9591,
      "mean_token_accuracy": 0.20029387921094893,
      "num_tokens": 23065430.0,
      "step": 13295
    },
    {
      "entropy": 5.489220762252808,
      "epoch": 1.0347792258315502,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004897503258765896,
      "loss": 5.0845,
      "mean_token_accuracy": 0.18796505630016327,
      "num_tokens": 23074533.0,
      "step": 13300
    },
    {
      "entropy": 5.55793137550354,
      "epoch": 1.035168255203268,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004897420543582413,
      "loss": 5.0892,
      "mean_token_accuracy": 0.19122830927371978,
      "num_tokens": 23083020.0,
      "step": 13305
    },
    {
      "entropy": 5.533450269699097,
      "epoch": 1.0355572845749854,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489733779581524,
      "loss": 5.1358,
      "mean_token_accuracy": 0.1904940977692604,
      "num_tokens": 23091992.0,
      "step": 13310
    },
    {
      "entropy": 5.49093861579895,
      "epoch": 1.035946313946703,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004897255015465635,
      "loss": 5.1079,
      "mean_token_accuracy": 0.18566958904266356,
      "num_tokens": 23100411.0,
      "step": 13315
    },
    {
      "entropy": 5.536679458618164,
      "epoch": 1.0363353433184206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004897172202534853,
      "loss": 5.2,
      "mean_token_accuracy": 0.19183483123779296,
      "num_tokens": 23108867.0,
      "step": 13320
    },
    {
      "entropy": 5.553509473800659,
      "epoch": 1.036724372690138,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000489708935702415,
      "loss": 5.0427,
      "mean_token_accuracy": 0.19434165507555007,
      "num_tokens": 23116692.0,
      "step": 13325
    },
    {
      "entropy": 5.567583990097046,
      "epoch": 1.0371134020618558,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004897006478934784,
      "loss": 5.1178,
      "mean_token_accuracy": 0.19184658229351043,
      "num_tokens": 23125138.0,
      "step": 13330
    },
    {
      "entropy": 5.523353481292725,
      "epoch": 1.0375024314335732,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004896923568268011,
      "loss": 5.1292,
      "mean_token_accuracy": 0.1848578304052353,
      "num_tokens": 23133263.0,
      "step": 13335
    },
    {
      "entropy": 5.570543670654297,
      "epoch": 1.0378914608052907,
      "grad_norm": 1.109375,
      "learning_rate": 0.000489684062502509,
      "loss": 5.1076,
      "mean_token_accuracy": 0.18882636576890946,
      "num_tokens": 23141984.0,
      "step": 13340
    },
    {
      "entropy": 5.54804482460022,
      "epoch": 1.0382804901770084,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004896757649207281,
      "loss": 5.0817,
      "mean_token_accuracy": 0.19435103237628937,
      "num_tokens": 23151598.0,
      "step": 13345
    },
    {
      "entropy": 5.589086961746216,
      "epoch": 1.0386695195487259,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000489667464081584,
      "loss": 5.2228,
      "mean_token_accuracy": 0.18364229947328567,
      "num_tokens": 23159737.0,
      "step": 13350
    },
    {
      "entropy": 5.554082155227661,
      "epoch": 1.0390585489204436,
      "grad_norm": 1.0,
      "learning_rate": 0.0004896591599852029,
      "loss": 5.0743,
      "mean_token_accuracy": 0.19256037026643752,
      "num_tokens": 23168852.0,
      "step": 13355
    },
    {
      "entropy": 5.624046325683594,
      "epoch": 1.039447578292161,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004896508526317107,
      "loss": 5.1585,
      "mean_token_accuracy": 0.18748216480016708,
      "num_tokens": 23178227.0,
      "step": 13360
    },
    {
      "entropy": 5.567367076873779,
      "epoch": 1.0398366076638785,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004896425420212334,
      "loss": 5.0713,
      "mean_token_accuracy": 0.19270658642053604,
      "num_tokens": 23187286.0,
      "step": 13365
    },
    {
      "entropy": 5.575743103027344,
      "epoch": 1.0402256370355962,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004896342281538973,
      "loss": 5.1224,
      "mean_token_accuracy": 0.19453513473272324,
      "num_tokens": 23196163.0,
      "step": 13370
    },
    {
      "entropy": 5.56591157913208,
      "epoch": 1.0406146664073137,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004896259110298283,
      "loss": 5.0689,
      "mean_token_accuracy": 0.19389081746339798,
      "num_tokens": 23203742.0,
      "step": 13375
    },
    {
      "entropy": 5.644067621231079,
      "epoch": 1.0410036957790314,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004896175906491528,
      "loss": 5.2359,
      "mean_token_accuracy": 0.18386470526456833,
      "num_tokens": 23212860.0,
      "step": 13380
    },
    {
      "entropy": 5.61379861831665,
      "epoch": 1.0413927251507489,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004896092670119968,
      "loss": 5.1104,
      "mean_token_accuracy": 0.1852978527545929,
      "num_tokens": 23222558.0,
      "step": 13385
    },
    {
      "entropy": 5.571009683609009,
      "epoch": 1.0417817545224664,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004896009401184869,
      "loss": 5.1585,
      "mean_token_accuracy": 0.18610741943120956,
      "num_tokens": 23231413.0,
      "step": 13390
    },
    {
      "entropy": 5.546082162857056,
      "epoch": 1.042170783894184,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004895926099687493,
      "loss": 5.1014,
      "mean_token_accuracy": 0.19446830302476883,
      "num_tokens": 23240783.0,
      "step": 13395
    },
    {
      "entropy": 5.5714925765991214,
      "epoch": 1.0425598132659015,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004895842765629104,
      "loss": 5.2161,
      "mean_token_accuracy": 0.1871562495827675,
      "num_tokens": 23249125.0,
      "step": 13400
    },
    {
      "entropy": 5.57609338760376,
      "epoch": 1.0429488426376192,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004895759399010966,
      "loss": 5.0759,
      "mean_token_accuracy": 0.18885018080472946,
      "num_tokens": 23257793.0,
      "step": 13405
    },
    {
      "entropy": 5.483095693588257,
      "epoch": 1.0433378720093367,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004895675999834345,
      "loss": 5.046,
      "mean_token_accuracy": 0.19669701755046845,
      "num_tokens": 23266361.0,
      "step": 13410
    },
    {
      "entropy": 5.5332506656646725,
      "epoch": 1.0437269013810542,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004895592568100505,
      "loss": 5.0144,
      "mean_token_accuracy": 0.19876361936330794,
      "num_tokens": 23274914.0,
      "step": 13415
    },
    {
      "entropy": 5.540232467651367,
      "epoch": 1.0441159307527719,
      "grad_norm": 1.125,
      "learning_rate": 0.0004895509103810714,
      "loss": 5.1125,
      "mean_token_accuracy": 0.1861565440893173,
      "num_tokens": 23283067.0,
      "step": 13420
    },
    {
      "entropy": 5.55697603225708,
      "epoch": 1.0445049601244893,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004895425606966237,
      "loss": 5.1262,
      "mean_token_accuracy": 0.18422970175743103,
      "num_tokens": 23291114.0,
      "step": 13425
    },
    {
      "entropy": 5.493901014328003,
      "epoch": 1.044893989496207,
      "grad_norm": 1.0,
      "learning_rate": 0.000489534207756834,
      "loss": 4.9969,
      "mean_token_accuracy": 0.19421214163303374,
      "num_tokens": 23300525.0,
      "step": 13430
    },
    {
      "entropy": 5.550111293792725,
      "epoch": 1.0452830188679245,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004895258515618293,
      "loss": 5.092,
      "mean_token_accuracy": 0.19004881232976914,
      "num_tokens": 23308984.0,
      "step": 13435
    },
    {
      "entropy": 5.575098037719727,
      "epoch": 1.045672048239642,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004895174921117362,
      "loss": 5.1936,
      "mean_token_accuracy": 0.17996650338172912,
      "num_tokens": 23318137.0,
      "step": 13440
    },
    {
      "entropy": 5.6501775741577145,
      "epoch": 1.0460610776113597,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004895091294066816,
      "loss": 5.2549,
      "mean_token_accuracy": 0.17623160183429717,
      "num_tokens": 23326276.0,
      "step": 13445
    },
    {
      "entropy": 5.569949579238892,
      "epoch": 1.0464501069830772,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004895007634467924,
      "loss": 5.1273,
      "mean_token_accuracy": 0.18646959215402603,
      "num_tokens": 23334578.0,
      "step": 13450
    },
    {
      "entropy": 5.541098260879517,
      "epoch": 1.0468391363547949,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004894923942321955,
      "loss": 5.1495,
      "mean_token_accuracy": 0.1853663593530655,
      "num_tokens": 23344173.0,
      "step": 13455
    },
    {
      "entropy": 5.519333457946777,
      "epoch": 1.0472281657265123,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004894840217630179,
      "loss": 5.0135,
      "mean_token_accuracy": 0.1950036659836769,
      "num_tokens": 23353098.0,
      "step": 13460
    },
    {
      "entropy": 5.616471624374389,
      "epoch": 1.0476171950982298,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004894756460393868,
      "loss": 5.1416,
      "mean_token_accuracy": 0.18826956450939178,
      "num_tokens": 23362364.0,
      "step": 13465
    },
    {
      "entropy": 5.59700984954834,
      "epoch": 1.0480062244699475,
      "grad_norm": 1.015625,
      "learning_rate": 0.000489467267061429,
      "loss": 5.1801,
      "mean_token_accuracy": 0.180839404463768,
      "num_tokens": 23371903.0,
      "step": 13470
    },
    {
      "entropy": 5.537702465057373,
      "epoch": 1.048395253841665,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004894588848292718,
      "loss": 5.0958,
      "mean_token_accuracy": 0.19802405536174775,
      "num_tokens": 23381711.0,
      "step": 13475
    },
    {
      "entropy": 5.548175239562989,
      "epoch": 1.0487842832133827,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004894504993430425,
      "loss": 5.0529,
      "mean_token_accuracy": 0.18945176750421525,
      "num_tokens": 23389812.0,
      "step": 13480
    },
    {
      "entropy": 5.565440273284912,
      "epoch": 1.0491733125851002,
      "grad_norm": 1.078125,
      "learning_rate": 0.000489442110602868,
      "loss": 5.0834,
      "mean_token_accuracy": 0.20017406791448594,
      "num_tokens": 23399231.0,
      "step": 13485
    },
    {
      "entropy": 5.53766565322876,
      "epoch": 1.0495623419568176,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004894337186088759,
      "loss": 5.0657,
      "mean_token_accuracy": 0.18777775764465332,
      "num_tokens": 23407764.0,
      "step": 13490
    },
    {
      "entropy": 5.541928052902222,
      "epoch": 1.0499513713285353,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004894253233611934,
      "loss": 5.024,
      "mean_token_accuracy": 0.19526999145746232,
      "num_tokens": 23416368.0,
      "step": 13495
    },
    {
      "entropy": 5.633498764038086,
      "epoch": 1.0503404007002528,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000489416924859948,
      "loss": 5.2654,
      "mean_token_accuracy": 0.18364062011241913,
      "num_tokens": 23425415.0,
      "step": 13500
    },
    {
      "entropy": 5.5485481262207035,
      "epoch": 1.0507294300719705,
      "grad_norm": 1.03125,
      "learning_rate": 0.000489408523105267,
      "loss": 5.0778,
      "mean_token_accuracy": 0.18930239826440812,
      "num_tokens": 23433435.0,
      "step": 13505
    },
    {
      "entropy": 5.591840982437134,
      "epoch": 1.051118459443688,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004894001180972779,
      "loss": 5.1359,
      "mean_token_accuracy": 0.18588989675045015,
      "num_tokens": 23442002.0,
      "step": 13510
    },
    {
      "entropy": 5.486305904388428,
      "epoch": 1.0515074888154055,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004893917098361082,
      "loss": 4.929,
      "mean_token_accuracy": 0.19889771640300752,
      "num_tokens": 23450437.0,
      "step": 13515
    },
    {
      "entropy": 5.476190757751465,
      "epoch": 1.0518965181871232,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004893832983218856,
      "loss": 5.0278,
      "mean_token_accuracy": 0.19660004675388337,
      "num_tokens": 23458755.0,
      "step": 13520
    },
    {
      "entropy": 5.580454349517822,
      "epoch": 1.0522855475588406,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004893748835547377,
      "loss": 5.1623,
      "mean_token_accuracy": 0.18904335498809816,
      "num_tokens": 23467138.0,
      "step": 13525
    },
    {
      "entropy": 5.533040857315063,
      "epoch": 1.0526745769305583,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004893664655347921,
      "loss": 5.0258,
      "mean_token_accuracy": 0.19522487968206406,
      "num_tokens": 23475494.0,
      "step": 13530
    },
    {
      "entropy": 5.481938695907592,
      "epoch": 1.0530636063022758,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004893580442621767,
      "loss": 4.9874,
      "mean_token_accuracy": 0.19702944606542588,
      "num_tokens": 23484215.0,
      "step": 13535
    },
    {
      "entropy": 5.486063432693482,
      "epoch": 1.0534526356739935,
      "grad_norm": 1.015625,
      "learning_rate": 0.000489349619737019,
      "loss": 5.0709,
      "mean_token_accuracy": 0.1844639077782631,
      "num_tokens": 23492604.0,
      "step": 13540
    },
    {
      "entropy": 5.644566297531128,
      "epoch": 1.053841665045711,
      "grad_norm": 1.09375,
      "learning_rate": 0.000489341191959447,
      "loss": 5.2725,
      "mean_token_accuracy": 0.18335204273462297,
      "num_tokens": 23501817.0,
      "step": 13545
    },
    {
      "entropy": 5.647152423858643,
      "epoch": 1.0542306944174284,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004893327609295887,
      "loss": 5.1743,
      "mean_token_accuracy": 0.1869557186961174,
      "num_tokens": 23509996.0,
      "step": 13550
    },
    {
      "entropy": 5.493410587310791,
      "epoch": 1.0546197237891461,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004893243266475719,
      "loss": 4.9317,
      "mean_token_accuracy": 0.20967918485403061,
      "num_tokens": 23518316.0,
      "step": 13555
    },
    {
      "entropy": 5.500581550598144,
      "epoch": 1.0550087531608636,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004893158891135245,
      "loss": 5.1489,
      "mean_token_accuracy": 0.18843015730381013,
      "num_tokens": 23526851.0,
      "step": 13560
    },
    {
      "entropy": 5.575391435623169,
      "epoch": 1.055397782532581,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004893074483275746,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19761413782835008,
      "num_tokens": 23534757.0,
      "step": 13565
    },
    {
      "entropy": 5.55781044960022,
      "epoch": 1.0557868119042988,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004892990042898503,
      "loss": 5.1213,
      "mean_token_accuracy": 0.1814389169216156,
      "num_tokens": 23544102.0,
      "step": 13570
    },
    {
      "entropy": 5.526807880401611,
      "epoch": 1.0561758412760163,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004892905570004798,
      "loss": 5.1574,
      "mean_token_accuracy": 0.17815360724925994,
      "num_tokens": 23552663.0,
      "step": 13575
    },
    {
      "entropy": 5.540577507019043,
      "epoch": 1.056564870647734,
      "grad_norm": 1.046875,
      "learning_rate": 0.000489282106459591,
      "loss": 5.0421,
      "mean_token_accuracy": 0.19477053433656694,
      "num_tokens": 23561144.0,
      "step": 13580
    },
    {
      "entropy": 5.485770654678345,
      "epoch": 1.0569539000194514,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004892736526673124,
      "loss": 5.0383,
      "mean_token_accuracy": 0.19195435792207718,
      "num_tokens": 23569833.0,
      "step": 13585
    },
    {
      "entropy": 5.5496666431427,
      "epoch": 1.0573429293911691,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004892651956237721,
      "loss": 5.1038,
      "mean_token_accuracy": 0.1859749212861061,
      "num_tokens": 23578643.0,
      "step": 13590
    },
    {
      "entropy": 5.611403131484986,
      "epoch": 1.0577319587628866,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004892567353290986,
      "loss": 5.171,
      "mean_token_accuracy": 0.18489499092102052,
      "num_tokens": 23587324.0,
      "step": 13595
    },
    {
      "entropy": 5.515933990478516,
      "epoch": 1.058120988134604,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004892482717834201,
      "loss": 5.1094,
      "mean_token_accuracy": 0.18617239743471145,
      "num_tokens": 23595670.0,
      "step": 13600
    },
    {
      "entropy": 5.52309217453003,
      "epoch": 1.0585100175063218,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004892398049868651,
      "loss": 5.0229,
      "mean_token_accuracy": 0.2074963331222534,
      "num_tokens": 23603833.0,
      "step": 13605
    },
    {
      "entropy": 5.542495965957642,
      "epoch": 1.0588990468780393,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000489231334939562,
      "loss": 5.1126,
      "mean_token_accuracy": 0.18682459890842437,
      "num_tokens": 23612815.0,
      "step": 13610
    },
    {
      "entropy": 5.570399570465088,
      "epoch": 1.059288076249757,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004892228616416395,
      "loss": 5.1166,
      "mean_token_accuracy": 0.18690780103206633,
      "num_tokens": 23621178.0,
      "step": 13615
    },
    {
      "entropy": 5.559569931030273,
      "epoch": 1.0596771056214744,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004892143850932259,
      "loss": 5.3037,
      "mean_token_accuracy": 0.18021937012672423,
      "num_tokens": 23630681.0,
      "step": 13620
    },
    {
      "entropy": 5.570667314529419,
      "epoch": 1.060066134993192,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048920590529445,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19716931283473968,
      "num_tokens": 23639374.0,
      "step": 13625
    },
    {
      "entropy": 5.550150299072266,
      "epoch": 1.0604551643649096,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004891974222454406,
      "loss": 5.1074,
      "mean_token_accuracy": 0.19388217777013778,
      "num_tokens": 23648217.0,
      "step": 13630
    },
    {
      "entropy": 5.576891994476318,
      "epoch": 1.060844193736627,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004891889359463261,
      "loss": 5.0819,
      "mean_token_accuracy": 0.1954309433698654,
      "num_tokens": 23656688.0,
      "step": 13635
    },
    {
      "entropy": 5.533861351013184,
      "epoch": 1.0612332231083448,
      "grad_norm": 1.0,
      "learning_rate": 0.0004891804463972354,
      "loss": 5.152,
      "mean_token_accuracy": 0.191353702545166,
      "num_tokens": 23666646.0,
      "step": 13640
    },
    {
      "entropy": 5.5231880187988285,
      "epoch": 1.0616222524800623,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004891719535982974,
      "loss": 5.0516,
      "mean_token_accuracy": 0.19825712740421295,
      "num_tokens": 23675278.0,
      "step": 13645
    },
    {
      "entropy": 5.559094047546386,
      "epoch": 1.0620112818517797,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004891634575496408,
      "loss": 5.169,
      "mean_token_accuracy": 0.1882852330803871,
      "num_tokens": 23684263.0,
      "step": 13650
    },
    {
      "entropy": 5.542915916442871,
      "epoch": 1.0624003112234974,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004891549582513946,
      "loss": 4.9945,
      "mean_token_accuracy": 0.19987936913967133,
      "num_tokens": 23692249.0,
      "step": 13655
    },
    {
      "entropy": 5.5894450664520265,
      "epoch": 1.062789340595215,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000489146455703688,
      "loss": 5.2109,
      "mean_token_accuracy": 0.18558669835329056,
      "num_tokens": 23701082.0,
      "step": 13660
    },
    {
      "entropy": 5.557166004180909,
      "epoch": 1.0631783699669326,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004891379499066495,
      "loss": 5.1697,
      "mean_token_accuracy": 0.17952576279640198,
      "num_tokens": 23710277.0,
      "step": 13665
    },
    {
      "entropy": 5.542935180664062,
      "epoch": 1.06356739933865,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004891294408604087,
      "loss": 5.1395,
      "mean_token_accuracy": 0.18682847768068314,
      "num_tokens": 23719131.0,
      "step": 13670
    },
    {
      "entropy": 5.5245277881622314,
      "epoch": 1.0639564287103676,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004891209285650942,
      "loss": 5.0844,
      "mean_token_accuracy": 0.18916123509407043,
      "num_tokens": 23727674.0,
      "step": 13675
    },
    {
      "entropy": 5.5888899803161625,
      "epoch": 1.0643454580820852,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004891124130208355,
      "loss": 5.2052,
      "mean_token_accuracy": 0.18104732632637024,
      "num_tokens": 23736181.0,
      "step": 13680
    },
    {
      "entropy": 5.561683225631714,
      "epoch": 1.0647344874538027,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004891038942277618,
      "loss": 5.0908,
      "mean_token_accuracy": 0.19476557672023773,
      "num_tokens": 23744552.0,
      "step": 13685
    },
    {
      "entropy": 5.640599250793457,
      "epoch": 1.0651235168255204,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004890953721860022,
      "loss": 5.2097,
      "mean_token_accuracy": 0.18445058166980743,
      "num_tokens": 23753755.0,
      "step": 13690
    },
    {
      "entropy": 5.5744880676269535,
      "epoch": 1.065512546197238,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004890868468956862,
      "loss": 5.1122,
      "mean_token_accuracy": 0.19210155457258224,
      "num_tokens": 23761759.0,
      "step": 13695
    },
    {
      "entropy": 5.5430450439453125,
      "epoch": 1.0659015755689554,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489078318356943,
      "loss": 5.096,
      "mean_token_accuracy": 0.19025285840034484,
      "num_tokens": 23769952.0,
      "step": 13700
    },
    {
      "entropy": 5.516364002227784,
      "epoch": 1.066290604940673,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004890697865699021,
      "loss": 5.1414,
      "mean_token_accuracy": 0.18873618841171264,
      "num_tokens": 23778045.0,
      "step": 13705
    },
    {
      "entropy": 5.560581254959106,
      "epoch": 1.0666796343123905,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004890612515346929,
      "loss": 5.0779,
      "mean_token_accuracy": 0.18887751549482346,
      "num_tokens": 23787477.0,
      "step": 13710
    },
    {
      "entropy": 5.555035829544067,
      "epoch": 1.0670686636841082,
      "grad_norm": 1.125,
      "learning_rate": 0.0004890527132514448,
      "loss": 5.0814,
      "mean_token_accuracy": 0.20074543058872224,
      "num_tokens": 23795905.0,
      "step": 13715
    },
    {
      "entropy": 5.627185678482055,
      "epoch": 1.0674576930558257,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004890441717202876,
      "loss": 5.1546,
      "mean_token_accuracy": 0.1903741791844368,
      "num_tokens": 23804339.0,
      "step": 13720
    },
    {
      "entropy": 5.582160139083863,
      "epoch": 1.0678467224275432,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004890356269413507,
      "loss": 5.1562,
      "mean_token_accuracy": 0.185725799202919,
      "num_tokens": 23813032.0,
      "step": 13725
    },
    {
      "entropy": 5.56419734954834,
      "epoch": 1.068235751799261,
      "grad_norm": 1.078125,
      "learning_rate": 0.000489027078914764,
      "loss": 5.0371,
      "mean_token_accuracy": 0.2049105256795883,
      "num_tokens": 23822047.0,
      "step": 13730
    },
    {
      "entropy": 5.462968206405639,
      "epoch": 1.0686247811709784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004890185276406569,
      "loss": 5.011,
      "mean_token_accuracy": 0.1926239013671875,
      "num_tokens": 23831220.0,
      "step": 13735
    },
    {
      "entropy": 5.540319967269897,
      "epoch": 1.069013810542696,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004890099731191592,
      "loss": 5.17,
      "mean_token_accuracy": 0.18684637397527695,
      "num_tokens": 23840118.0,
      "step": 13740
    },
    {
      "entropy": 5.587307167053223,
      "epoch": 1.0694028399144135,
      "grad_norm": 1.0,
      "learning_rate": 0.000489001415350401,
      "loss": 5.1306,
      "mean_token_accuracy": 0.19016505479812623,
      "num_tokens": 23848826.0,
      "step": 13745
    },
    {
      "entropy": 5.62038516998291,
      "epoch": 1.069791869286131,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004889928543345118,
      "loss": 5.2569,
      "mean_token_accuracy": 0.184573432803154,
      "num_tokens": 23857462.0,
      "step": 13750
    },
    {
      "entropy": 5.540248727798462,
      "epoch": 1.0701808986578487,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004889842900716217,
      "loss": 5.1117,
      "mean_token_accuracy": 0.19019486606121064,
      "num_tokens": 23866842.0,
      "step": 13755
    },
    {
      "entropy": 5.574837350845337,
      "epoch": 1.0705699280295662,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004889757225618606,
      "loss": 5.1394,
      "mean_token_accuracy": 0.18294884860515595,
      "num_tokens": 23875079.0,
      "step": 13760
    },
    {
      "entropy": 5.615604066848755,
      "epoch": 1.0709589574012839,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004889671518053584,
      "loss": 5.1147,
      "mean_token_accuracy": 0.1861308217048645,
      "num_tokens": 23883963.0,
      "step": 13765
    },
    {
      "entropy": 5.589229679107666,
      "epoch": 1.0713479867730014,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004889585778022453,
      "loss": 5.0853,
      "mean_token_accuracy": 0.20030197203159333,
      "num_tokens": 23892709.0,
      "step": 13770
    },
    {
      "entropy": 5.57830696105957,
      "epoch": 1.0717370161447188,
      "grad_norm": 1.125,
      "learning_rate": 0.0004889500005526514,
      "loss": 5.2215,
      "mean_token_accuracy": 0.1834134966135025,
      "num_tokens": 23901124.0,
      "step": 13775
    },
    {
      "entropy": 5.6007640838623045,
      "epoch": 1.0721260455164365,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004889414200567067,
      "loss": 5.1574,
      "mean_token_accuracy": 0.1839173659682274,
      "num_tokens": 23909442.0,
      "step": 13780
    },
    {
      "entropy": 5.5681853771209715,
      "epoch": 1.072515074888154,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004889328363145415,
      "loss": 5.1293,
      "mean_token_accuracy": 0.19386406540870665,
      "num_tokens": 23918143.0,
      "step": 13785
    },
    {
      "entropy": 5.580930900573731,
      "epoch": 1.0729041042598717,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004889242493262859,
      "loss": 5.0733,
      "mean_token_accuracy": 0.1941058561205864,
      "num_tokens": 23926432.0,
      "step": 13790
    },
    {
      "entropy": 5.6241645336151125,
      "epoch": 1.0732931336315892,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004889156590920704,
      "loss": 5.2217,
      "mean_token_accuracy": 0.17931180894374849,
      "num_tokens": 23935362.0,
      "step": 13795
    },
    {
      "entropy": 5.488368797302246,
      "epoch": 1.0736821630033067,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004889070656120253,
      "loss": 4.9684,
      "mean_token_accuracy": 0.198727485537529,
      "num_tokens": 23943926.0,
      "step": 13800
    },
    {
      "entropy": 5.5760266304016115,
      "epoch": 1.0740711923750244,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004888984688862809,
      "loss": 5.135,
      "mean_token_accuracy": 0.1842544123530388,
      "num_tokens": 23952513.0,
      "step": 13805
    },
    {
      "entropy": 5.537561082839966,
      "epoch": 1.0744602217467418,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004888898689149677,
      "loss": 5.1714,
      "mean_token_accuracy": 0.18406044393777848,
      "num_tokens": 23961529.0,
      "step": 13810
    },
    {
      "entropy": 5.676672124862671,
      "epoch": 1.0748492511184595,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004888812656982162,
      "loss": 5.19,
      "mean_token_accuracy": 0.18808944076299666,
      "num_tokens": 23971156.0,
      "step": 13815
    },
    {
      "entropy": 5.6050958156585695,
      "epoch": 1.075238280490177,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004888726592361569,
      "loss": 5.1075,
      "mean_token_accuracy": 0.18496610671281816,
      "num_tokens": 23979676.0,
      "step": 13820
    },
    {
      "entropy": 5.5462747573852536,
      "epoch": 1.0756273098618945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004888640495289204,
      "loss": 5.0621,
      "mean_token_accuracy": 0.19622142910957335,
      "num_tokens": 23987764.0,
      "step": 13825
    },
    {
      "entropy": 5.562421226501465,
      "epoch": 1.0760163392336122,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004888554365766374,
      "loss": 5.108,
      "mean_token_accuracy": 0.18574558049440384,
      "num_tokens": 23995844.0,
      "step": 13830
    },
    {
      "entropy": 5.608630514144897,
      "epoch": 1.0764053686053296,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004888468203794385,
      "loss": 5.1845,
      "mean_token_accuracy": 0.18551842570304872,
      "num_tokens": 24003824.0,
      "step": 13835
    },
    {
      "entropy": 5.561627149581909,
      "epoch": 1.0767943979770473,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004888382009374545,
      "loss": 5.1338,
      "mean_token_accuracy": 0.1885688602924347,
      "num_tokens": 24012471.0,
      "step": 13840
    },
    {
      "entropy": 5.571304273605347,
      "epoch": 1.0771834273487648,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004888295782508161,
      "loss": 5.0982,
      "mean_token_accuracy": 0.1925314575433731,
      "num_tokens": 24021344.0,
      "step": 13845
    },
    {
      "entropy": 5.529728031158447,
      "epoch": 1.0775724567204823,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004888209523196542,
      "loss": 5.0805,
      "mean_token_accuracy": 0.19462160766124725,
      "num_tokens": 24029598.0,
      "step": 13850
    },
    {
      "entropy": 5.506401872634887,
      "epoch": 1.0779614860922,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004888123231440997,
      "loss": 5.1042,
      "mean_token_accuracy": 0.19367956668138503,
      "num_tokens": 24038097.0,
      "step": 13855
    },
    {
      "entropy": 5.55786166191101,
      "epoch": 1.0783505154639175,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004888036907242834,
      "loss": 5.1556,
      "mean_token_accuracy": 0.19234763830900192,
      "num_tokens": 24046812.0,
      "step": 13860
    },
    {
      "entropy": 5.597179555892945,
      "epoch": 1.0787395448356352,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004887950550603366,
      "loss": 5.1237,
      "mean_token_accuracy": 0.18817501217126847,
      "num_tokens": 24054908.0,
      "step": 13865
    },
    {
      "entropy": 5.503034448623657,
      "epoch": 1.0791285742073526,
      "grad_norm": 1.0,
      "learning_rate": 0.0004887864161523901,
      "loss": 5.1082,
      "mean_token_accuracy": 0.19433763027191162,
      "num_tokens": 24063695.0,
      "step": 13870
    },
    {
      "entropy": 5.530128383636475,
      "epoch": 1.0795176035790701,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004887777740005749,
      "loss": 5.1087,
      "mean_token_accuracy": 0.19398031383752823,
      "num_tokens": 24072481.0,
      "step": 13875
    },
    {
      "entropy": 5.5003636360168455,
      "epoch": 1.0799066329507878,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004887691286050224,
      "loss": 5.1208,
      "mean_token_accuracy": 0.194350366294384,
      "num_tokens": 24080751.0,
      "step": 13880
    },
    {
      "entropy": 5.544689130783081,
      "epoch": 1.0802956623225053,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004887604799658635,
      "loss": 5.0922,
      "mean_token_accuracy": 0.19087753295898438,
      "num_tokens": 24088618.0,
      "step": 13885
    },
    {
      "entropy": 5.578266620635986,
      "epoch": 1.080684691694223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004887518280832295,
      "loss": 5.1842,
      "mean_token_accuracy": 0.1857525885105133,
      "num_tokens": 24096754.0,
      "step": 13890
    },
    {
      "entropy": 5.551661491394043,
      "epoch": 1.0810737210659405,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004887431729572519,
      "loss": 5.1186,
      "mean_token_accuracy": 0.1966030180454254,
      "num_tokens": 24104793.0,
      "step": 13895
    },
    {
      "entropy": 5.707340955734253,
      "epoch": 1.081462750437658,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004887345145880617,
      "loss": 5.2533,
      "mean_token_accuracy": 0.17408431321382523,
      "num_tokens": 24113816.0,
      "step": 13900
    },
    {
      "entropy": 5.557224750518799,
      "epoch": 1.0818517798093756,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004887258529757904,
      "loss": 5.0394,
      "mean_token_accuracy": 0.19136426597833633,
      "num_tokens": 24123002.0,
      "step": 13905
    },
    {
      "entropy": 5.581175184249878,
      "epoch": 1.082240809181093,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004887171881205696,
      "loss": 5.0965,
      "mean_token_accuracy": 0.1916172832250595,
      "num_tokens": 24130841.0,
      "step": 13910
    },
    {
      "entropy": 5.579694938659668,
      "epoch": 1.0826298385528108,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004887085200225306,
      "loss": 5.1689,
      "mean_token_accuracy": 0.1794822633266449,
      "num_tokens": 24139236.0,
      "step": 13915
    },
    {
      "entropy": 5.5275708675384525,
      "epoch": 1.0830188679245283,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004886998486818049,
      "loss": 5.027,
      "mean_token_accuracy": 0.19782494753599167,
      "num_tokens": 24147824.0,
      "step": 13920
    },
    {
      "entropy": 5.553222513198852,
      "epoch": 1.083407897296246,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004886911740985242,
      "loss": 5.1548,
      "mean_token_accuracy": 0.1861773267388344,
      "num_tokens": 24156555.0,
      "step": 13925
    },
    {
      "entropy": 5.620675897598266,
      "epoch": 1.0837969266679635,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048868249627282,
      "loss": 5.0816,
      "mean_token_accuracy": 0.19274846464395523,
      "num_tokens": 24164424.0,
      "step": 13930
    },
    {
      "entropy": 5.591138029098511,
      "epoch": 1.084185956039681,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000488673815204824,
      "loss": 5.1033,
      "mean_token_accuracy": 0.19383617043495177,
      "num_tokens": 24173325.0,
      "step": 13935
    },
    {
      "entropy": 5.611918878555298,
      "epoch": 1.0845749854113986,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004886651308946681,
      "loss": 5.2591,
      "mean_token_accuracy": 0.18532256484031678,
      "num_tokens": 24181781.0,
      "step": 13940
    },
    {
      "entropy": 5.58632264137268,
      "epoch": 1.084964014783116,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004886564433424839,
      "loss": 5.0742,
      "mean_token_accuracy": 0.18997456282377242,
      "num_tokens": 24190065.0,
      "step": 13945
    },
    {
      "entropy": 5.585769128799439,
      "epoch": 1.0853530441548336,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004886477525484032,
      "loss": 5.212,
      "mean_token_accuracy": 0.17957908660173416,
      "num_tokens": 24198814.0,
      "step": 13950
    },
    {
      "entropy": 5.560888624191284,
      "epoch": 1.0857420735265513,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004886390585125579,
      "loss": 5.0894,
      "mean_token_accuracy": 0.18675375282764434,
      "num_tokens": 24208008.0,
      "step": 13955
    },
    {
      "entropy": 5.625029468536377,
      "epoch": 1.0861311028982688,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004886303612350799,
      "loss": 5.0937,
      "mean_token_accuracy": 0.1943022146821022,
      "num_tokens": 24216098.0,
      "step": 13960
    },
    {
      "entropy": 5.485174608230591,
      "epoch": 1.0865201322699864,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004886216607161013,
      "loss": 5.0299,
      "mean_token_accuracy": 0.19588403701782225,
      "num_tokens": 24224943.0,
      "step": 13965
    },
    {
      "entropy": 5.528639411926269,
      "epoch": 1.086909161641704,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004886129569557538,
      "loss": 5.0989,
      "mean_token_accuracy": 0.18515908420085908,
      "num_tokens": 24232950.0,
      "step": 13970
    },
    {
      "entropy": 5.625828123092651,
      "epoch": 1.0872981910134216,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004886042499541699,
      "loss": 5.2439,
      "mean_token_accuracy": 0.1840382382273674,
      "num_tokens": 24242355.0,
      "step": 13975
    },
    {
      "entropy": 5.56804723739624,
      "epoch": 1.087687220385139,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004885955397114813,
      "loss": 5.0947,
      "mean_token_accuracy": 0.1971299722790718,
      "num_tokens": 24251247.0,
      "step": 13980
    },
    {
      "entropy": 5.551373243331909,
      "epoch": 1.0880762497568566,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004885868262278205,
      "loss": 5.1746,
      "mean_token_accuracy": 0.18218241333961488,
      "num_tokens": 24260281.0,
      "step": 13985
    },
    {
      "entropy": 5.621750020980835,
      "epoch": 1.0884652791285743,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004885781095033195,
      "loss": 5.1733,
      "mean_token_accuracy": 0.18397517502307892,
      "num_tokens": 24268474.0,
      "step": 13990
    },
    {
      "entropy": 5.635916090011596,
      "epoch": 1.0888543085002917,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004885693895381108,
      "loss": 5.1351,
      "mean_token_accuracy": 0.19216557443141938,
      "num_tokens": 24276897.0,
      "step": 13995
    },
    {
      "entropy": 5.551297521591186,
      "epoch": 1.0892433378720092,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004885606663323263,
      "loss": 5.122,
      "mean_token_accuracy": 0.1863850697875023,
      "num_tokens": 24285892.0,
      "step": 14000
    },
    {
      "entropy": 5.537345504760742,
      "epoch": 1.089632367243727,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004885519398860987,
      "loss": 5.1681,
      "mean_token_accuracy": 0.18805530667304993,
      "num_tokens": 24294141.0,
      "step": 14005
    },
    {
      "entropy": 5.517060804367065,
      "epoch": 1.0900213966154444,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004885432101995604,
      "loss": 5.062,
      "mean_token_accuracy": 0.19237542003393174,
      "num_tokens": 24302803.0,
      "step": 14010
    },
    {
      "entropy": 5.67067322731018,
      "epoch": 1.090410425987162,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004885344772728436,
      "loss": 5.2398,
      "mean_token_accuracy": 0.18113906681537628,
      "num_tokens": 24311755.0,
      "step": 14015
    },
    {
      "entropy": 5.579156446456909,
      "epoch": 1.0907994553588796,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004885257411060812,
      "loss": 5.0567,
      "mean_token_accuracy": 0.1981242999434471,
      "num_tokens": 24320913.0,
      "step": 14020
    },
    {
      "entropy": 5.614050960540771,
      "epoch": 1.0911884847305973,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004885170016994053,
      "loss": 5.2668,
      "mean_token_accuracy": 0.17813585549592972,
      "num_tokens": 24331112.0,
      "step": 14025
    },
    {
      "entropy": 5.5172981262207035,
      "epoch": 1.0915775141023147,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004885082590529489,
      "loss": 5.0573,
      "mean_token_accuracy": 0.19136620163917542,
      "num_tokens": 24339286.0,
      "step": 14030
    },
    {
      "entropy": 5.529982995986939,
      "epoch": 1.0919665434740322,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004884995131668445,
      "loss": 5.074,
      "mean_token_accuracy": 0.18799741417169571,
      "num_tokens": 24347787.0,
      "step": 14035
    },
    {
      "entropy": 5.553857660293579,
      "epoch": 1.09235557284575,
      "grad_norm": 1.125,
      "learning_rate": 0.0004884907640412248,
      "loss": 5.0428,
      "mean_token_accuracy": 0.195489402115345,
      "num_tokens": 24356340.0,
      "step": 14040
    },
    {
      "entropy": 5.5461619853973385,
      "epoch": 1.0927446022174674,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004884820116762225,
      "loss": 5.1257,
      "mean_token_accuracy": 0.18353502899408342,
      "num_tokens": 24365182.0,
      "step": 14045
    },
    {
      "entropy": 5.62374062538147,
      "epoch": 1.093133631589185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004884732560719706,
      "loss": 5.1581,
      "mean_token_accuracy": 0.1798996075987816,
      "num_tokens": 24373455.0,
      "step": 14050
    },
    {
      "entropy": 5.601822280883789,
      "epoch": 1.0935226609609026,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004884644972286017,
      "loss": 5.2276,
      "mean_token_accuracy": 0.18316431790590287,
      "num_tokens": 24382433.0,
      "step": 14055
    },
    {
      "entropy": 5.547022914886474,
      "epoch": 1.09391169033262,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004884557351462488,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19143435209989548,
      "num_tokens": 24390971.0,
      "step": 14060
    },
    {
      "entropy": 5.612474679946899,
      "epoch": 1.0943007197043377,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004884469698250449,
      "loss": 5.1079,
      "mean_token_accuracy": 0.18284114748239516,
      "num_tokens": 24399508.0,
      "step": 14065
    },
    {
      "entropy": 5.532827472686767,
      "epoch": 1.0946897490760552,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000488438201265123,
      "loss": 5.051,
      "mean_token_accuracy": 0.19327640384435654,
      "num_tokens": 24408424.0,
      "step": 14070
    },
    {
      "entropy": 5.491205787658691,
      "epoch": 1.095078778447773,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004884294294666161,
      "loss": 5.1679,
      "mean_token_accuracy": 0.18431971073150635,
      "num_tokens": 24416918.0,
      "step": 14075
    },
    {
      "entropy": 5.537044429779053,
      "epoch": 1.0954678078194904,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004884206544296573,
      "loss": 5.0223,
      "mean_token_accuracy": 0.19488463699817657,
      "num_tokens": 24425779.0,
      "step": 14080
    },
    {
      "entropy": 5.506648254394531,
      "epoch": 1.0958568371912079,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004884118761543797,
      "loss": 5.0387,
      "mean_token_accuracy": 0.18994079381227494,
      "num_tokens": 24434567.0,
      "step": 14085
    },
    {
      "entropy": 5.549871635437012,
      "epoch": 1.0962458665629256,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004884030946409167,
      "loss": 5.1301,
      "mean_token_accuracy": 0.18966175019741058,
      "num_tokens": 24442980.0,
      "step": 14090
    },
    {
      "entropy": 5.537383699417115,
      "epoch": 1.096634895934643,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004883943098894013,
      "loss": 5.059,
      "mean_token_accuracy": 0.18882790356874465,
      "num_tokens": 24452273.0,
      "step": 14095
    },
    {
      "entropy": 5.616352272033692,
      "epoch": 1.0970239253063607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000488385521899967,
      "loss": 5.1594,
      "mean_token_accuracy": 0.19060663729906083,
      "num_tokens": 24461854.0,
      "step": 14100
    },
    {
      "entropy": 5.616894388198853,
      "epoch": 1.0974129546780782,
      "grad_norm": 1.0625,
      "learning_rate": 0.000488376730672747,
      "loss": 5.2052,
      "mean_token_accuracy": 0.1858152389526367,
      "num_tokens": 24470468.0,
      "step": 14105
    },
    {
      "entropy": 5.534484910964966,
      "epoch": 1.0978019840497957,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004883679362078746,
      "loss": 5.0745,
      "mean_token_accuracy": 0.19345633685588837,
      "num_tokens": 24479282.0,
      "step": 14110
    },
    {
      "entropy": 5.629233884811401,
      "epoch": 1.0981910134215134,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004883591385054836,
      "loss": 5.1488,
      "mean_token_accuracy": 0.19297229498624802,
      "num_tokens": 24487331.0,
      "step": 14115
    },
    {
      "entropy": 5.605197095870972,
      "epoch": 1.0985800427932308,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004883503375657072,
      "loss": 5.0553,
      "mean_token_accuracy": 0.19714493453502654,
      "num_tokens": 24496101.0,
      "step": 14120
    },
    {
      "entropy": 5.623606109619141,
      "epoch": 1.0989690721649485,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004883415333886789,
      "loss": 5.154,
      "mean_token_accuracy": 0.1851455107331276,
      "num_tokens": 24504321.0,
      "step": 14125
    },
    {
      "entropy": 5.566852378845215,
      "epoch": 1.099358101536666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004883327259745325,
      "loss": 5.054,
      "mean_token_accuracy": 0.19597918838262557,
      "num_tokens": 24511887.0,
      "step": 14130
    },
    {
      "entropy": 5.549468278884888,
      "epoch": 1.0997471309083835,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004883239153234015,
      "loss": 5.0602,
      "mean_token_accuracy": 0.1902603030204773,
      "num_tokens": 24521188.0,
      "step": 14135
    },
    {
      "entropy": 5.576074504852295,
      "epoch": 1.1001361602801012,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004883151014354197,
      "loss": 5.0593,
      "mean_token_accuracy": 0.19443640857934952,
      "num_tokens": 24530656.0,
      "step": 14140
    },
    {
      "entropy": 5.494708251953125,
      "epoch": 1.1005251896518187,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004883062843107207,
      "loss": 5.0538,
      "mean_token_accuracy": 0.18540543913841248,
      "num_tokens": 24539438.0,
      "step": 14145
    },
    {
      "entropy": 5.562297916412353,
      "epoch": 1.1009142190235364,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004882974639494383,
      "loss": 5.1366,
      "mean_token_accuracy": 0.1859801784157753,
      "num_tokens": 24548005.0,
      "step": 14150
    },
    {
      "entropy": 5.628852224349975,
      "epoch": 1.1013032483952538,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004882886403517065,
      "loss": 5.1877,
      "mean_token_accuracy": 0.18561447858810426,
      "num_tokens": 24556507.0,
      "step": 14155
    },
    {
      "entropy": 5.534324502944946,
      "epoch": 1.1016922777669713,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004882798135176589,
      "loss": 5.1231,
      "mean_token_accuracy": 0.19144049137830735,
      "num_tokens": 24565228.0,
      "step": 14160
    },
    {
      "entropy": 5.533432674407959,
      "epoch": 1.102081307138689,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048827098344742973,
      "loss": 5.0794,
      "mean_token_accuracy": 0.1909134492278099,
      "num_tokens": 24573977.0,
      "step": 14165
    },
    {
      "entropy": 5.512648105621338,
      "epoch": 1.1024703365104065,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004882621501411528,
      "loss": 5.02,
      "mean_token_accuracy": 0.1979286938905716,
      "num_tokens": 24582897.0,
      "step": 14170
    },
    {
      "entropy": 5.532407331466675,
      "epoch": 1.1028593658821242,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004882533135989622,
      "loss": 5.031,
      "mean_token_accuracy": 0.19007385969161988,
      "num_tokens": 24591132.0,
      "step": 14175
    },
    {
      "entropy": 5.494268703460693,
      "epoch": 1.1032483952538417,
      "grad_norm": 1.09375,
      "learning_rate": 0.000488244473820992,
      "loss": 5.0741,
      "mean_token_accuracy": 0.19256412535905837,
      "num_tokens": 24600111.0,
      "step": 14180
    },
    {
      "entropy": 5.511301851272583,
      "epoch": 1.1036374246255591,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048823563080737634,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19070682376623155,
      "num_tokens": 24609229.0,
      "step": 14185
    },
    {
      "entropy": 5.547101640701294,
      "epoch": 1.1040264539972768,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004882267845582493,
      "loss": 5.0289,
      "mean_token_accuracy": 0.19263579696416855,
      "num_tokens": 24618012.0,
      "step": 14190
    },
    {
      "entropy": 5.555389785766602,
      "epoch": 1.1044154833689943,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048821793507374526,
      "loss": 5.0968,
      "mean_token_accuracy": 0.1878509536385536,
      "num_tokens": 24626801.0,
      "step": 14195
    },
    {
      "entropy": 5.610468816757202,
      "epoch": 1.104804512740712,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004882090823539984,
      "loss": 5.1662,
      "mean_token_accuracy": 0.1899107202887535,
      "num_tokens": 24635365.0,
      "step": 14200
    },
    {
      "entropy": 5.546853160858154,
      "epoch": 1.1051935421124295,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004882002263991431,
      "loss": 5.0937,
      "mean_token_accuracy": 0.1900238573551178,
      "num_tokens": 24643745.0,
      "step": 14205
    },
    {
      "entropy": 5.497986125946045,
      "epoch": 1.105582571484147,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048819136720931364,
      "loss": 5.0049,
      "mean_token_accuracy": 0.18967562168836594,
      "num_tokens": 24652760.0,
      "step": 14210
    },
    {
      "entropy": 5.535041666030883,
      "epoch": 1.1059716008558647,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048818250478464457,
      "loss": 5.1166,
      "mean_token_accuracy": 0.19113752394914627,
      "num_tokens": 24661326.0,
      "step": 14215
    },
    {
      "entropy": 5.569070148468017,
      "epoch": 1.1063606302275821,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004881736391252703,
      "loss": 5.0152,
      "mean_token_accuracy": 0.2029775857925415,
      "num_tokens": 24669611.0,
      "step": 14220
    },
    {
      "entropy": 5.584267473220825,
      "epoch": 1.1067496595992998,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004881647702313253,
      "loss": 5.1536,
      "mean_token_accuracy": 0.1834797292947769,
      "num_tokens": 24677780.0,
      "step": 14225
    },
    {
      "entropy": 5.50883297920227,
      "epoch": 1.1071386889710173,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048815589810294427,
      "loss": 5.0188,
      "mean_token_accuracy": 0.1955699771642685,
      "num_tokens": 24685807.0,
      "step": 14230
    },
    {
      "entropy": 5.543837785720825,
      "epoch": 1.1075277183427348,
      "grad_norm": 1.0,
      "learning_rate": 0.00048814702274026174,
      "loss": 5.0901,
      "mean_token_accuracy": 0.1868273437023163,
      "num_tokens": 24694383.0,
      "step": 14235
    },
    {
      "entropy": 5.51952052116394,
      "epoch": 1.1079167477144525,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004881381441434124,
      "loss": 5.0359,
      "mean_token_accuracy": 0.19846682846546174,
      "num_tokens": 24702325.0,
      "step": 14240
    },
    {
      "entropy": 5.617363166809082,
      "epoch": 1.10830577708617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000488129262312531,
      "loss": 5.0643,
      "mean_token_accuracy": 0.19092031121253966,
      "num_tokens": 24710725.0,
      "step": 14245
    },
    {
      "entropy": 5.570712566375732,
      "epoch": 1.1086948064578876,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048812037724775217,
      "loss": 5.1146,
      "mean_token_accuracy": 0.19066088199615477,
      "num_tokens": 24720230.0,
      "step": 14250
    },
    {
      "entropy": 5.540719270706177,
      "epoch": 1.1090838358296051,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004881114889492109,
      "loss": 5.0723,
      "mean_token_accuracy": 0.19068222045898436,
      "num_tokens": 24728896.0,
      "step": 14255
    },
    {
      "entropy": 5.617240476608276,
      "epoch": 1.1094728652013226,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048810259741704197,
      "loss": 5.1538,
      "mean_token_accuracy": 0.1849449783563614,
      "num_tokens": 24737731.0,
      "step": 14260
    },
    {
      "entropy": 5.559365606307983,
      "epoch": 1.1098618945730403,
      "grad_norm": 1.0,
      "learning_rate": 0.0004880937026513803,
      "loss": 5.0182,
      "mean_token_accuracy": 0.2024766504764557,
      "num_tokens": 24746335.0,
      "step": 14265
    },
    {
      "entropy": 5.525782108306885,
      "epoch": 1.1102509239447578,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004880848046523609,
      "loss": 5.0314,
      "mean_token_accuracy": 0.19644611775875093,
      "num_tokens": 24755917.0,
      "step": 14270
    },
    {
      "entropy": 5.5859959602355955,
      "epoch": 1.1106399533164755,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048807590342011875,
      "loss": 5.1782,
      "mean_token_accuracy": 0.18299742490053178,
      "num_tokens": 24764713.0,
      "step": 14275
    },
    {
      "entropy": 5.661219358444214,
      "epoch": 1.111028982688193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048806699895478895,
      "loss": 5.2017,
      "mean_token_accuracy": 0.18422999680042268,
      "num_tokens": 24773464.0,
      "step": 14280
    },
    {
      "entropy": 5.59569935798645,
      "epoch": 1.1114180120599104,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004880580912565065,
      "loss": 5.1056,
      "mean_token_accuracy": 0.19189320802688598,
      "num_tokens": 24781437.0,
      "step": 14285
    },
    {
      "entropy": 5.471013593673706,
      "epoch": 1.1118070414316281,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004880491803254067,
      "loss": 4.9623,
      "mean_token_accuracy": 0.20401892960071563,
      "num_tokens": 24789505.0,
      "step": 14290
    },
    {
      "entropy": 5.582314920425415,
      "epoch": 1.1121960708033456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004880402661616246,
      "loss": 5.1777,
      "mean_token_accuracy": 0.18516246229410172,
      "num_tokens": 24798492.0,
      "step": 14295
    },
    {
      "entropy": 5.582237386703492,
      "epoch": 1.1125851001750633,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004880313487652956,
      "loss": 5.1427,
      "mean_token_accuracy": 0.18301793336868286,
      "num_tokens": 24807472.0,
      "step": 14300
    },
    {
      "entropy": 5.621257877349853,
      "epoch": 1.1129741295467808,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048802242813655503,
      "loss": 5.1447,
      "mean_token_accuracy": 0.1889468625187874,
      "num_tokens": 24815838.0,
      "step": 14305
    },
    {
      "entropy": 5.535931158065796,
      "epoch": 1.1133631589184985,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000488013504275538,
      "loss": 5.0308,
      "mean_token_accuracy": 0.20103780925273895,
      "num_tokens": 24823953.0,
      "step": 14310
    },
    {
      "entropy": 5.493774938583374,
      "epoch": 1.113752188290216,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004880045771823802,
      "loss": 4.9757,
      "mean_token_accuracy": 0.20277907103300094,
      "num_tokens": 24832414.0,
      "step": 14315
    },
    {
      "entropy": 5.5106306076049805,
      "epoch": 1.1141412176619334,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048799564685721695,
      "loss": 5.105,
      "mean_token_accuracy": 0.19672067314386368,
      "num_tokens": 24840503.0,
      "step": 14320
    },
    {
      "entropy": 5.58501615524292,
      "epoch": 1.1145302470336511,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004879867133001837,
      "loss": 5.2269,
      "mean_token_accuracy": 0.1800301879644394,
      "num_tokens": 24849272.0,
      "step": 14325
    },
    {
      "entropy": 5.645293569564819,
      "epoch": 1.1149192764053686,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004879777765114162,
      "loss": 5.1811,
      "mean_token_accuracy": 0.18523346483707429,
      "num_tokens": 24858565.0,
      "step": 14330
    },
    {
      "entropy": 5.595496988296508,
      "epoch": 1.115308305777086,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048796883649104996,
      "loss": 5.1225,
      "mean_token_accuracy": 0.19080182015895844,
      "num_tokens": 24867811.0,
      "step": 14335
    },
    {
      "entropy": 5.551525211334228,
      "epoch": 1.1156973351488038,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004879598932392206,
      "loss": 5.1684,
      "mean_token_accuracy": 0.18083980083465576,
      "num_tokens": 24876776.0,
      "step": 14340
    },
    {
      "entropy": 5.563112735748291,
      "epoch": 1.1160863645205212,
      "grad_norm": 1.0,
      "learning_rate": 0.00048795094675606364,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19113542437553405,
      "num_tokens": 24885030.0,
      "step": 14345
    },
    {
      "entropy": 5.628605318069458,
      "epoch": 1.116475393892239,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004879419970417152,
      "loss": 5.2177,
      "mean_token_accuracy": 0.18255677670240403,
      "num_tokens": 24894336.0,
      "step": 14350
    },
    {
      "entropy": 5.632161855697632,
      "epoch": 1.1168644232639564,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048793304409631077,
      "loss": 5.1768,
      "mean_token_accuracy": 0.1853221356868744,
      "num_tokens": 24903107.0,
      "step": 14355
    },
    {
      "entropy": 5.688526201248169,
      "epoch": 1.117253452635674,
      "grad_norm": 1.0,
      "learning_rate": 0.0004879240879199863,
      "loss": 5.2455,
      "mean_token_accuracy": 0.18908555656671525,
      "num_tokens": 24912727.0,
      "step": 14360
    },
    {
      "entropy": 5.521405267715454,
      "epoch": 1.1176424820073916,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048791512851287786,
      "loss": 5.0194,
      "mean_token_accuracy": 0.192735193669796,
      "num_tokens": 24921534.0,
      "step": 14365
    },
    {
      "entropy": 5.5539459705352785,
      "epoch": 1.118031511379109,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048790616587512107,
      "loss": 5.1014,
      "mean_token_accuracy": 0.18986730128526688,
      "num_tokens": 24930093.0,
      "step": 14370
    },
    {
      "entropy": 5.610888242721558,
      "epoch": 1.1184205407508268,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004878972000068521,
      "loss": 5.1886,
      "mean_token_accuracy": 0.19122879803180695,
      "num_tokens": 24939230.0,
      "step": 14375
    },
    {
      "entropy": 5.621839952468872,
      "epoch": 1.1188095701225442,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048788823090820707,
      "loss": 5.2321,
      "mean_token_accuracy": 0.19314769208431243,
      "num_tokens": 24947595.0,
      "step": 14380
    },
    {
      "entropy": 5.619896697998047,
      "epoch": 1.1191985994942617,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048787925857932194,
      "loss": 5.1861,
      "mean_token_accuracy": 0.18565509021282195,
      "num_tokens": 24957099.0,
      "step": 14385
    },
    {
      "entropy": 5.55630464553833,
      "epoch": 1.1195876288659794,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048787028302033287,
      "loss": 5.0709,
      "mean_token_accuracy": 0.18945772796869279,
      "num_tokens": 24965701.0,
      "step": 14390
    },
    {
      "entropy": 5.638362216949463,
      "epoch": 1.1199766582376969,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048786130423137606,
      "loss": 5.1916,
      "mean_token_accuracy": 0.1828547954559326,
      "num_tokens": 24974025.0,
      "step": 14395
    },
    {
      "entropy": 5.664219856262207,
      "epoch": 1.1203656876094146,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004878523222125879,
      "loss": 5.2263,
      "mean_token_accuracy": 0.18391305655241014,
      "num_tokens": 24982930.0,
      "step": 14400
    },
    {
      "entropy": 5.615021991729736,
      "epoch": 1.120754716981132,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004878433369641044,
      "loss": 5.0937,
      "mean_token_accuracy": 0.1945231392979622,
      "num_tokens": 24991332.0,
      "step": 14405
    },
    {
      "entropy": 5.56526026725769,
      "epoch": 1.1211437463528497,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004878343484860621,
      "loss": 5.0918,
      "mean_token_accuracy": 0.19387697875499726,
      "num_tokens": 25000130.0,
      "step": 14410
    },
    {
      "entropy": 5.546545886993409,
      "epoch": 1.1215327757245672,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004878253567785972,
      "loss": 5.108,
      "mean_token_accuracy": 0.19131711423397063,
      "num_tokens": 25008423.0,
      "step": 14415
    },
    {
      "entropy": 5.634926509857178,
      "epoch": 1.1219218050962847,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048781636184184644,
      "loss": 5.1681,
      "mean_token_accuracy": 0.18848209381103515,
      "num_tokens": 25017399.0,
      "step": 14420
    },
    {
      "entropy": 5.649090909957886,
      "epoch": 1.1223108344680024,
      "grad_norm": 1.03125,
      "learning_rate": 0.000487807363675946,
      "loss": 5.1544,
      "mean_token_accuracy": 0.1925899416208267,
      "num_tokens": 25025957.0,
      "step": 14425
    },
    {
      "entropy": 5.600171184539795,
      "epoch": 1.1226998638397199,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004877983622810326,
      "loss": 5.0783,
      "mean_token_accuracy": 0.19370086640119552,
      "num_tokens": 25034246.0,
      "step": 14430
    },
    {
      "entropy": 5.523709964752197,
      "epoch": 1.1230888932114376,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004877893576572427,
      "loss": 5.083,
      "mean_token_accuracy": 0.19385031163692473,
      "num_tokens": 25042461.0,
      "step": 14435
    },
    {
      "entropy": 5.569584846496582,
      "epoch": 1.123477922583155,
      "grad_norm": 1.03125,
      "learning_rate": 0.000487780349804713,
      "loss": 5.0519,
      "mean_token_accuracy": 0.19904048144817352,
      "num_tokens": 25051291.0,
      "step": 14440
    },
    {
      "entropy": 5.587459659576416,
      "epoch": 1.1238669519548725,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004877713387235802,
      "loss": 5.1772,
      "mean_token_accuracy": 0.18640127629041672,
      "num_tokens": 25060538.0,
      "step": 14445
    },
    {
      "entropy": 5.608622407913208,
      "epoch": 1.1242559813265902,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000487762324413981,
      "loss": 5.1562,
      "mean_token_accuracy": 0.18106972724199294,
      "num_tokens": 25068965.0,
      "step": 14450
    },
    {
      "entropy": 5.610302543640136,
      "epoch": 1.1246450106983077,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048775330687605213,
      "loss": 5.1423,
      "mean_token_accuracy": 0.18641403913497925,
      "num_tokens": 25077021.0,
      "step": 14455
    },
    {
      "entropy": 5.569644498825073,
      "epoch": 1.1250340400700254,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004877442861099305,
      "loss": 5.1383,
      "mean_token_accuracy": 0.19138455837965013,
      "num_tokens": 25086009.0,
      "step": 14460
    },
    {
      "entropy": 5.501922464370727,
      "epoch": 1.1254230694417429,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004877352621157529,
      "loss": 5.06,
      "mean_token_accuracy": 0.1943339228630066,
      "num_tokens": 25094745.0,
      "step": 14465
    },
    {
      "entropy": 5.564437532424927,
      "epoch": 1.1258120988134603,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048772623489365635,
      "loss": 5.0715,
      "mean_token_accuracy": 0.18664079457521437,
      "num_tokens": 25103087.0,
      "step": 14470
    },
    {
      "entropy": 5.6123435497283936,
      "epoch": 1.126201128185178,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004877172044437777,
      "loss": 5.1098,
      "mean_token_accuracy": 0.18710518032312393,
      "num_tokens": 25111822.0,
      "step": 14475
    },
    {
      "entropy": 5.572814893722534,
      "epoch": 1.1265901575568955,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048770817076625416,
      "loss": 5.0947,
      "mean_token_accuracy": 0.19054839611053467,
      "num_tokens": 25119975.0,
      "step": 14480
    },
    {
      "entropy": 5.5661180973052975,
      "epoch": 1.126979186928613,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048769913386122253,
      "loss": 5.1575,
      "mean_token_accuracy": 0.18483465164899826,
      "num_tokens": 25129031.0,
      "step": 14485
    },
    {
      "entropy": 5.5998358726501465,
      "epoch": 1.1273682163003307,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004876900937288202,
      "loss": 5.0812,
      "mean_token_accuracy": 0.1930053025484085,
      "num_tokens": 25138230.0,
      "step": 14490
    },
    {
      "entropy": 5.630468511581421,
      "epoch": 1.1277572456720482,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048768105036918426,
      "loss": 5.1217,
      "mean_token_accuracy": 0.18948798328638078,
      "num_tokens": 25147120.0,
      "step": 14495
    },
    {
      "entropy": 5.574945020675659,
      "epoch": 1.1281462750437659,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048767200378245187,
      "loss": 5.1208,
      "mean_token_accuracy": 0.1898228719830513,
      "num_tokens": 25155623.0,
      "step": 14500
    },
    {
      "entropy": 5.4905647277832035,
      "epoch": 1.1285353044154833,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048766295396876025,
      "loss": 5.0076,
      "mean_token_accuracy": 0.19737488478422166,
      "num_tokens": 25163867.0,
      "step": 14505
    },
    {
      "entropy": 5.560219097137451,
      "epoch": 1.128924333787201,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048765390092824683,
      "loss": 5.1268,
      "mean_token_accuracy": 0.19247578084468842,
      "num_tokens": 25172838.0,
      "step": 14510
    },
    {
      "entropy": 5.551562786102295,
      "epoch": 1.1293133631589185,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000487644844661049,
      "loss": 5.0555,
      "mean_token_accuracy": 0.19907177835702897,
      "num_tokens": 25181168.0,
      "step": 14515
    },
    {
      "entropy": 5.569566059112549,
      "epoch": 1.129702392530636,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000487635785167304,
      "loss": 5.1521,
      "mean_token_accuracy": 0.1940109983086586,
      "num_tokens": 25190115.0,
      "step": 14520
    },
    {
      "entropy": 5.578721714019776,
      "epoch": 1.1300914219023537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048762672244714953,
      "loss": 5.1406,
      "mean_token_accuracy": 0.19143279492855073,
      "num_tokens": 25199594.0,
      "step": 14525
    },
    {
      "entropy": 5.60958251953125,
      "epoch": 1.1304804512740712,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004876176565007229,
      "loss": 5.1136,
      "mean_token_accuracy": 0.1821112722158432,
      "num_tokens": 25208060.0,
      "step": 14530
    },
    {
      "entropy": 5.574132633209229,
      "epoch": 1.1308694806457888,
      "grad_norm": 1.0,
      "learning_rate": 0.00048760858732816174,
      "loss": 5.0652,
      "mean_token_accuracy": 0.18881643563508987,
      "num_tokens": 25216257.0,
      "step": 14535
    },
    {
      "entropy": 5.541465616226196,
      "epoch": 1.1312585100175063,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004875995149296037,
      "loss": 5.087,
      "mean_token_accuracy": 0.18855876475572586,
      "num_tokens": 25225032.0,
      "step": 14540
    },
    {
      "entropy": 5.584880208969116,
      "epoch": 1.1316475393892238,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004875904393051864,
      "loss": 5.1337,
      "mean_token_accuracy": 0.19188373386859894,
      "num_tokens": 25233624.0,
      "step": 14545
    },
    {
      "entropy": 5.533587455749512,
      "epoch": 1.1320365687609415,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004875813604550476,
      "loss": 5.1012,
      "mean_token_accuracy": 0.18869900852441787,
      "num_tokens": 25242132.0,
      "step": 14550
    },
    {
      "entropy": 5.584107303619385,
      "epoch": 1.132425598132659,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048757227837932494,
      "loss": 5.0851,
      "mean_token_accuracy": 0.19440909922122956,
      "num_tokens": 25250923.0,
      "step": 14555
    },
    {
      "entropy": 5.575410795211792,
      "epoch": 1.1328146275043767,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004875631930781563,
      "loss": 5.098,
      "mean_token_accuracy": 0.18965144455432892,
      "num_tokens": 25259783.0,
      "step": 14560
    },
    {
      "entropy": 5.655840301513672,
      "epoch": 1.1332036568760941,
      "grad_norm": 1.0,
      "learning_rate": 0.00048755410455167967,
      "loss": 5.286,
      "mean_token_accuracy": 0.1829624816775322,
      "num_tokens": 25268801.0,
      "step": 14565
    },
    {
      "entropy": 5.567112684249878,
      "epoch": 1.1335926862478116,
      "grad_norm": 1.0,
      "learning_rate": 0.0004875450128000326,
      "loss": 5.0975,
      "mean_token_accuracy": 0.20123474895954133,
      "num_tokens": 25277226.0,
      "step": 14570
    },
    {
      "entropy": 5.580487775802612,
      "epoch": 1.1339817156195293,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004875359178233534,
      "loss": 5.1523,
      "mean_token_accuracy": 0.182919679582119,
      "num_tokens": 25285887.0,
      "step": 14575
    },
    {
      "entropy": 5.517343759536743,
      "epoch": 1.1343707449912468,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048752681962177985,
      "loss": 5.0093,
      "mean_token_accuracy": 0.19050014466047288,
      "num_tokens": 25294025.0,
      "step": 14580
    },
    {
      "entropy": 5.6035301208496096,
      "epoch": 1.1347597743629645,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048751771819545013,
      "loss": 5.2135,
      "mean_token_accuracy": 0.18079658150672911,
      "num_tokens": 25303016.0,
      "step": 14585
    },
    {
      "entropy": 5.54065465927124,
      "epoch": 1.135148803734682,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048750861354450236,
      "loss": 5.0172,
      "mean_token_accuracy": 0.19706076085567475,
      "num_tokens": 25311193.0,
      "step": 14590
    },
    {
      "entropy": 5.550218296051026,
      "epoch": 1.1355378331063994,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004874995056690746,
      "loss": 5.0666,
      "mean_token_accuracy": 0.19193121343851088,
      "num_tokens": 25319380.0,
      "step": 14595
    },
    {
      "entropy": 5.535642480850219,
      "epoch": 1.1359268624781171,
      "grad_norm": 1.09375,
      "learning_rate": 0.000487490394569305,
      "loss": 5.08,
      "mean_token_accuracy": 0.19244492948055267,
      "num_tokens": 25327854.0,
      "step": 14600
    },
    {
      "entropy": 5.582655620574951,
      "epoch": 1.1363158918498346,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004874812802453319,
      "loss": 5.1514,
      "mean_token_accuracy": 0.18457458466291427,
      "num_tokens": 25335912.0,
      "step": 14605
    },
    {
      "entropy": 5.563676929473877,
      "epoch": 1.1367049212215523,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004874721626972936,
      "loss": 5.1668,
      "mean_token_accuracy": 0.19149231761693955,
      "num_tokens": 25344104.0,
      "step": 14610
    },
    {
      "entropy": 5.452754831314087,
      "epoch": 1.1370939505932698,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004874630419253284,
      "loss": 4.9252,
      "mean_token_accuracy": 0.20484225004911422,
      "num_tokens": 25352583.0,
      "step": 14615
    },
    {
      "entropy": 5.541375732421875,
      "epoch": 1.1374829799649873,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048745391792957474,
      "loss": 5.0434,
      "mean_token_accuracy": 0.19370035380125045,
      "num_tokens": 25361035.0,
      "step": 14620
    },
    {
      "entropy": 5.563179588317871,
      "epoch": 1.137872009336705,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048744479071017097,
      "loss": 5.0871,
      "mean_token_accuracy": 0.18786347210407256,
      "num_tokens": 25369511.0,
      "step": 14625
    },
    {
      "entropy": 5.586024761199951,
      "epoch": 1.1382610387084224,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004874356602672556,
      "loss": 5.0854,
      "mean_token_accuracy": 0.1938834547996521,
      "num_tokens": 25378023.0,
      "step": 14630
    },
    {
      "entropy": 5.580712842941284,
      "epoch": 1.1386500680801401,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004874265266009673,
      "loss": 5.117,
      "mean_token_accuracy": 0.19111802577972412,
      "num_tokens": 25386791.0,
      "step": 14635
    },
    {
      "entropy": 5.606944227218628,
      "epoch": 1.1390390974518576,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004874173897114445,
      "loss": 5.1464,
      "mean_token_accuracy": 0.18752903640270233,
      "num_tokens": 25395401.0,
      "step": 14640
    },
    {
      "entropy": 5.575720643997192,
      "epoch": 1.1394281268235753,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048740824959882603,
      "loss": 5.1135,
      "mean_token_accuracy": 0.18792559206485748,
      "num_tokens": 25404193.0,
      "step": 14645
    },
    {
      "entropy": 5.639514636993408,
      "epoch": 1.1398171561952928,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004873991062632504,
      "loss": 5.1714,
      "mean_token_accuracy": 0.18551381379365922,
      "num_tokens": 25413073.0,
      "step": 14650
    },
    {
      "entropy": 5.546651935577392,
      "epoch": 1.1402061855670103,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048738995970485635,
      "loss": 5.0803,
      "mean_token_accuracy": 0.19260105192661287,
      "num_tokens": 25422064.0,
      "step": 14655
    },
    {
      "entropy": 5.52317624092102,
      "epoch": 1.140595214938728,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004873808099237827,
      "loss": 5.0614,
      "mean_token_accuracy": 0.19399910420179367,
      "num_tokens": 25430711.0,
      "step": 14660
    },
    {
      "entropy": 5.608675146102906,
      "epoch": 1.1409842443104454,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004873716569201684,
      "loss": 5.1719,
      "mean_token_accuracy": 0.1873139202594757,
      "num_tokens": 25439527.0,
      "step": 14665
    },
    {
      "entropy": 5.634265327453614,
      "epoch": 1.141373273682163,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048736250069415213,
      "loss": 5.1565,
      "mean_token_accuracy": 0.18710326701402663,
      "num_tokens": 25448633.0,
      "step": 14670
    },
    {
      "entropy": 5.564380216598511,
      "epoch": 1.1417623030538806,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004873533412458729,
      "loss": 5.1007,
      "mean_token_accuracy": 0.19030878245830535,
      "num_tokens": 25457657.0,
      "step": 14675
    },
    {
      "entropy": 5.5843555450439455,
      "epoch": 1.142151332425598,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048734417857546986,
      "loss": 5.1574,
      "mean_token_accuracy": 0.18433075696229934,
      "num_tokens": 25466698.0,
      "step": 14680
    },
    {
      "entropy": 5.59642915725708,
      "epoch": 1.1425403617973158,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004873350126830818,
      "loss": 5.1226,
      "mean_token_accuracy": 0.18787850588560104,
      "num_tokens": 25474921.0,
      "step": 14685
    },
    {
      "entropy": 5.551491212844849,
      "epoch": 1.1429293911690332,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004873258435688479,
      "loss": 5.02,
      "mean_token_accuracy": 0.19826652556657792,
      "num_tokens": 25483553.0,
      "step": 14690
    },
    {
      "entropy": 5.591440629959107,
      "epoch": 1.143318420540751,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004873166712329073,
      "loss": 5.0939,
      "mean_token_accuracy": 0.19593354761600495,
      "num_tokens": 25492408.0,
      "step": 14695
    },
    {
      "entropy": 5.541330003738404,
      "epoch": 1.1437074499124684,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048730749567539914,
      "loss": 5.1244,
      "mean_token_accuracy": 0.18857344835996628,
      "num_tokens": 25501404.0,
      "step": 14700
    },
    {
      "entropy": 5.667981386184692,
      "epoch": 1.144096479284186,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048729831689646257,
      "loss": 5.1248,
      "mean_token_accuracy": 0.18747548907995223,
      "num_tokens": 25510176.0,
      "step": 14705
    },
    {
      "entropy": 5.521900749206543,
      "epoch": 1.1444855086559036,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048728913489623705,
      "loss": 5.041,
      "mean_token_accuracy": 0.19366063326597213,
      "num_tokens": 25518809.0,
      "step": 14710
    },
    {
      "entropy": 5.551668262481689,
      "epoch": 1.144874538027621,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004872799496748618,
      "loss": 5.1532,
      "mean_token_accuracy": 0.19095004498958587,
      "num_tokens": 25527415.0,
      "step": 14715
    },
    {
      "entropy": 5.5533545970916744,
      "epoch": 1.1452635673993385,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004872707612324761,
      "loss": 5.1018,
      "mean_token_accuracy": 0.18428247421979904,
      "num_tokens": 25535618.0,
      "step": 14720
    },
    {
      "entropy": 5.580552387237549,
      "epoch": 1.1456525967710562,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004872615695692196,
      "loss": 5.0831,
      "mean_token_accuracy": 0.19384828060865403,
      "num_tokens": 25544228.0,
      "step": 14725
    },
    {
      "entropy": 5.616612195968628,
      "epoch": 1.1460416261427737,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004872523746852315,
      "loss": 5.1458,
      "mean_token_accuracy": 0.1881529062986374,
      "num_tokens": 25553557.0,
      "step": 14730
    },
    {
      "entropy": 5.554841327667236,
      "epoch": 1.1464306555144914,
      "grad_norm": 1.0,
      "learning_rate": 0.00048724317658065146,
      "loss": 5.0768,
      "mean_token_accuracy": 0.1937004417181015,
      "num_tokens": 25562018.0,
      "step": 14735
    },
    {
      "entropy": 5.597393321990967,
      "epoch": 1.146819684886209,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048723397525561916,
      "loss": 5.1132,
      "mean_token_accuracy": 0.19164858311414718,
      "num_tokens": 25569382.0,
      "step": 14740
    },
    {
      "entropy": 5.48429045677185,
      "epoch": 1.1472087142579266,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048722477071027394,
      "loss": 5.0543,
      "mean_token_accuracy": 0.1927467867732048,
      "num_tokens": 25578172.0,
      "step": 14745
    },
    {
      "entropy": 5.522347116470337,
      "epoch": 1.147597743629644,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048721556294475574,
      "loss": 5.0996,
      "mean_token_accuracy": 0.18739671856164933,
      "num_tokens": 25587572.0,
      "step": 14750
    },
    {
      "entropy": 5.575719165802002,
      "epoch": 1.1479867730013615,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048720635195920403,
      "loss": 5.1471,
      "mean_token_accuracy": 0.19230797588825227,
      "num_tokens": 25596609.0,
      "step": 14755
    },
    {
      "entropy": 5.567278146743774,
      "epoch": 1.1483758023730792,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004871971377537588,
      "loss": 5.1423,
      "mean_token_accuracy": 0.18543420732021332,
      "num_tokens": 25605707.0,
      "step": 14760
    },
    {
      "entropy": 5.57260684967041,
      "epoch": 1.1487648317447967,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004871879203285597,
      "loss": 5.1156,
      "mean_token_accuracy": 0.19459976702928544,
      "num_tokens": 25614218.0,
      "step": 14765
    },
    {
      "entropy": 5.5321691036224365,
      "epoch": 1.1491538611165142,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048717869968374655,
      "loss": 5.0297,
      "mean_token_accuracy": 0.1946216568350792,
      "num_tokens": 25622613.0,
      "step": 14770
    },
    {
      "entropy": 5.664019012451172,
      "epoch": 1.1495428904882319,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048716947581945954,
      "loss": 5.1729,
      "mean_token_accuracy": 0.19284721463918686,
      "num_tokens": 25631396.0,
      "step": 14775
    },
    {
      "entropy": 5.569239282608033,
      "epoch": 1.1499319198599494,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004871602487358383,
      "loss": 5.0749,
      "mean_token_accuracy": 0.1919866144657135,
      "num_tokens": 25640870.0,
      "step": 14780
    },
    {
      "entropy": 5.560152864456176,
      "epoch": 1.150320949231667,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000487151018433023,
      "loss": 5.1432,
      "mean_token_accuracy": 0.18698510825634002,
      "num_tokens": 25649395.0,
      "step": 14785
    },
    {
      "entropy": 5.61124620437622,
      "epoch": 1.1507099786033845,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048714178491115367,
      "loss": 5.2358,
      "mean_token_accuracy": 0.1810119032859802,
      "num_tokens": 25659087.0,
      "step": 14790
    },
    {
      "entropy": 5.641227674484253,
      "epoch": 1.1510990079751022,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048713254817037046,
      "loss": 5.1564,
      "mean_token_accuracy": 0.1962727501988411,
      "num_tokens": 25667847.0,
      "step": 14795
    },
    {
      "entropy": 5.6062705516815186,
      "epoch": 1.1514880373468197,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004871233082108135,
      "loss": 5.106,
      "mean_token_accuracy": 0.19782232046127318,
      "num_tokens": 25676786.0,
      "step": 14800
    },
    {
      "entropy": 5.605106401443481,
      "epoch": 1.1518770667185372,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004871140650326228,
      "loss": 5.1476,
      "mean_token_accuracy": 0.1856660410761833,
      "num_tokens": 25685223.0,
      "step": 14805
    },
    {
      "entropy": 5.559368562698364,
      "epoch": 1.1522660960902549,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004871048186359389,
      "loss": 5.1032,
      "mean_token_accuracy": 0.18552485704421998,
      "num_tokens": 25694243.0,
      "step": 14810
    },
    {
      "entropy": 5.5484109878540036,
      "epoch": 1.1526551254619724,
      "grad_norm": 1.0625,
      "learning_rate": 0.000487095569020902,
      "loss": 4.9898,
      "mean_token_accuracy": 0.19948240220546723,
      "num_tokens": 25702735.0,
      "step": 14815
    },
    {
      "entropy": 5.526841926574707,
      "epoch": 1.1530441548336898,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004870863161876524,
      "loss": 5.0639,
      "mean_token_accuracy": 0.19685641378164292,
      "num_tokens": 25711577.0,
      "step": 14820
    },
    {
      "entropy": 5.489689493179322,
      "epoch": 1.1534331842054075,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004870770601363305,
      "loss": 4.9949,
      "mean_token_accuracy": 0.20101726055145264,
      "num_tokens": 25719927.0,
      "step": 14825
    },
    {
      "entropy": 5.541572952270508,
      "epoch": 1.153822213577125,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048706780086707675,
      "loss": 5.0797,
      "mean_token_accuracy": 0.19204696416854858,
      "num_tokens": 25728144.0,
      "step": 14830
    },
    {
      "entropy": 5.552884578704834,
      "epoch": 1.1542112429488427,
      "grad_norm": 1.125,
      "learning_rate": 0.0004870585383800317,
      "loss": 5.0184,
      "mean_token_accuracy": 0.19972701370716095,
      "num_tokens": 25736736.0,
      "step": 14835
    },
    {
      "entropy": 5.6133159637451175,
      "epoch": 1.1546002723205602,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048704927267533585,
      "loss": 5.1934,
      "mean_token_accuracy": 0.19172807186841964,
      "num_tokens": 25745031.0,
      "step": 14840
    },
    {
      "entropy": 5.633868312835693,
      "epoch": 1.1549893016922779,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048704000375312975,
      "loss": 5.1924,
      "mean_token_accuracy": 0.19258706420660018,
      "num_tokens": 25754265.0,
      "step": 14845
    },
    {
      "entropy": 5.642695093154908,
      "epoch": 1.1553783310639953,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048703073161355414,
      "loss": 5.177,
      "mean_token_accuracy": 0.17840857654809952,
      "num_tokens": 25763084.0,
      "step": 14850
    },
    {
      "entropy": 5.656079578399658,
      "epoch": 1.1557673604357128,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004870214562567497,
      "loss": 5.1534,
      "mean_token_accuracy": 0.18367359340190886,
      "num_tokens": 25771626.0,
      "step": 14855
    },
    {
      "entropy": 5.564869213104248,
      "epoch": 1.1561563898074305,
      "grad_norm": 1.125,
      "learning_rate": 0.000487012177682857,
      "loss": 5.1143,
      "mean_token_accuracy": 0.18735800236463546,
      "num_tokens": 25780326.0,
      "step": 14860
    },
    {
      "entropy": 5.624983549118042,
      "epoch": 1.156545419179148,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004870028958920171,
      "loss": 5.2342,
      "mean_token_accuracy": 0.1776147499680519,
      "num_tokens": 25789318.0,
      "step": 14865
    },
    {
      "entropy": 5.592816352844238,
      "epoch": 1.1569344485508655,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004869936108843706,
      "loss": 5.1929,
      "mean_token_accuracy": 0.18489024490118028,
      "num_tokens": 25798025.0,
      "step": 14870
    },
    {
      "entropy": 5.615821504592896,
      "epoch": 1.1573234779225832,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048698432266005854,
      "loss": 5.1002,
      "mean_token_accuracy": 0.19362777918577195,
      "num_tokens": 25806669.0,
      "step": 14875
    },
    {
      "entropy": 5.618700551986694,
      "epoch": 1.1577125072943006,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004869750312192217,
      "loss": 5.0887,
      "mean_token_accuracy": 0.19141381680965425,
      "num_tokens": 25815938.0,
      "step": 14880
    },
    {
      "entropy": 5.552509355545044,
      "epoch": 1.1581015366660183,
      "grad_norm": 1.0,
      "learning_rate": 0.00048696573656200123,
      "loss": 5.0467,
      "mean_token_accuracy": 0.19463123083114625,
      "num_tokens": 25824293.0,
      "step": 14885
    },
    {
      "entropy": 5.640071105957031,
      "epoch": 1.1584905660377358,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048695643868853806,
      "loss": 5.1488,
      "mean_token_accuracy": 0.18800119161605836,
      "num_tokens": 25833081.0,
      "step": 14890
    },
    {
      "entropy": 5.630659198760986,
      "epoch": 1.1588795954094535,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004869471375989733,
      "loss": 5.1858,
      "mean_token_accuracy": 0.18351871520280838,
      "num_tokens": 25841344.0,
      "step": 14895
    },
    {
      "entropy": 5.59793872833252,
      "epoch": 1.159268624781171,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004869378332934481,
      "loss": 5.1557,
      "mean_token_accuracy": 0.18592580854892732,
      "num_tokens": 25850617.0,
      "step": 14900
    },
    {
      "entropy": 5.564720201492309,
      "epoch": 1.1596576541528885,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048692852577210356,
      "loss": 5.1403,
      "mean_token_accuracy": 0.1918509930372238,
      "num_tokens": 25858569.0,
      "step": 14905
    },
    {
      "entropy": 5.53714919090271,
      "epoch": 1.1600466835246062,
      "grad_norm": 1.046875,
      "learning_rate": 0.000486919215035081,
      "loss": 5.1155,
      "mean_token_accuracy": 0.19092709422111512,
      "num_tokens": 25866892.0,
      "step": 14910
    },
    {
      "entropy": 5.631830263137817,
      "epoch": 1.1604357128963236,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004869099010825217,
      "loss": 5.1942,
      "mean_token_accuracy": 0.18311539739370347,
      "num_tokens": 25875661.0,
      "step": 14915
    },
    {
      "entropy": 5.524275255203247,
      "epoch": 1.1608247422680413,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048690058391456684,
      "loss": 5.0049,
      "mean_token_accuracy": 0.20060795843601226,
      "num_tokens": 25884951.0,
      "step": 14920
    },
    {
      "entropy": 5.635675573348999,
      "epoch": 1.1612137716397588,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004868912635313579,
      "loss": 5.2372,
      "mean_token_accuracy": 0.18609603941440583,
      "num_tokens": 25893652.0,
      "step": 14925
    },
    {
      "entropy": 5.639912080764771,
      "epoch": 1.1616028010114763,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004868819399330364,
      "loss": 5.0547,
      "mean_token_accuracy": 0.19008962363004683,
      "num_tokens": 25902375.0,
      "step": 14930
    },
    {
      "entropy": 5.61219596862793,
      "epoch": 1.161991830383194,
      "grad_norm": 1.0,
      "learning_rate": 0.0004868726131197436,
      "loss": 5.0558,
      "mean_token_accuracy": 0.1957837834954262,
      "num_tokens": 25911322.0,
      "step": 14935
    },
    {
      "entropy": 5.620619106292724,
      "epoch": 1.1623808597549115,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004868632830916212,
      "loss": 5.1609,
      "mean_token_accuracy": 0.19244829267263414,
      "num_tokens": 25920444.0,
      "step": 14940
    },
    {
      "entropy": 5.567115926742554,
      "epoch": 1.1627698891266292,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004868539498488106,
      "loss": 5.1674,
      "mean_token_accuracy": 0.18978559076786042,
      "num_tokens": 25928986.0,
      "step": 14945
    },
    {
      "entropy": 5.591844654083252,
      "epoch": 1.1631589184983466,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004868446133914536,
      "loss": 5.1573,
      "mean_token_accuracy": 0.18468926697969437,
      "num_tokens": 25938209.0,
      "step": 14950
    },
    {
      "entropy": 5.637552547454834,
      "epoch": 1.163547947870064,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048683527371969185,
      "loss": 5.1772,
      "mean_token_accuracy": 0.1779983326792717,
      "num_tokens": 25946909.0,
      "step": 14955
    },
    {
      "entropy": 5.591159296035767,
      "epoch": 1.1639369772417818,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048682593083366685,
      "loss": 5.2094,
      "mean_token_accuracy": 0.1895134285092354,
      "num_tokens": 25956154.0,
      "step": 14960
    },
    {
      "entropy": 5.592004442214966,
      "epoch": 1.1643260066134993,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048681658473352055,
      "loss": 5.1077,
      "mean_token_accuracy": 0.19457264244556427,
      "num_tokens": 25964478.0,
      "step": 14965
    },
    {
      "entropy": 5.553935861587524,
      "epoch": 1.164715035985217,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004868072354193947,
      "loss": 5.1047,
      "mean_token_accuracy": 0.18316972255706787,
      "num_tokens": 25973162.0,
      "step": 14970
    },
    {
      "entropy": 5.5709902286529545,
      "epoch": 1.1651040653569344,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004867978828914312,
      "loss": 5.0624,
      "mean_token_accuracy": 0.1913768991827965,
      "num_tokens": 25982790.0,
      "step": 14975
    },
    {
      "entropy": 5.5766767978668215,
      "epoch": 1.165493094728652,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004867885271497719,
      "loss": 5.0162,
      "mean_token_accuracy": 0.1961001604795456,
      "num_tokens": 25992111.0,
      "step": 14980
    },
    {
      "entropy": 5.630577659606933,
      "epoch": 1.1658821241003696,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004867791681945588,
      "loss": 5.1463,
      "mean_token_accuracy": 0.18628059178590775,
      "num_tokens": 26000803.0,
      "step": 14985
    },
    {
      "entropy": 5.654449033737182,
      "epoch": 1.166271153472087,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048676980602593395,
      "loss": 5.2573,
      "mean_token_accuracy": 0.18055839836597443,
      "num_tokens": 26010152.0,
      "step": 14990
    },
    {
      "entropy": 5.589032411575317,
      "epoch": 1.1666601828438048,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048676044064403936,
      "loss": 5.1552,
      "mean_token_accuracy": 0.18115283101797103,
      "num_tokens": 26019183.0,
      "step": 14995
    },
    {
      "entropy": 5.618133401870727,
      "epoch": 1.1670492122155223,
      "grad_norm": 1.125,
      "learning_rate": 0.0004867510720490171,
      "loss": 5.084,
      "mean_token_accuracy": 0.19277227520942689,
      "num_tokens": 26027459.0,
      "step": 15000
    },
    {
      "epoch": 1.1670492122155223,
      "eval_entropy": 5.38063917646509,
      "eval_loss": 5.199825763702393,
      "eval_mean_token_accuracy": 0.19610792320146367,
      "eval_num_tokens": 26027459.0,
      "eval_runtime": 21.6837,
      "eval_samples_per_second": 1916.558,
      "eval_steps_per_second": 239.581,
      "step": 15000
    },
    {
      "entropy": 5.6070170402526855,
      "epoch": 1.1674382415872397,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004867417002410094,
      "loss": 5.153,
      "mean_token_accuracy": 0.18308807909488678,
      "num_tokens": 26036421.0,
      "step": 15005
    },
    {
      "entropy": 5.571762180328369,
      "epoch": 1.1678272709589574,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048673232522015845,
      "loss": 5.1572,
      "mean_token_accuracy": 0.1901620551943779,
      "num_tokens": 26044971.0,
      "step": 15010
    },
    {
      "entropy": 5.558289480209351,
      "epoch": 1.168216300330675,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004867229469866064,
      "loss": 5.1052,
      "mean_token_accuracy": 0.18862562179565429,
      "num_tokens": 26053597.0,
      "step": 15015
    },
    {
      "entropy": 5.6364421367645265,
      "epoch": 1.1686053297023926,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004867135655404956,
      "loss": 5.1571,
      "mean_token_accuracy": 0.1851444274187088,
      "num_tokens": 26061928.0,
      "step": 15020
    },
    {
      "entropy": 5.576354598999023,
      "epoch": 1.16899435907411,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048670418088196844,
      "loss": 5.1127,
      "mean_token_accuracy": 0.19274311810731887,
      "num_tokens": 26070917.0,
      "step": 15025
    },
    {
      "entropy": 5.500616121292114,
      "epoch": 1.1693833884458276,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004866947930111674,
      "loss": 5.0449,
      "mean_token_accuracy": 0.19592256247997283,
      "num_tokens": 26079556.0,
      "step": 15030
    },
    {
      "entropy": 5.5612077713012695,
      "epoch": 1.1697724178175453,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048668540192823467,
      "loss": 5.1433,
      "mean_token_accuracy": 0.1903790056705475,
      "num_tokens": 26087963.0,
      "step": 15035
    },
    {
      "entropy": 5.6055991649627686,
      "epoch": 1.1701614471892627,
      "grad_norm": 1.0625,
      "learning_rate": 0.000486676007633313,
      "loss": 5.084,
      "mean_token_accuracy": 0.19391100704669953,
      "num_tokens": 26096208.0,
      "step": 15040
    },
    {
      "entropy": 5.6063727855682375,
      "epoch": 1.1705504765609804,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004866666101265448,
      "loss": 5.1093,
      "mean_token_accuracy": 0.18813618272542953,
      "num_tokens": 26104961.0,
      "step": 15045
    },
    {
      "entropy": 5.5672554016113285,
      "epoch": 1.170939505932698,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048665720940807274,
      "loss": 5.1264,
      "mean_token_accuracy": 0.18212450444698333,
      "num_tokens": 26113889.0,
      "step": 15050
    },
    {
      "entropy": 5.521347856521606,
      "epoch": 1.1713285353044154,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048664780547803935,
      "loss": 5.0879,
      "mean_token_accuracy": 0.19350726306438445,
      "num_tokens": 26122139.0,
      "step": 15055
    },
    {
      "entropy": 5.633164596557617,
      "epoch": 1.171717564676133,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004866383983365873,
      "loss": 5.1458,
      "mean_token_accuracy": 0.19945055693387986,
      "num_tokens": 26130817.0,
      "step": 15060
    },
    {
      "entropy": 5.600032997131348,
      "epoch": 1.1721065940478506,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048662898798385956,
      "loss": 5.1501,
      "mean_token_accuracy": 0.19003477841615676,
      "num_tokens": 26140035.0,
      "step": 15065
    },
    {
      "entropy": 5.669366121292114,
      "epoch": 1.1724956234195683,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048661957441999875,
      "loss": 5.2063,
      "mean_token_accuracy": 0.18841926008462906,
      "num_tokens": 26149575.0,
      "step": 15070
    },
    {
      "entropy": 5.637383031845093,
      "epoch": 1.1728846527912857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048661015764514765,
      "loss": 5.1419,
      "mean_token_accuracy": 0.19596780240535736,
      "num_tokens": 26158743.0,
      "step": 15075
    },
    {
      "entropy": 5.611569261550903,
      "epoch": 1.1732736821630034,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048660073765944927,
      "loss": 5.1557,
      "mean_token_accuracy": 0.18850169628858565,
      "num_tokens": 26167380.0,
      "step": 15080
    },
    {
      "entropy": 5.4767015933990475,
      "epoch": 1.173662711534721,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004865913144630465,
      "loss": 5.071,
      "mean_token_accuracy": 0.19493190497159957,
      "num_tokens": 26175613.0,
      "step": 15085
    },
    {
      "entropy": 5.618897724151611,
      "epoch": 1.1740517409064384,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048658188805608227,
      "loss": 5.1873,
      "mean_token_accuracy": 0.18750355243682862,
      "num_tokens": 26184347.0,
      "step": 15090
    },
    {
      "entropy": 5.597576522827149,
      "epoch": 1.174440770278156,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004865724584386998,
      "loss": 5.1111,
      "mean_token_accuracy": 0.19106998145580292,
      "num_tokens": 26193513.0,
      "step": 15095
    },
    {
      "entropy": 5.611871528625488,
      "epoch": 1.1748297996498736,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004865630256110418,
      "loss": 5.1127,
      "mean_token_accuracy": 0.18844806253910065,
      "num_tokens": 26202790.0,
      "step": 15100
    },
    {
      "entropy": 5.544135904312133,
      "epoch": 1.175218829021591,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048655358957325174,
      "loss": 5.0763,
      "mean_token_accuracy": 0.19279582798480988,
      "num_tokens": 26211032.0,
      "step": 15105
    },
    {
      "entropy": 5.50651330947876,
      "epoch": 1.1756078583933087,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048654415032547277,
      "loss": 5.0169,
      "mean_token_accuracy": 0.19430835843086242,
      "num_tokens": 26219963.0,
      "step": 15110
    },
    {
      "entropy": 5.617864322662354,
      "epoch": 1.1759968877650262,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048653470786784794,
      "loss": 5.0614,
      "mean_token_accuracy": 0.1897891029715538,
      "num_tokens": 26228164.0,
      "step": 15115
    },
    {
      "entropy": 5.569379043579102,
      "epoch": 1.176385917136744,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004865252622005206,
      "loss": 5.0896,
      "mean_token_accuracy": 0.19182399213314055,
      "num_tokens": 26236294.0,
      "step": 15120
    },
    {
      "entropy": 5.516499376296997,
      "epoch": 1.1767749465084614,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048651581332363405,
      "loss": 5.0441,
      "mean_token_accuracy": 0.19946387708187102,
      "num_tokens": 26244955.0,
      "step": 15125
    },
    {
      "entropy": 5.5880265712738035,
      "epoch": 1.177163975880179,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048650636123733176,
      "loss": 5.1306,
      "mean_token_accuracy": 0.18992896676063536,
      "num_tokens": 26253221.0,
      "step": 15130
    },
    {
      "entropy": 5.627907800674438,
      "epoch": 1.1775530052518965,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004864969059417571,
      "loss": 5.2193,
      "mean_token_accuracy": 0.17644216865301132,
      "num_tokens": 26262544.0,
      "step": 15135
    },
    {
      "entropy": 5.611646747589111,
      "epoch": 1.177942034623614,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004864874474370534,
      "loss": 5.1093,
      "mean_token_accuracy": 0.19003526717424393,
      "num_tokens": 26271864.0,
      "step": 15140
    },
    {
      "entropy": 5.655444192886352,
      "epoch": 1.1783310639953317,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048647798572336445,
      "loss": 5.1658,
      "mean_token_accuracy": 0.18868221789598466,
      "num_tokens": 26279713.0,
      "step": 15145
    },
    {
      "entropy": 5.545048809051513,
      "epoch": 1.1787200933670492,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048646852080083355,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20163817703723907,
      "num_tokens": 26288654.0,
      "step": 15150
    },
    {
      "entropy": 5.581171369552612,
      "epoch": 1.1791091227387667,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004864590526696045,
      "loss": 5.1769,
      "mean_token_accuracy": 0.18162493258714676,
      "num_tokens": 26297007.0,
      "step": 15155
    },
    {
      "entropy": 5.5863746166229244,
      "epoch": 1.1794981521104844,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048644958132982094,
      "loss": 5.1028,
      "mean_token_accuracy": 0.19216720908880233,
      "num_tokens": 26305601.0,
      "step": 15160
    },
    {
      "entropy": 5.618317174911499,
      "epoch": 1.1798871814822018,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048644010678162644,
      "loss": 5.1424,
      "mean_token_accuracy": 0.1963210791349411,
      "num_tokens": 26313455.0,
      "step": 15165
    },
    {
      "entropy": 5.6111854076385494,
      "epoch": 1.1802762108539195,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004864306290251649,
      "loss": 5.1522,
      "mean_token_accuracy": 0.188117116689682,
      "num_tokens": 26322450.0,
      "step": 15170
    },
    {
      "entropy": 5.63731575012207,
      "epoch": 1.180665240225637,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048642114806058014,
      "loss": 5.1653,
      "mean_token_accuracy": 0.1797630086541176,
      "num_tokens": 26331301.0,
      "step": 15175
    },
    {
      "entropy": 5.722377347946167,
      "epoch": 1.1810542695973547,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004864116638880159,
      "loss": 5.248,
      "mean_token_accuracy": 0.18856421262025833,
      "num_tokens": 26339937.0,
      "step": 15180
    },
    {
      "entropy": 5.626636075973511,
      "epoch": 1.1814432989690722,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004864021765076162,
      "loss": 5.1145,
      "mean_token_accuracy": 0.19093227237462998,
      "num_tokens": 26348541.0,
      "step": 15185
    },
    {
      "entropy": 5.558046579360962,
      "epoch": 1.1818323283407897,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048639268591952494,
      "loss": 5.0793,
      "mean_token_accuracy": 0.19542826861143112,
      "num_tokens": 26356621.0,
      "step": 15190
    },
    {
      "entropy": 5.613065147399903,
      "epoch": 1.1822213577125074,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048638319212388614,
      "loss": 5.1801,
      "mean_token_accuracy": 0.18269234746694565,
      "num_tokens": 26366112.0,
      "step": 15195
    },
    {
      "entropy": 5.665078783035279,
      "epoch": 1.1826103870842248,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004863736951208438,
      "loss": 5.1534,
      "mean_token_accuracy": 0.18836449235677719,
      "num_tokens": 26374768.0,
      "step": 15200
    },
    {
      "entropy": 5.597625112533569,
      "epoch": 1.1829994164559423,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048636419491054217,
      "loss": 5.1359,
      "mean_token_accuracy": 0.1941370978951454,
      "num_tokens": 26383097.0,
      "step": 15205
    },
    {
      "entropy": 5.6053431034088135,
      "epoch": 1.18338844582766,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004863546914931252,
      "loss": 5.1141,
      "mean_token_accuracy": 0.1876509353518486,
      "num_tokens": 26391361.0,
      "step": 15210
    },
    {
      "entropy": 5.5784852504730225,
      "epoch": 1.1837774751993775,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004863451848687373,
      "loss": 5.0642,
      "mean_token_accuracy": 0.1935330033302307,
      "num_tokens": 26400818.0,
      "step": 15215
    },
    {
      "entropy": 5.60666708946228,
      "epoch": 1.1841665045710952,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048633567503752253,
      "loss": 5.1392,
      "mean_token_accuracy": 0.1896408572793007,
      "num_tokens": 26409650.0,
      "step": 15220
    },
    {
      "entropy": 5.5267926216125485,
      "epoch": 1.1845555339428127,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004863261619996253,
      "loss": 5.0136,
      "mean_token_accuracy": 0.19559521079063416,
      "num_tokens": 26418237.0,
      "step": 15225
    },
    {
      "entropy": 5.470397090911865,
      "epoch": 1.1849445633145304,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004863166457551899,
      "loss": 4.9812,
      "mean_token_accuracy": 0.20518125891685485,
      "num_tokens": 26426366.0,
      "step": 15230
    },
    {
      "entropy": 5.4904931545257565,
      "epoch": 1.1853335926862478,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004863071263043608,
      "loss": 5.0764,
      "mean_token_accuracy": 0.18565648496150972,
      "num_tokens": 26435603.0,
      "step": 15235
    },
    {
      "entropy": 5.609816455841065,
      "epoch": 1.1857226220579653,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004862976036472823,
      "loss": 5.1902,
      "mean_token_accuracy": 0.18635938614606856,
      "num_tokens": 26444230.0,
      "step": 15240
    },
    {
      "entropy": 5.592506742477417,
      "epoch": 1.186111651429683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048628807778409907,
      "loss": 5.0532,
      "mean_token_accuracy": 0.18922004848718643,
      "num_tokens": 26452779.0,
      "step": 15245
    },
    {
      "entropy": 5.605657768249512,
      "epoch": 1.1865006808014005,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004862785487149555,
      "loss": 5.2057,
      "mean_token_accuracy": 0.18610675036907195,
      "num_tokens": 26461977.0,
      "step": 15250
    },
    {
      "entropy": 5.576980113983154,
      "epoch": 1.186889710173118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004862690164399963,
      "loss": 5.0114,
      "mean_token_accuracy": 0.19747587144374848,
      "num_tokens": 26470182.0,
      "step": 15255
    },
    {
      "entropy": 5.629970455169678,
      "epoch": 1.1872787395448356,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048625948095936593,
      "loss": 5.2197,
      "mean_token_accuracy": 0.18533240109682084,
      "num_tokens": 26479445.0,
      "step": 15260
    },
    {
      "entropy": 5.594922065734863,
      "epoch": 1.1876677689165531,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048624994227320923,
      "loss": 5.0857,
      "mean_token_accuracy": 0.19384056627750396,
      "num_tokens": 26488085.0,
      "step": 15265
    },
    {
      "entropy": 5.6941886901855465,
      "epoch": 1.1880567982882708,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048624040038167094,
      "loss": 5.2301,
      "mean_token_accuracy": 0.18443275094032288,
      "num_tokens": 26496849.0,
      "step": 15270
    },
    {
      "entropy": 5.56822657585144,
      "epoch": 1.1884458276599883,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048623085528489584,
      "loss": 5.0957,
      "mean_token_accuracy": 0.19489249885082244,
      "num_tokens": 26505412.0,
      "step": 15275
    },
    {
      "entropy": 5.5129741668701175,
      "epoch": 1.188834857031706,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048622130698302863,
      "loss": 5.0667,
      "mean_token_accuracy": 0.1912377417087555,
      "num_tokens": 26513564.0,
      "step": 15280
    },
    {
      "entropy": 5.533912181854248,
      "epoch": 1.1892238864034235,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048621175547621426,
      "loss": 5.0909,
      "mean_token_accuracy": 0.18857007175683976,
      "num_tokens": 26522230.0,
      "step": 15285
    },
    {
      "entropy": 5.56089072227478,
      "epoch": 1.189612915775141,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004862022007645978,
      "loss": 5.0803,
      "mean_token_accuracy": 0.19006049931049346,
      "num_tokens": 26531719.0,
      "step": 15290
    },
    {
      "entropy": 5.511069583892822,
      "epoch": 1.1900019451468586,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048619264284832403,
      "loss": 5.0378,
      "mean_token_accuracy": 0.20560531914234162,
      "num_tokens": 26540352.0,
      "step": 15295
    },
    {
      "entropy": 5.583624649047851,
      "epoch": 1.1903909745185761,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048618308172753804,
      "loss": 5.0254,
      "mean_token_accuracy": 0.1968090355396271,
      "num_tokens": 26548252.0,
      "step": 15300
    },
    {
      "entropy": 5.596045255661011,
      "epoch": 1.1907800038902936,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004861735174023849,
      "loss": 5.0925,
      "mean_token_accuracy": 0.1928489938378334,
      "num_tokens": 26557267.0,
      "step": 15305
    },
    {
      "entropy": 5.607676887512207,
      "epoch": 1.1911690332620113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048616394987300985,
      "loss": 5.2073,
      "mean_token_accuracy": 0.18672688454389572,
      "num_tokens": 26566221.0,
      "step": 15310
    },
    {
      "entropy": 5.597143888473511,
      "epoch": 1.1915580626337288,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004861543791395579,
      "loss": 5.2152,
      "mean_token_accuracy": 0.18726943284273148,
      "num_tokens": 26574826.0,
      "step": 15315
    },
    {
      "entropy": 5.59898796081543,
      "epoch": 1.1919470920054465,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004861448052021743,
      "loss": 5.1869,
      "mean_token_accuracy": 0.179260778427124,
      "num_tokens": 26583442.0,
      "step": 15320
    },
    {
      "entropy": 5.639116430282593,
      "epoch": 1.192336121377164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004861352280610044,
      "loss": 5.127,
      "mean_token_accuracy": 0.19415039718151092,
      "num_tokens": 26592106.0,
      "step": 15325
    },
    {
      "entropy": 5.587483024597168,
      "epoch": 1.1927251507488816,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048612564771619346,
      "loss": 5.0137,
      "mean_token_accuracy": 0.19882869422435762,
      "num_tokens": 26601275.0,
      "step": 15330
    },
    {
      "entropy": 5.475168609619141,
      "epoch": 1.193114180120599,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048611606416788686,
      "loss": 4.9694,
      "mean_token_accuracy": 0.19517303258180618,
      "num_tokens": 26610179.0,
      "step": 15335
    },
    {
      "entropy": 5.545497083663941,
      "epoch": 1.1935032094923166,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048610647741622996,
      "loss": 5.1411,
      "mean_token_accuracy": 0.18698107302188874,
      "num_tokens": 26619463.0,
      "step": 15340
    },
    {
      "entropy": 5.551906251907349,
      "epoch": 1.1938922388640343,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004860968874613683,
      "loss": 4.985,
      "mean_token_accuracy": 0.20008267164230348,
      "num_tokens": 26627597.0,
      "step": 15345
    },
    {
      "entropy": 5.566534519195557,
      "epoch": 1.1942812682357518,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004860872943034474,
      "loss": 5.0884,
      "mean_token_accuracy": 0.19655673056840897,
      "num_tokens": 26636180.0,
      "step": 15350
    },
    {
      "entropy": 5.571224308013916,
      "epoch": 1.1946702976074695,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004860776979426128,
      "loss": 5.0286,
      "mean_token_accuracy": 0.19708898961544036,
      "num_tokens": 26644803.0,
      "step": 15355
    },
    {
      "entropy": 5.5151488304138185,
      "epoch": 1.195059326979187,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004860680983790101,
      "loss": 5.0747,
      "mean_token_accuracy": 0.18061135560274125,
      "num_tokens": 26653619.0,
      "step": 15360
    },
    {
      "entropy": 5.598951387405395,
      "epoch": 1.1954483563509044,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004860584956127849,
      "loss": 5.0944,
      "mean_token_accuracy": 0.1840843752026558,
      "num_tokens": 26662291.0,
      "step": 15365
    },
    {
      "entropy": 5.610071134567261,
      "epoch": 1.195837385722622,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048604888964408306,
      "loss": 5.0501,
      "mean_token_accuracy": 0.20193059146404266,
      "num_tokens": 26670592.0,
      "step": 15370
    },
    {
      "entropy": 5.564420509338379,
      "epoch": 1.1962264150943396,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004860392804730502,
      "loss": 5.1194,
      "mean_token_accuracy": 0.19373519122600555,
      "num_tokens": 26679816.0,
      "step": 15375
    },
    {
      "entropy": 5.608189535140991,
      "epoch": 1.1966154444660573,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048602966809983204,
      "loss": 5.0841,
      "mean_token_accuracy": 0.1990629717707634,
      "num_tokens": 26688185.0,
      "step": 15380
    },
    {
      "entropy": 5.628693962097168,
      "epoch": 1.1970044738377748,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004860200525245746,
      "loss": 5.1441,
      "mean_token_accuracy": 0.19673731327056884,
      "num_tokens": 26697157.0,
      "step": 15385
    },
    {
      "entropy": 5.575571823120117,
      "epoch": 1.1973935032094922,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004860104337474239,
      "loss": 5.09,
      "mean_token_accuracy": 0.18258867263793946,
      "num_tokens": 26705868.0,
      "step": 15390
    },
    {
      "entropy": 5.5600487232208256,
      "epoch": 1.19778253258121,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048600081176852555,
      "loss": 5.1417,
      "mean_token_accuracy": 0.18435254096984863,
      "num_tokens": 26714364.0,
      "step": 15395
    },
    {
      "entropy": 5.573964262008667,
      "epoch": 1.1981715619529274,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048599118658802575,
      "loss": 5.1365,
      "mean_token_accuracy": 0.18718570321798325,
      "num_tokens": 26722866.0,
      "step": 15400
    },
    {
      "entropy": 5.631946086883545,
      "epoch": 1.198560591324645,
      "grad_norm": 1.125,
      "learning_rate": 0.0004859815582060706,
      "loss": 5.0898,
      "mean_token_accuracy": 0.1913563922047615,
      "num_tokens": 26731305.0,
      "step": 15405
    },
    {
      "entropy": 5.62173490524292,
      "epoch": 1.1989496206963626,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004859719266228061,
      "loss": 5.0882,
      "mean_token_accuracy": 0.19028402119874954,
      "num_tokens": 26740807.0,
      "step": 15410
    },
    {
      "entropy": 5.558030414581299,
      "epoch": 1.19933865006808,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004859622918383784,
      "loss": 5.0828,
      "mean_token_accuracy": 0.19723034054040908,
      "num_tokens": 26749748.0,
      "step": 15415
    },
    {
      "entropy": 5.599880599975586,
      "epoch": 1.1997276794397977,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004859526538529337,
      "loss": 5.1279,
      "mean_token_accuracy": 0.19151920974254608,
      "num_tokens": 26758597.0,
      "step": 15420
    },
    {
      "entropy": 5.590807342529297,
      "epoch": 1.2001167088115152,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004859430126666182,
      "loss": 5.1801,
      "mean_token_accuracy": 0.1843524843454361,
      "num_tokens": 26766772.0,
      "step": 15425
    },
    {
      "entropy": 5.476952457427979,
      "epoch": 1.200505738183233,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048593336827957824,
      "loss": 4.9198,
      "mean_token_accuracy": 0.2183031901717186,
      "num_tokens": 26775229.0,
      "step": 15430
    },
    {
      "entropy": 5.520895195007324,
      "epoch": 1.2008947675549504,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048592372069196024,
      "loss": 5.0846,
      "mean_token_accuracy": 0.18325729370117189,
      "num_tokens": 26784188.0,
      "step": 15435
    },
    {
      "entropy": 5.460034704208374,
      "epoch": 1.2012837969266679,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004859140699039104,
      "loss": 5.0273,
      "mean_token_accuracy": 0.19336565285921098,
      "num_tokens": 26792687.0,
      "step": 15440
    },
    {
      "entropy": 5.5107306957244875,
      "epoch": 1.2016728262983856,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048590441591557526,
      "loss": 5.0786,
      "mean_token_accuracy": 0.19534944593906403,
      "num_tokens": 26801931.0,
      "step": 15445
    },
    {
      "entropy": 5.600342416763306,
      "epoch": 1.202061855670103,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048589475872710134,
      "loss": 5.2971,
      "mean_token_accuracy": 0.182644072920084,
      "num_tokens": 26810689.0,
      "step": 15450
    },
    {
      "entropy": 5.60272331237793,
      "epoch": 1.2024508850418207,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004858850983386351,
      "loss": 5.111,
      "mean_token_accuracy": 0.18917934447526932,
      "num_tokens": 26819038.0,
      "step": 15455
    },
    {
      "entropy": 5.579953956604004,
      "epoch": 1.2028399144135382,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004858754347503231,
      "loss": 5.0171,
      "mean_token_accuracy": 0.20105472803115845,
      "num_tokens": 26827443.0,
      "step": 15460
    },
    {
      "entropy": 5.565564441680908,
      "epoch": 1.203228943785256,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048586576796231216,
      "loss": 5.0161,
      "mean_token_accuracy": 0.19953968226909638,
      "num_tokens": 26836306.0,
      "step": 15465
    },
    {
      "entropy": 5.584036779403687,
      "epoch": 1.2036179731569734,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004858560979747487,
      "loss": 5.1187,
      "mean_token_accuracy": 0.19076377153396606,
      "num_tokens": 26845043.0,
      "step": 15470
    },
    {
      "entropy": 5.587303256988525,
      "epoch": 1.2040070025286909,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048584642478777967,
      "loss": 5.0504,
      "mean_token_accuracy": 0.19465643614530564,
      "num_tokens": 26853978.0,
      "step": 15475
    },
    {
      "entropy": 5.553425025939942,
      "epoch": 1.2043960319004086,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004858367484015517,
      "loss": 4.9753,
      "mean_token_accuracy": 0.1934100016951561,
      "num_tokens": 26862487.0,
      "step": 15480
    },
    {
      "entropy": 5.534739875793457,
      "epoch": 1.204785061272126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048582706881621166,
      "loss": 5.1476,
      "mean_token_accuracy": 0.190867018699646,
      "num_tokens": 26871137.0,
      "step": 15485
    },
    {
      "entropy": 5.539325904846192,
      "epoch": 1.2051740906438435,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004858173860319063,
      "loss": 5.0504,
      "mean_token_accuracy": 0.19186416566371917,
      "num_tokens": 26880091.0,
      "step": 15490
    },
    {
      "entropy": 5.566355228424072,
      "epoch": 1.2055631200155612,
      "grad_norm": 1.125,
      "learning_rate": 0.00048580770004878277,
      "loss": 5.0923,
      "mean_token_accuracy": 0.19103320837020873,
      "num_tokens": 26888777.0,
      "step": 15495
    },
    {
      "entropy": 5.552337503433227,
      "epoch": 1.2059521493872787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004857980108669879,
      "loss": 5.049,
      "mean_token_accuracy": 0.19066378921270372,
      "num_tokens": 26897689.0,
      "step": 15500
    },
    {
      "entropy": 5.637480401992798,
      "epoch": 1.2063411787589964,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048578831848666875,
      "loss": 5.0968,
      "mean_token_accuracy": 0.18980053514242173,
      "num_tokens": 26906209.0,
      "step": 15505
    },
    {
      "entropy": 5.627501392364502,
      "epoch": 1.2067302081307139,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048577862290797235,
      "loss": 5.242,
      "mean_token_accuracy": 0.18738613426685333,
      "num_tokens": 26916340.0,
      "step": 15510
    },
    {
      "entropy": 5.617512178421021,
      "epoch": 1.2071192375024316,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048576892413104586,
      "loss": 5.1177,
      "mean_token_accuracy": 0.19289989918470382,
      "num_tokens": 26924482.0,
      "step": 15515
    },
    {
      "entropy": 5.549185800552368,
      "epoch": 1.207508266874149,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004857592221560364,
      "loss": 5.0774,
      "mean_token_accuracy": 0.18911547362804412,
      "num_tokens": 26932844.0,
      "step": 15520
    },
    {
      "entropy": 5.451512193679809,
      "epoch": 1.2078972962458665,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004857495169830912,
      "loss": 4.9645,
      "mean_token_accuracy": 0.1984395354986191,
      "num_tokens": 26940916.0,
      "step": 15525
    },
    {
      "entropy": 5.590522146224975,
      "epoch": 1.2082863256175842,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004857398086123575,
      "loss": 5.1085,
      "mean_token_accuracy": 0.19080109745264054,
      "num_tokens": 26949182.0,
      "step": 15530
    },
    {
      "entropy": 5.591371726989746,
      "epoch": 1.2086753549893017,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004857300970439827,
      "loss": 5.0546,
      "mean_token_accuracy": 0.19957173317670823,
      "num_tokens": 26957531.0,
      "step": 15535
    },
    {
      "entropy": 5.589437866210938,
      "epoch": 1.2090643843610192,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048572038227811404,
      "loss": 5.1522,
      "mean_token_accuracy": 0.19046340137720108,
      "num_tokens": 26965430.0,
      "step": 15540
    },
    {
      "entropy": 5.615444135665894,
      "epoch": 1.2094534137327368,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000485710664314899,
      "loss": 5.1805,
      "mean_token_accuracy": 0.1888631597161293,
      "num_tokens": 26974132.0,
      "step": 15545
    },
    {
      "entropy": 5.57098274230957,
      "epoch": 1.2098424431044543,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000485700943154485,
      "loss": 5.0712,
      "mean_token_accuracy": 0.1968693792819977,
      "num_tokens": 26982277.0,
      "step": 15550
    },
    {
      "entropy": 5.4820122718811035,
      "epoch": 1.210231472476172,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004856912187970195,
      "loss": 5.0301,
      "mean_token_accuracy": 0.1910757064819336,
      "num_tokens": 26991827.0,
      "step": 15555
    },
    {
      "entropy": 5.625962591171264,
      "epoch": 1.2106205018478895,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004856814912426502,
      "loss": 5.1834,
      "mean_token_accuracy": 0.18780433088541032,
      "num_tokens": 27000697.0,
      "step": 15560
    },
    {
      "entropy": 5.576309251785278,
      "epoch": 1.2110095312196072,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004856717604915245,
      "loss": 5.1232,
      "mean_token_accuracy": 0.19405819326639176,
      "num_tokens": 27009461.0,
      "step": 15565
    },
    {
      "entropy": 5.516828155517578,
      "epoch": 1.2113985605913247,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004856620265437902,
      "loss": 5.0538,
      "mean_token_accuracy": 0.19772937297821044,
      "num_tokens": 27017970.0,
      "step": 15570
    },
    {
      "entropy": 5.588766288757324,
      "epoch": 1.2117875899630421,
      "grad_norm": 1.0625,
      "learning_rate": 0.000485652289399595,
      "loss": 5.1133,
      "mean_token_accuracy": 0.19586780816316604,
      "num_tokens": 27026439.0,
      "step": 15575
    },
    {
      "entropy": 5.580862903594971,
      "epoch": 1.2121766193347598,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048564254905908655,
      "loss": 5.0592,
      "mean_token_accuracy": 0.19007128924131395,
      "num_tokens": 27034985.0,
      "step": 15580
    },
    {
      "entropy": 5.638749790191651,
      "epoch": 1.2125656487064773,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048563280552241266,
      "loss": 5.2045,
      "mean_token_accuracy": 0.18281891345977783,
      "num_tokens": 27044171.0,
      "step": 15585
    },
    {
      "entropy": 5.635423803329468,
      "epoch": 1.2129546780781948,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004856230587897212,
      "loss": 5.1648,
      "mean_token_accuracy": 0.18812349885702134,
      "num_tokens": 27053219.0,
      "step": 15590
    },
    {
      "entropy": 5.518210744857788,
      "epoch": 1.2133437074499125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004856133088611602,
      "loss": 5.0267,
      "mean_token_accuracy": 0.20125340521335602,
      "num_tokens": 27062012.0,
      "step": 15595
    },
    {
      "entropy": 5.556315279006958,
      "epoch": 1.21373273682163,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004856035557368773,
      "loss": 5.1216,
      "mean_token_accuracy": 0.18618103116750717,
      "num_tokens": 27070854.0,
      "step": 15600
    },
    {
      "entropy": 5.640985441207886,
      "epoch": 1.2141217661933477,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048559379941702074,
      "loss": 5.1629,
      "mean_token_accuracy": 0.19010813981294633,
      "num_tokens": 27079295.0,
      "step": 15605
    },
    {
      "entropy": 5.572699546813965,
      "epoch": 1.2145107955650651,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048558403990173844,
      "loss": 5.0501,
      "mean_token_accuracy": 0.19180653244256973,
      "num_tokens": 27087909.0,
      "step": 15610
    },
    {
      "entropy": 5.512327384948731,
      "epoch": 1.2148998249367828,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048557427719117855,
      "loss": 5.0837,
      "mean_token_accuracy": 0.1947572946548462,
      "num_tokens": 27096461.0,
      "step": 15615
    },
    {
      "entropy": 5.608134174346924,
      "epoch": 1.2152888543085003,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004855645112854891,
      "loss": 5.214,
      "mean_token_accuracy": 0.18693210631608964,
      "num_tokens": 27104846.0,
      "step": 15620
    },
    {
      "entropy": 5.619349527359009,
      "epoch": 1.2156778836802178,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004855547421848184,
      "loss": 5.0074,
      "mean_token_accuracy": 0.20129285305738448,
      "num_tokens": 27112623.0,
      "step": 15625
    },
    {
      "entropy": 5.5897434711456295,
      "epoch": 1.2160669130519355,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048554496988931456,
      "loss": 5.0528,
      "mean_token_accuracy": 0.1938340410590172,
      "num_tokens": 27120752.0,
      "step": 15630
    },
    {
      "entropy": 5.640067768096924,
      "epoch": 1.216455942423653,
      "grad_norm": 1.125,
      "learning_rate": 0.00048553519439912597,
      "loss": 5.2194,
      "mean_token_accuracy": 0.1824770301580429,
      "num_tokens": 27129583.0,
      "step": 15635
    },
    {
      "entropy": 5.614911413192749,
      "epoch": 1.2168449717953704,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004855254157144009,
      "loss": 5.1265,
      "mean_token_accuracy": 0.19445567578077316,
      "num_tokens": 27137807.0,
      "step": 15640
    },
    {
      "entropy": 5.541811513900757,
      "epoch": 1.2172340011670881,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004855156338352877,
      "loss": 4.9878,
      "mean_token_accuracy": 0.19631571918725968,
      "num_tokens": 27146246.0,
      "step": 15645
    },
    {
      "entropy": 5.572632741928101,
      "epoch": 1.2176230305388056,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048550584876193493,
      "loss": 5.0661,
      "mean_token_accuracy": 0.19595494270324706,
      "num_tokens": 27154811.0,
      "step": 15650
    },
    {
      "entropy": 5.608317852020264,
      "epoch": 1.2180120599105233,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048549606049449085,
      "loss": 5.0746,
      "mean_token_accuracy": 0.20396005660295485,
      "num_tokens": 27162849.0,
      "step": 15655
    },
    {
      "entropy": 5.552761554718018,
      "epoch": 1.2184010892822408,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004854862690331041,
      "loss": 5.1461,
      "mean_token_accuracy": 0.18890870958566666,
      "num_tokens": 27171268.0,
      "step": 15660
    },
    {
      "entropy": 5.582083988189697,
      "epoch": 1.2187901186539585,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048547647437792327,
      "loss": 5.0932,
      "mean_token_accuracy": 0.19241475462913513,
      "num_tokens": 27179571.0,
      "step": 15665
    },
    {
      "entropy": 5.561008262634277,
      "epoch": 1.219179148025676,
      "grad_norm": 1.109375,
      "learning_rate": 0.000485466676529097,
      "loss": 5.0763,
      "mean_token_accuracy": 0.1964916095137596,
      "num_tokens": 27188441.0,
      "step": 15670
    },
    {
      "entropy": 5.629905891418457,
      "epoch": 1.2195681773973934,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048545687548677373,
      "loss": 5.0965,
      "mean_token_accuracy": 0.18833362311124802,
      "num_tokens": 27196960.0,
      "step": 15675
    },
    {
      "entropy": 5.569472551345825,
      "epoch": 1.2199572067691111,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004854470712511025,
      "loss": 5.0958,
      "mean_token_accuracy": 0.194248528778553,
      "num_tokens": 27205129.0,
      "step": 15680
    },
    {
      "entropy": 5.5625584602355955,
      "epoch": 1.2203462361408286,
      "grad_norm": 1.0,
      "learning_rate": 0.00048543726382223193,
      "loss": 5.0546,
      "mean_token_accuracy": 0.19435477405786514,
      "num_tokens": 27214213.0,
      "step": 15685
    },
    {
      "entropy": 5.556509065628052,
      "epoch": 1.220735265512546,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048542745320031075,
      "loss": 5.1004,
      "mean_token_accuracy": 0.18486985117197036,
      "num_tokens": 27222447.0,
      "step": 15690
    },
    {
      "entropy": 5.584448385238647,
      "epoch": 1.2211242948842638,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048541763938548795,
      "loss": 5.0802,
      "mean_token_accuracy": 0.19131069481372834,
      "num_tokens": 27230872.0,
      "step": 15695
    },
    {
      "entropy": 5.589754295349121,
      "epoch": 1.2215133242559812,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048540782237791244,
      "loss": 5.1528,
      "mean_token_accuracy": 0.18907450288534164,
      "num_tokens": 27239171.0,
      "step": 15700
    },
    {
      "entropy": 5.532102870941162,
      "epoch": 1.221902353627699,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048539800217773293,
      "loss": 5.0568,
      "mean_token_accuracy": 0.18893016129732132,
      "num_tokens": 27247530.0,
      "step": 15705
    },
    {
      "entropy": 5.595793914794922,
      "epoch": 1.2222913829994164,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004853881787850988,
      "loss": 5.1411,
      "mean_token_accuracy": 0.18888894766569136,
      "num_tokens": 27255579.0,
      "step": 15710
    },
    {
      "entropy": 5.534402656555176,
      "epoch": 1.2226804123711341,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048537835220015886,
      "loss": 5.0785,
      "mean_token_accuracy": 0.19326250106096268,
      "num_tokens": 27264209.0,
      "step": 15715
    },
    {
      "entropy": 5.625450134277344,
      "epoch": 1.2230694417428516,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004853685224230623,
      "loss": 5.1456,
      "mean_token_accuracy": 0.18266841471195222,
      "num_tokens": 27272424.0,
      "step": 15720
    },
    {
      "entropy": 5.530358362197876,
      "epoch": 1.223458471114569,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048535868945395825,
      "loss": 4.9674,
      "mean_token_accuracy": 0.201402847468853,
      "num_tokens": 27281265.0,
      "step": 15725
    },
    {
      "entropy": 5.625441598892212,
      "epoch": 1.2238475004862868,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048534885329299586,
      "loss": 5.2232,
      "mean_token_accuracy": 0.19027862846851348,
      "num_tokens": 27290026.0,
      "step": 15730
    },
    {
      "entropy": 5.608558654785156,
      "epoch": 1.2242365298580042,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004853390139403245,
      "loss": 5.1688,
      "mean_token_accuracy": 0.18795347958803177,
      "num_tokens": 27298597.0,
      "step": 15735
    },
    {
      "entropy": 5.633943796157837,
      "epoch": 1.224625559229722,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048532917139609334,
      "loss": 5.1335,
      "mean_token_accuracy": 0.18765166252851487,
      "num_tokens": 27306763.0,
      "step": 15740
    },
    {
      "entropy": 5.580550622940064,
      "epoch": 1.2250145886014394,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004853193256604518,
      "loss": 5.1385,
      "mean_token_accuracy": 0.18764326423406602,
      "num_tokens": 27315782.0,
      "step": 15745
    },
    {
      "entropy": 5.681466007232666,
      "epoch": 1.2254036179731569,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048530947673354924,
      "loss": 5.1861,
      "mean_token_accuracy": 0.18378168642520903,
      "num_tokens": 27324487.0,
      "step": 15750
    },
    {
      "entropy": 5.61579475402832,
      "epoch": 1.2257926473448746,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004852996246155351,
      "loss": 5.1106,
      "mean_token_accuracy": 0.18874807208776473,
      "num_tokens": 27333099.0,
      "step": 15755
    },
    {
      "entropy": 5.582251024246216,
      "epoch": 1.226181676716592,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048528976930655896,
      "loss": 5.1176,
      "mean_token_accuracy": 0.18901627659797668,
      "num_tokens": 27341697.0,
      "step": 15760
    },
    {
      "entropy": 5.618842029571534,
      "epoch": 1.2265707060883098,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048527991080677015,
      "loss": 5.0962,
      "mean_token_accuracy": 0.1931592270731926,
      "num_tokens": 27349811.0,
      "step": 15765
    },
    {
      "entropy": 5.562830066680908,
      "epoch": 1.2269597354600272,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048527004911631843,
      "loss": 5.1045,
      "mean_token_accuracy": 0.19176326990127562,
      "num_tokens": 27358257.0,
      "step": 15770
    },
    {
      "entropy": 5.616780614852905,
      "epoch": 1.2273487648317447,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004852601842353534,
      "loss": 5.07,
      "mean_token_accuracy": 0.19709960520267486,
      "num_tokens": 27366993.0,
      "step": 15775
    },
    {
      "entropy": 5.610050439834595,
      "epoch": 1.2277377942034624,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048525031616402476,
      "loss": 5.1003,
      "mean_token_accuracy": 0.1969912141561508,
      "num_tokens": 27375780.0,
      "step": 15780
    },
    {
      "entropy": 5.576451301574707,
      "epoch": 1.2281268235751799,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048524044490248227,
      "loss": 5.1585,
      "mean_token_accuracy": 0.18969890773296355,
      "num_tokens": 27384641.0,
      "step": 15785
    },
    {
      "entropy": 5.638945150375366,
      "epoch": 1.2285158529468976,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048523057045087557,
      "loss": 5.1405,
      "mean_token_accuracy": 0.19688313007354735,
      "num_tokens": 27392990.0,
      "step": 15790
    },
    {
      "entropy": 5.650594854354859,
      "epoch": 1.228904882318615,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048522069280935466,
      "loss": 5.285,
      "mean_token_accuracy": 0.18183117657899855,
      "num_tokens": 27401012.0,
      "step": 15795
    },
    {
      "entropy": 5.54598650932312,
      "epoch": 1.2292939116903325,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004852108119780693,
      "loss": 5.0812,
      "mean_token_accuracy": 0.19106172472238542,
      "num_tokens": 27409758.0,
      "step": 15800
    },
    {
      "entropy": 5.614936780929566,
      "epoch": 1.2296829410620502,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004852009279571694,
      "loss": 5.0934,
      "mean_token_accuracy": 0.19357936829328537,
      "num_tokens": 27417592.0,
      "step": 15805
    },
    {
      "entropy": 5.570390510559082,
      "epoch": 1.2300719704337677,
      "grad_norm": 1.0625,
      "learning_rate": 0.000485191040746805,
      "loss": 5.058,
      "mean_token_accuracy": 0.19234727174043656,
      "num_tokens": 27426822.0,
      "step": 15810
    },
    {
      "entropy": 5.536443281173706,
      "epoch": 1.2304609998054854,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004851811503471262,
      "loss": 5.0443,
      "mean_token_accuracy": 0.20273693054914474,
      "num_tokens": 27435446.0,
      "step": 15815
    },
    {
      "entropy": 5.585339832305908,
      "epoch": 1.2308500291772029,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048517125675828294,
      "loss": 5.0434,
      "mean_token_accuracy": 0.19286508113145828,
      "num_tokens": 27443156.0,
      "step": 15820
    },
    {
      "entropy": 5.659043884277343,
      "epoch": 1.2312390585489204,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048516135998042536,
      "loss": 5.1712,
      "mean_token_accuracy": 0.19028818011283874,
      "num_tokens": 27451500.0,
      "step": 15825
    },
    {
      "entropy": 5.6254126071929935,
      "epoch": 1.231628087920638,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004851514600137037,
      "loss": 5.1307,
      "mean_token_accuracy": 0.19801352322101592,
      "num_tokens": 27460237.0,
      "step": 15830
    },
    {
      "entropy": 5.588210487365723,
      "epoch": 1.2320171172923555,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048514155685826807,
      "loss": 5.0884,
      "mean_token_accuracy": 0.19894171953201295,
      "num_tokens": 27468434.0,
      "step": 15835
    },
    {
      "entropy": 5.6366798877716064,
      "epoch": 1.2324061466640732,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004851316505142688,
      "loss": 5.1157,
      "mean_token_accuracy": 0.19308706521987914,
      "num_tokens": 27477667.0,
      "step": 15840
    },
    {
      "entropy": 5.618930673599243,
      "epoch": 1.2327951760357907,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048512174098185615,
      "loss": 5.0811,
      "mean_token_accuracy": 0.18397969752550125,
      "num_tokens": 27486365.0,
      "step": 15845
    },
    {
      "entropy": 5.530434226989746,
      "epoch": 1.2331842054075082,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048511182826118055,
      "loss": 5.0673,
      "mean_token_accuracy": 0.19726494401693345,
      "num_tokens": 27495310.0,
      "step": 15850
    },
    {
      "entropy": 5.4918054103851315,
      "epoch": 1.2335732347792259,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048510191235239246,
      "loss": 5.0325,
      "mean_token_accuracy": 0.19891120940446855,
      "num_tokens": 27504175.0,
      "step": 15855
    },
    {
      "entropy": 5.657974195480347,
      "epoch": 1.2339622641509433,
      "grad_norm": 1.125,
      "learning_rate": 0.00048509199325564217,
      "loss": 5.0955,
      "mean_token_accuracy": 0.18868385255336761,
      "num_tokens": 27512042.0,
      "step": 15860
    },
    {
      "entropy": 5.62535662651062,
      "epoch": 1.234351293522661,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004850820709710803,
      "loss": 5.1832,
      "mean_token_accuracy": 0.19400513917207718,
      "num_tokens": 27520721.0,
      "step": 15865
    },
    {
      "entropy": 5.512140846252441,
      "epoch": 1.2347403228943785,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004850721454988574,
      "loss": 5.0726,
      "mean_token_accuracy": 0.18600696921348572,
      "num_tokens": 27529431.0,
      "step": 15870
    },
    {
      "entropy": 5.520215559005737,
      "epoch": 1.235129352266096,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048506221683912405,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20258515030145646,
      "num_tokens": 27537380.0,
      "step": 15875
    },
    {
      "entropy": 5.509256362915039,
      "epoch": 1.2355183816378137,
      "grad_norm": 0.875,
      "learning_rate": 0.0004850522849920309,
      "loss": 5.0145,
      "mean_token_accuracy": 0.1896832525730133,
      "num_tokens": 27547487.0,
      "step": 15880
    },
    {
      "entropy": 5.645970439910888,
      "epoch": 1.2359074110095312,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048504234995772863,
      "loss": 5.0863,
      "mean_token_accuracy": 0.19388291388750076,
      "num_tokens": 27556813.0,
      "step": 15885
    },
    {
      "entropy": 5.626342058181763,
      "epoch": 1.2362964403812489,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004850324117363681,
      "loss": 5.2079,
      "mean_token_accuracy": 0.18508358299732208,
      "num_tokens": 27566602.0,
      "step": 15890
    },
    {
      "entropy": 5.575374269485474,
      "epoch": 1.2366854697529663,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048502247032809997,
      "loss": 5.1123,
      "mean_token_accuracy": 0.19680604785680772,
      "num_tokens": 27574577.0,
      "step": 15895
    },
    {
      "entropy": 5.685038471221924,
      "epoch": 1.237074499124684,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004850125257330751,
      "loss": 5.2692,
      "mean_token_accuracy": 0.17783930599689485,
      "num_tokens": 27582702.0,
      "step": 15900
    },
    {
      "entropy": 5.592756843566894,
      "epoch": 1.2374635284964015,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048500257795144446,
      "loss": 5.1617,
      "mean_token_accuracy": 0.18681564629077912,
      "num_tokens": 27591457.0,
      "step": 15905
    },
    {
      "entropy": 5.553035497665405,
      "epoch": 1.237852557868119,
      "grad_norm": 1.03125,
      "learning_rate": 0.000484992626983359,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19970177859067917,
      "num_tokens": 27600643.0,
      "step": 15910
    },
    {
      "entropy": 5.587669134140015,
      "epoch": 1.2382415872398367,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004849826728289696,
      "loss": 5.0241,
      "mean_token_accuracy": 0.19343722462654114,
      "num_tokens": 27608438.0,
      "step": 15915
    },
    {
      "entropy": 5.614461994171142,
      "epoch": 1.2386306166115542,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048497271548842737,
      "loss": 5.1477,
      "mean_token_accuracy": 0.19096991866827012,
      "num_tokens": 27617149.0,
      "step": 15920
    },
    {
      "entropy": 5.523227977752685,
      "epoch": 1.2390196459832716,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004849627549618834,
      "loss": 5.0878,
      "mean_token_accuracy": 0.1934687003493309,
      "num_tokens": 27625188.0,
      "step": 15925
    },
    {
      "entropy": 5.523476934432983,
      "epoch": 1.2394086753549893,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048495279124948886,
      "loss": 5.1571,
      "mean_token_accuracy": 0.19009485989809036,
      "num_tokens": 27633865.0,
      "step": 15930
    },
    {
      "entropy": 5.57327356338501,
      "epoch": 1.2397977047267068,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004849428243513948,
      "loss": 5.1101,
      "mean_token_accuracy": 0.19219367057085038,
      "num_tokens": 27642528.0,
      "step": 15935
    },
    {
      "entropy": 5.608651161193848,
      "epoch": 1.2401867340984245,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048493285426775255,
      "loss": 5.1983,
      "mean_token_accuracy": 0.17797445505857468,
      "num_tokens": 27651291.0,
      "step": 15940
    },
    {
      "entropy": 5.641522073745728,
      "epoch": 1.240575763470142,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004849228809987135,
      "loss": 5.194,
      "mean_token_accuracy": 0.1832926467061043,
      "num_tokens": 27660725.0,
      "step": 15945
    },
    {
      "entropy": 5.6311310768127445,
      "epoch": 1.2409647928418597,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004849129045444288,
      "loss": 5.0392,
      "mean_token_accuracy": 0.19656172543764114,
      "num_tokens": 27669282.0,
      "step": 15950
    },
    {
      "entropy": 5.563568258285523,
      "epoch": 1.2413538222135772,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004849029249050498,
      "loss": 5.0433,
      "mean_token_accuracy": 0.2009149357676506,
      "num_tokens": 27677987.0,
      "step": 15955
    },
    {
      "entropy": 5.500935697555542,
      "epoch": 1.2417428515852946,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048489294208072805,
      "loss": 5.0502,
      "mean_token_accuracy": 0.19451956152915956,
      "num_tokens": 27686409.0,
      "step": 15960
    },
    {
      "entropy": 5.554556941986084,
      "epoch": 1.2421318809570123,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048488295607161495,
      "loss": 5.0513,
      "mean_token_accuracy": 0.19104095846414565,
      "num_tokens": 27695485.0,
      "step": 15965
    },
    {
      "entropy": 5.547342348098755,
      "epoch": 1.2425209103287298,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004848729668778621,
      "loss": 4.9517,
      "mean_token_accuracy": 0.19947921484708786,
      "num_tokens": 27703962.0,
      "step": 15970
    },
    {
      "entropy": 5.5795379161834715,
      "epoch": 1.2429099397004473,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004848629744996211,
      "loss": 5.0794,
      "mean_token_accuracy": 0.1934753343462944,
      "num_tokens": 27713632.0,
      "step": 15975
    },
    {
      "entropy": 5.557565307617187,
      "epoch": 1.243298969072165,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004848529789370434,
      "loss": 5.1575,
      "mean_token_accuracy": 0.19070208668708802,
      "num_tokens": 27722890.0,
      "step": 15980
    },
    {
      "entropy": 5.58310489654541,
      "epoch": 1.2436879984438824,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004848429801902808,
      "loss": 5.0829,
      "mean_token_accuracy": 0.19657804667949677,
      "num_tokens": 27731325.0,
      "step": 15985
    },
    {
      "entropy": 5.600124549865723,
      "epoch": 1.2440770278156001,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048483297825948485,
      "loss": 5.163,
      "mean_token_accuracy": 0.18975530713796615,
      "num_tokens": 27739600.0,
      "step": 15990
    },
    {
      "entropy": 5.583254814147949,
      "epoch": 1.2444660571873176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048482297314480756,
      "loss": 5.145,
      "mean_token_accuracy": 0.1875674992799759,
      "num_tokens": 27747860.0,
      "step": 15995
    },
    {
      "entropy": 5.552810192108154,
      "epoch": 1.2448550865590353,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004848129648464006,
      "loss": 5.0608,
      "mean_token_accuracy": 0.1939297452569008,
      "num_tokens": 27757141.0,
      "step": 16000
    },
    {
      "entropy": 5.619099903106689,
      "epoch": 1.2452441159307528,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004848029533644159,
      "loss": 5.2286,
      "mean_token_accuracy": 0.1821589231491089,
      "num_tokens": 27766832.0,
      "step": 16005
    },
    {
      "entropy": 5.600907993316651,
      "epoch": 1.2456331453024703,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004847929386990052,
      "loss": 5.0999,
      "mean_token_accuracy": 0.18735098242759704,
      "num_tokens": 27775487.0,
      "step": 16010
    },
    {
      "entropy": 5.578440952301025,
      "epoch": 1.246022174674188,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048478292085032065,
      "loss": 5.0388,
      "mean_token_accuracy": 0.19397705793380737,
      "num_tokens": 27783601.0,
      "step": 16015
    },
    {
      "entropy": 5.626968955993652,
      "epoch": 1.2464112040459054,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004847728998185142,
      "loss": 5.1514,
      "mean_token_accuracy": 0.18919097930192946,
      "num_tokens": 27792931.0,
      "step": 16020
    },
    {
      "entropy": 5.642937469482422,
      "epoch": 1.246800233417623,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048476287560373786,
      "loss": 5.1112,
      "mean_token_accuracy": 0.1907007873058319,
      "num_tokens": 27801388.0,
      "step": 16025
    },
    {
      "entropy": 5.5731464385986325,
      "epoch": 1.2471892627893406,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048475284820614375,
      "loss": 5.0878,
      "mean_token_accuracy": 0.1909249857068062,
      "num_tokens": 27810047.0,
      "step": 16030
    },
    {
      "entropy": 5.588643646240234,
      "epoch": 1.247578292161058,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048474281762588407,
      "loss": 5.0863,
      "mean_token_accuracy": 0.1944271668791771,
      "num_tokens": 27818800.0,
      "step": 16035
    },
    {
      "entropy": 5.53579363822937,
      "epoch": 1.2479673215327758,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004847327838631109,
      "loss": 5.0497,
      "mean_token_accuracy": 0.19037897288799285,
      "num_tokens": 27827623.0,
      "step": 16040
    },
    {
      "entropy": 5.5154825210571286,
      "epoch": 1.2483563509044933,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004847227469179766,
      "loss": 5.0194,
      "mean_token_accuracy": 0.20358068197965623,
      "num_tokens": 27836491.0,
      "step": 16045
    },
    {
      "entropy": 5.57544322013855,
      "epoch": 1.248745380276211,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004847127067906334,
      "loss": 5.0835,
      "mean_token_accuracy": 0.19623470604419707,
      "num_tokens": 27845563.0,
      "step": 16050
    },
    {
      "entropy": 5.570050191879273,
      "epoch": 1.2491344096479284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048470266348123365,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19259732961654663,
      "num_tokens": 27854263.0,
      "step": 16055
    },
    {
      "entropy": 5.595818948745728,
      "epoch": 1.249523439019646,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004846926169899298,
      "loss": 5.0712,
      "mean_token_accuracy": 0.19274725317955016,
      "num_tokens": 27862868.0,
      "step": 16060
    },
    {
      "entropy": 5.605850887298584,
      "epoch": 1.2499124683913636,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048468256731687426,
      "loss": 5.1862,
      "mean_token_accuracy": 0.18923528492450714,
      "num_tokens": 27872053.0,
      "step": 16065
    },
    {
      "entropy": 5.554771852493286,
      "epoch": 1.250301497763081,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004846725144622194,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20081694722175597,
      "num_tokens": 27880953.0,
      "step": 16070
    },
    {
      "entropy": 5.5996613025665285,
      "epoch": 1.2506905271347986,
      "grad_norm": 1.078125,
      "learning_rate": 0.000484662458426118,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19625876247882842,
      "num_tokens": 27890102.0,
      "step": 16075
    },
    {
      "entropy": 5.630896520614624,
      "epoch": 1.2510795565065163,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048465239920872247,
      "loss": 5.17,
      "mean_token_accuracy": 0.1895263284444809,
      "num_tokens": 27899305.0,
      "step": 16080
    },
    {
      "entropy": 5.643464756011963,
      "epoch": 1.2514685858782337,
      "grad_norm": 1.125,
      "learning_rate": 0.00048464233681018545,
      "loss": 5.1111,
      "mean_token_accuracy": 0.19226248264312745,
      "num_tokens": 27907476.0,
      "step": 16085
    },
    {
      "entropy": 5.620686769485474,
      "epoch": 1.2518576152499514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004846322712306596,
      "loss": 5.1852,
      "mean_token_accuracy": 0.1901336207985878,
      "num_tokens": 27915353.0,
      "step": 16090
    },
    {
      "entropy": 5.546068525314331,
      "epoch": 1.252246644621669,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048462220247029783,
      "loss": 5.1022,
      "mean_token_accuracy": 0.1912755399942398,
      "num_tokens": 27923980.0,
      "step": 16095
    },
    {
      "entropy": 5.583468008041382,
      "epoch": 1.2526356739933866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048461213052925265,
      "loss": 5.1839,
      "mean_token_accuracy": 0.18978094011545182,
      "num_tokens": 27932760.0,
      "step": 16100
    },
    {
      "entropy": 5.566206312179565,
      "epoch": 1.253024703365104,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004846020554076771,
      "loss": 5.0548,
      "mean_token_accuracy": 0.1946013256907463,
      "num_tokens": 27941328.0,
      "step": 16105
    },
    {
      "entropy": 5.618392658233643,
      "epoch": 1.2534137327368216,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004845919771057239,
      "loss": 5.2321,
      "mean_token_accuracy": 0.18288879096508026,
      "num_tokens": 27949597.0,
      "step": 16110
    },
    {
      "entropy": 5.626197052001953,
      "epoch": 1.2538027621085392,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004845818956235462,
      "loss": 5.1073,
      "mean_token_accuracy": 0.19146668463945388,
      "num_tokens": 27958798.0,
      "step": 16115
    },
    {
      "entropy": 5.615036678314209,
      "epoch": 1.2541917914802567,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048457181096129664,
      "loss": 5.0582,
      "mean_token_accuracy": 0.19657903015613556,
      "num_tokens": 27967582.0,
      "step": 16120
    },
    {
      "entropy": 5.565688276290894,
      "epoch": 1.2545808208519742,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004845617231191285,
      "loss": 5.0449,
      "mean_token_accuracy": 0.18856090158224106,
      "num_tokens": 27975959.0,
      "step": 16125
    },
    {
      "entropy": 5.560848379135132,
      "epoch": 1.254969850223692,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004845516320971948,
      "loss": 5.0679,
      "mean_token_accuracy": 0.1995279684662819,
      "num_tokens": 27983980.0,
      "step": 16130
    },
    {
      "entropy": 5.565843915939331,
      "epoch": 1.2553588795954094,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004845415378956486,
      "loss": 4.9683,
      "mean_token_accuracy": 0.1910046696662903,
      "num_tokens": 27992563.0,
      "step": 16135
    },
    {
      "entropy": 5.548493051528931,
      "epoch": 1.255747908967127,
      "grad_norm": 1.015625,
      "learning_rate": 0.000484531440514643,
      "loss": 4.97,
      "mean_token_accuracy": 0.20675093978643416,
      "num_tokens": 28001457.0,
      "step": 16140
    },
    {
      "entropy": 5.664248275756836,
      "epoch": 1.2561369383388445,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048452133995433136,
      "loss": 5.1392,
      "mean_token_accuracy": 0.18932852298021316,
      "num_tokens": 28009507.0,
      "step": 16145
    },
    {
      "entropy": 5.508557224273682,
      "epoch": 1.2565259677105622,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004845112362148668,
      "loss": 5.0821,
      "mean_token_accuracy": 0.18782061338424683,
      "num_tokens": 28017858.0,
      "step": 16150
    },
    {
      "entropy": 5.619452667236328,
      "epoch": 1.2569149970822797,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004845011292964028,
      "loss": 5.147,
      "mean_token_accuracy": 0.19194145649671554,
      "num_tokens": 28026832.0,
      "step": 16155
    },
    {
      "entropy": 5.63121919631958,
      "epoch": 1.2573040264539972,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048449101919909265,
      "loss": 5.0282,
      "mean_token_accuracy": 0.19895150065422057,
      "num_tokens": 28034546.0,
      "step": 16160
    },
    {
      "entropy": 5.547819375991821,
      "epoch": 1.257693055825715,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048448090592308955,
      "loss": 5.102,
      "mean_token_accuracy": 0.18822899013757705,
      "num_tokens": 28043400.0,
      "step": 16165
    },
    {
      "entropy": 5.516633176803589,
      "epoch": 1.2580820851974324,
      "grad_norm": 1.125,
      "learning_rate": 0.0004844707894685473,
      "loss": 5.0692,
      "mean_token_accuracy": 0.19954805225133895,
      "num_tokens": 28052764.0,
      "step": 16170
    },
    {
      "entropy": 5.6272422790527346,
      "epoch": 1.2584711145691498,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004844606698356192,
      "loss": 5.148,
      "mean_token_accuracy": 0.1852506160736084,
      "num_tokens": 28061859.0,
      "step": 16175
    },
    {
      "entropy": 5.596032905578613,
      "epoch": 1.2588601439408675,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004844505470244588,
      "loss": 5.0273,
      "mean_token_accuracy": 0.1958564728498459,
      "num_tokens": 28070935.0,
      "step": 16180
    },
    {
      "entropy": 5.6125835418701175,
      "epoch": 1.2592491733125852,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004844404210352197,
      "loss": 5.0962,
      "mean_token_accuracy": 0.18744837492704391,
      "num_tokens": 28079864.0,
      "step": 16185
    },
    {
      "entropy": 5.515648889541626,
      "epoch": 1.2596382026843027,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048443029186805554,
      "loss": 4.9706,
      "mean_token_accuracy": 0.1960033118724823,
      "num_tokens": 28088938.0,
      "step": 16190
    },
    {
      "entropy": 5.558675241470337,
      "epoch": 1.2600272320560202,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004844201595231201,
      "loss": 5.0844,
      "mean_token_accuracy": 0.19341787099838256,
      "num_tokens": 28097213.0,
      "step": 16195
    },
    {
      "entropy": 5.5612386703491214,
      "epoch": 1.2604162614277379,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048441002400056706,
      "loss": 5.1093,
      "mean_token_accuracy": 0.19028161615133285,
      "num_tokens": 28105187.0,
      "step": 16200
    },
    {
      "entropy": 5.584390592575073,
      "epoch": 1.2608052907994554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004843998853005502,
      "loss": 5.0666,
      "mean_token_accuracy": 0.19468899965286254,
      "num_tokens": 28113238.0,
      "step": 16205
    },
    {
      "entropy": 5.540847206115723,
      "epoch": 1.2611943201711728,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004843897434232233,
      "loss": 5.0032,
      "mean_token_accuracy": 0.19913533926010132,
      "num_tokens": 28121502.0,
      "step": 16210
    },
    {
      "entropy": 5.5665552616119385,
      "epoch": 1.2615833495428905,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048437959836874047,
      "loss": 5.088,
      "mean_token_accuracy": 0.19445707499980927,
      "num_tokens": 28130269.0,
      "step": 16215
    },
    {
      "entropy": 5.601376152038574,
      "epoch": 1.261972378914608,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048436945013725544,
      "loss": 5.2229,
      "mean_token_accuracy": 0.18307623118162156,
      "num_tokens": 28139482.0,
      "step": 16220
    },
    {
      "entropy": 5.602341842651367,
      "epoch": 1.2623614082863255,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048435929872892226,
      "loss": 5.175,
      "mean_token_accuracy": 0.19212170094251632,
      "num_tokens": 28148443.0,
      "step": 16225
    },
    {
      "entropy": 5.628897380828858,
      "epoch": 1.2627504376580432,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004843491441438949,
      "loss": 5.031,
      "mean_token_accuracy": 0.19913185238838196,
      "num_tokens": 28156807.0,
      "step": 16230
    },
    {
      "entropy": 5.567248439788818,
      "epoch": 1.2631394670297609,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048433898638232755,
      "loss": 5.0874,
      "mean_token_accuracy": 0.19673034697771072,
      "num_tokens": 28165019.0,
      "step": 16235
    },
    {
      "entropy": 5.561327791213989,
      "epoch": 1.2635284964014784,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004843288254443742,
      "loss": 5.1295,
      "mean_token_accuracy": 0.1899797186255455,
      "num_tokens": 28173505.0,
      "step": 16240
    },
    {
      "entropy": 5.6576591491699215,
      "epoch": 1.2639175257731958,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004843186613301892,
      "loss": 5.1607,
      "mean_token_accuracy": 0.18639736324548722,
      "num_tokens": 28182524.0,
      "step": 16245
    },
    {
      "entropy": 5.662319087982178,
      "epoch": 1.2643065551449135,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004843084940399266,
      "loss": 5.207,
      "mean_token_accuracy": 0.18318491876125337,
      "num_tokens": 28191093.0,
      "step": 16250
    },
    {
      "entropy": 5.591050243377685,
      "epoch": 1.264695584516631,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004842983235737408,
      "loss": 5.0964,
      "mean_token_accuracy": 0.18838854879140854,
      "num_tokens": 28199092.0,
      "step": 16255
    },
    {
      "entropy": 5.657285737991333,
      "epoch": 1.2650846138883485,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004842881499317861,
      "loss": 5.1784,
      "mean_token_accuracy": 0.1909079894423485,
      "num_tokens": 28207341.0,
      "step": 16260
    },
    {
      "entropy": 5.6056070804595945,
      "epoch": 1.2654736432600662,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004842779731142168,
      "loss": 5.0872,
      "mean_token_accuracy": 0.1917978510260582,
      "num_tokens": 28216370.0,
      "step": 16265
    },
    {
      "entropy": 5.605053567886353,
      "epoch": 1.2658626726317836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048426779312118735,
      "loss": 5.1234,
      "mean_token_accuracy": 0.19039810448884964,
      "num_tokens": 28224554.0,
      "step": 16270
    },
    {
      "entropy": 5.663516330718994,
      "epoch": 1.2662517020035013,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004842576099528522,
      "loss": 5.1525,
      "mean_token_accuracy": 0.18665037900209427,
      "num_tokens": 28232672.0,
      "step": 16275
    },
    {
      "entropy": 5.551630687713623,
      "epoch": 1.2666407313752188,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004842474236093659,
      "loss": 4.9758,
      "mean_token_accuracy": 0.19758483916521072,
      "num_tokens": 28240730.0,
      "step": 16280
    },
    {
      "entropy": 5.5807586193084715,
      "epoch": 1.2670297607469365,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048423723409088306,
      "loss": 5.0974,
      "mean_token_accuracy": 0.1906809017062187,
      "num_tokens": 28248989.0,
      "step": 16285
    },
    {
      "entropy": 5.6339654445648195,
      "epoch": 1.267418790118654,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004842270413975582,
      "loss": 5.177,
      "mean_token_accuracy": 0.1912175014615059,
      "num_tokens": 28256900.0,
      "step": 16290
    },
    {
      "entropy": 5.648556184768677,
      "epoch": 1.2678078194903715,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000484216845529546,
      "loss": 5.1246,
      "mean_token_accuracy": 0.18360855877399446,
      "num_tokens": 28265832.0,
      "step": 16295
    },
    {
      "entropy": 5.56776704788208,
      "epoch": 1.2681968488620892,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048420664648700113,
      "loss": 5.0134,
      "mean_token_accuracy": 0.19365309178829193,
      "num_tokens": 28274661.0,
      "step": 16300
    },
    {
      "entropy": 5.614473009109497,
      "epoch": 1.2685858782338066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004841964442700784,
      "loss": 5.0773,
      "mean_token_accuracy": 0.19184325635433197,
      "num_tokens": 28283303.0,
      "step": 16305
    },
    {
      "entropy": 5.57660059928894,
      "epoch": 1.2689749076055241,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048418623887893264,
      "loss": 4.9794,
      "mean_token_accuracy": 0.19630232751369475,
      "num_tokens": 28291660.0,
      "step": 16310
    },
    {
      "entropy": 5.5983765602111815,
      "epoch": 1.2693639369772418,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048417603031371867,
      "loss": 5.1414,
      "mean_token_accuracy": 0.18618385791778563,
      "num_tokens": 28301314.0,
      "step": 16315
    },
    {
      "entropy": 5.598386764526367,
      "epoch": 1.2697529663489593,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004841658185745913,
      "loss": 5.0897,
      "mean_token_accuracy": 0.19332052618265153,
      "num_tokens": 28310044.0,
      "step": 16320
    },
    {
      "entropy": 5.720161008834839,
      "epoch": 1.270141995720677,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048415560366170564,
      "loss": 5.1888,
      "mean_token_accuracy": 0.19506587237119674,
      "num_tokens": 28319773.0,
      "step": 16325
    },
    {
      "entropy": 5.586122369766235,
      "epoch": 1.2705310250923945,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004841453855752165,
      "loss": 4.9882,
      "mean_token_accuracy": 0.19530003964900972,
      "num_tokens": 28328070.0,
      "step": 16330
    },
    {
      "entropy": 5.587516784667969,
      "epoch": 1.2709200544641122,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004841351643152791,
      "loss": 5.0815,
      "mean_token_accuracy": 0.1963294342160225,
      "num_tokens": 28337636.0,
      "step": 16335
    },
    {
      "entropy": 5.4862799644470215,
      "epoch": 1.2713090838358296,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004841249398820485,
      "loss": 5.0875,
      "mean_token_accuracy": 0.19166263937950134,
      "num_tokens": 28346609.0,
      "step": 16340
    },
    {
      "entropy": 5.587266969680786,
      "epoch": 1.271698113207547,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004841147122756797,
      "loss": 5.1128,
      "mean_token_accuracy": 0.19189873486757278,
      "num_tokens": 28355395.0,
      "step": 16345
    },
    {
      "entropy": 5.569062089920044,
      "epoch": 1.2720871425792648,
      "grad_norm": 1.09375,
      "learning_rate": 0.000484104481496328,
      "loss": 5.1861,
      "mean_token_accuracy": 0.1872741088271141,
      "num_tokens": 28364447.0,
      "step": 16350
    },
    {
      "entropy": 5.62692928314209,
      "epoch": 1.2724761719509823,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004840942475441486,
      "loss": 5.1193,
      "mean_token_accuracy": 0.19126724004745482,
      "num_tokens": 28372872.0,
      "step": 16355
    },
    {
      "entropy": 5.608555316925049,
      "epoch": 1.2728652013226998,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004840840104192969,
      "loss": 4.9748,
      "mean_token_accuracy": 0.19754622876644135,
      "num_tokens": 28380840.0,
      "step": 16360
    },
    {
      "entropy": 5.5594652652740475,
      "epoch": 1.2732542306944175,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000484073770121928,
      "loss": 5.0421,
      "mean_token_accuracy": 0.19092322736978531,
      "num_tokens": 28390417.0,
      "step": 16365
    },
    {
      "entropy": 5.574447917938232,
      "epoch": 1.273643260066135,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004840635266521975,
      "loss": 5.0974,
      "mean_token_accuracy": 0.18769218474626542,
      "num_tokens": 28400101.0,
      "step": 16370
    },
    {
      "entropy": 5.525232982635498,
      "epoch": 1.2740322894378526,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004840532800102607,
      "loss": 5.0204,
      "mean_token_accuracy": 0.19470336884260178,
      "num_tokens": 28408336.0,
      "step": 16375
    },
    {
      "entropy": 5.551790523529053,
      "epoch": 1.27442131880957,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048404303019627314,
      "loss": 5.044,
      "mean_token_accuracy": 0.19560120701789857,
      "num_tokens": 28416985.0,
      "step": 16380
    },
    {
      "entropy": 5.593158912658692,
      "epoch": 1.2748103481812878,
      "grad_norm": 1.125,
      "learning_rate": 0.00048403277721039036,
      "loss": 5.2012,
      "mean_token_accuracy": 0.18230695575475692,
      "num_tokens": 28425543.0,
      "step": 16385
    },
    {
      "entropy": 5.609423542022705,
      "epoch": 1.2751993775530053,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004840225210527679,
      "loss": 5.135,
      "mean_token_accuracy": 0.187672358751297,
      "num_tokens": 28435151.0,
      "step": 16390
    },
    {
      "entropy": 5.618977832794189,
      "epoch": 1.2755884069247228,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004840122617235613,
      "loss": 5.1311,
      "mean_token_accuracy": 0.18690426051616668,
      "num_tokens": 28444535.0,
      "step": 16395
    },
    {
      "entropy": 5.647767162322998,
      "epoch": 1.2759774362964404,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004840019992229263,
      "loss": 5.0739,
      "mean_token_accuracy": 0.19250009953975677,
      "num_tokens": 28452852.0,
      "step": 16400
    },
    {
      "entropy": 5.61885347366333,
      "epoch": 1.276366465668158,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048399173355101867,
      "loss": 5.1149,
      "mean_token_accuracy": 0.18428053855895996,
      "num_tokens": 28462573.0,
      "step": 16405
    },
    {
      "entropy": 5.649799156188965,
      "epoch": 1.2767554950398754,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048398146470799417,
      "loss": 5.1763,
      "mean_token_accuracy": 0.18333609849214555,
      "num_tokens": 28471714.0,
      "step": 16410
    },
    {
      "entropy": 5.566230726242066,
      "epoch": 1.277144524411593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048397119269400846,
      "loss": 5.0476,
      "mean_token_accuracy": 0.20105839818716048,
      "num_tokens": 28479989.0,
      "step": 16415
    },
    {
      "entropy": 5.571966981887817,
      "epoch": 1.2775335537833106,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048396091750921766,
      "loss": 5.0956,
      "mean_token_accuracy": 0.1905428558588028,
      "num_tokens": 28488707.0,
      "step": 16420
    },
    {
      "entropy": 5.672205829620362,
      "epoch": 1.2779225831550283,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004839506391537774,
      "loss": 5.2359,
      "mean_token_accuracy": 0.18538480401039123,
      "num_tokens": 28497228.0,
      "step": 16425
    },
    {
      "entropy": 5.660464096069336,
      "epoch": 1.2783116125267457,
      "grad_norm": 1.125,
      "learning_rate": 0.0004839403576278438,
      "loss": 5.0881,
      "mean_token_accuracy": 0.1925577163696289,
      "num_tokens": 28505604.0,
      "step": 16430
    },
    {
      "entropy": 5.5633667469024655,
      "epoch": 1.2787006418984634,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004839300729315729,
      "loss": 5.1376,
      "mean_token_accuracy": 0.18924558013677598,
      "num_tokens": 28514798.0,
      "step": 16435
    },
    {
      "entropy": 5.6681187629699705,
      "epoch": 1.279089671270181,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004839197850651206,
      "loss": 5.1366,
      "mean_token_accuracy": 0.1956767573952675,
      "num_tokens": 28523883.0,
      "step": 16440
    },
    {
      "entropy": 5.626384449005127,
      "epoch": 1.2794787006418984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048390949402864317,
      "loss": 5.1232,
      "mean_token_accuracy": 0.18650213479995728,
      "num_tokens": 28532290.0,
      "step": 16445
    },
    {
      "entropy": 5.53694715499878,
      "epoch": 1.279867730013616,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048389919982229666,
      "loss": 5.0439,
      "mean_token_accuracy": 0.19717105180025102,
      "num_tokens": 28540775.0,
      "step": 16450
    },
    {
      "entropy": 5.570633745193481,
      "epoch": 1.2802567593853336,
      "grad_norm": 1.125,
      "learning_rate": 0.00048388890244623723,
      "loss": 5.1526,
      "mean_token_accuracy": 0.187069371342659,
      "num_tokens": 28549517.0,
      "step": 16455
    },
    {
      "entropy": 5.612607765197754,
      "epoch": 1.280645788757051,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004838786019006213,
      "loss": 5.1072,
      "mean_token_accuracy": 0.19305062144994736,
      "num_tokens": 28557708.0,
      "step": 16460
    },
    {
      "entropy": 5.638970041275025,
      "epoch": 1.2810348181287687,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048386829818560493,
      "loss": 5.118,
      "mean_token_accuracy": 0.18778360188007354,
      "num_tokens": 28566912.0,
      "step": 16465
    },
    {
      "entropy": 5.556902313232422,
      "epoch": 1.2814238475004862,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048385799130134464,
      "loss": 5.1253,
      "mean_token_accuracy": 0.18722086399793625,
      "num_tokens": 28577104.0,
      "step": 16470
    },
    {
      "entropy": 5.558637285232544,
      "epoch": 1.281812876872204,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004838476812479968,
      "loss": 5.1151,
      "mean_token_accuracy": 0.18765891641378402,
      "num_tokens": 28585984.0,
      "step": 16475
    },
    {
      "entropy": 5.636599922180176,
      "epoch": 1.2822019062439214,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048383736802571774,
      "loss": 5.1455,
      "mean_token_accuracy": 0.18856384754180908,
      "num_tokens": 28594945.0,
      "step": 16480
    },
    {
      "entropy": 5.701941633224488,
      "epoch": 1.282590935615639,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000483827051634664,
      "loss": 5.1687,
      "mean_token_accuracy": 0.186891108751297,
      "num_tokens": 28603546.0,
      "step": 16485
    },
    {
      "entropy": 5.6071686267852785,
      "epoch": 1.2829799649873566,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048381673207499224,
      "loss": 5.1071,
      "mean_token_accuracy": 0.19561906456947326,
      "num_tokens": 28612057.0,
      "step": 16490
    },
    {
      "entropy": 5.540022325515747,
      "epoch": 1.283368994359074,
      "grad_norm": 1.125,
      "learning_rate": 0.0004838064093468588,
      "loss": 5.063,
      "mean_token_accuracy": 0.18930589854717256,
      "num_tokens": 28620945.0,
      "step": 16495
    },
    {
      "entropy": 5.525605964660644,
      "epoch": 1.2837580237307917,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004837960834504206,
      "loss": 4.9635,
      "mean_token_accuracy": 0.2052335560321808,
      "num_tokens": 28629028.0,
      "step": 16500
    },
    {
      "entropy": 5.527144479751587,
      "epoch": 1.2841470531025092,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004837857543858341,
      "loss": 5.0221,
      "mean_token_accuracy": 0.19998676031827928,
      "num_tokens": 28637533.0,
      "step": 16505
    },
    {
      "entropy": 5.626755571365356,
      "epoch": 1.2845360824742267,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004837754221532561,
      "loss": 5.1699,
      "mean_token_accuracy": 0.18968948125839233,
      "num_tokens": 28646906.0,
      "step": 16510
    },
    {
      "entropy": 5.558996725082397,
      "epoch": 1.2849251118459444,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048376508675284334,
      "loss": 5.0473,
      "mean_token_accuracy": 0.199790096282959,
      "num_tokens": 28655597.0,
      "step": 16515
    },
    {
      "entropy": 5.644053602218628,
      "epoch": 1.2853141412176619,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048375474818475274,
      "loss": 5.1807,
      "mean_token_accuracy": 0.18855624198913573,
      "num_tokens": 28664806.0,
      "step": 16520
    },
    {
      "entropy": 5.5634369373321535,
      "epoch": 1.2857031705893796,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048374440644914104,
      "loss": 5.0241,
      "mean_token_accuracy": 0.20249858498573303,
      "num_tokens": 28673610.0,
      "step": 16525
    },
    {
      "entropy": 5.611365222930909,
      "epoch": 1.286092199961097,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048373406154616523,
      "loss": 5.0798,
      "mean_token_accuracy": 0.19193932712078093,
      "num_tokens": 28681726.0,
      "step": 16530
    },
    {
      "entropy": 5.653905534744263,
      "epoch": 1.2864812293328147,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048372371347598226,
      "loss": 5.2393,
      "mean_token_accuracy": 0.17713529914617537,
      "num_tokens": 28691014.0,
      "step": 16535
    },
    {
      "entropy": 5.632786226272583,
      "epoch": 1.2868702587045322,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048371336223874923,
      "loss": 5.1223,
      "mean_token_accuracy": 0.18437448889017105,
      "num_tokens": 28699304.0,
      "step": 16540
    },
    {
      "entropy": 5.633014917373657,
      "epoch": 1.2872592880762497,
      "grad_norm": 1.109375,
      "learning_rate": 0.000483703007834623,
      "loss": 5.0416,
      "mean_token_accuracy": 0.19493970274925232,
      "num_tokens": 28707950.0,
      "step": 16545
    },
    {
      "entropy": 5.540950584411621,
      "epoch": 1.2876483174479674,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004836926502637609,
      "loss": 5.0564,
      "mean_token_accuracy": 0.19928259402513504,
      "num_tokens": 28716678.0,
      "step": 16550
    },
    {
      "entropy": 5.5410106658935545,
      "epoch": 1.2880373468196848,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048368228952632005,
      "loss": 5.0783,
      "mean_token_accuracy": 0.20132799297571183,
      "num_tokens": 28725658.0,
      "step": 16555
    },
    {
      "entropy": 5.645573425292969,
      "epoch": 1.2884263761914023,
      "grad_norm": 1.125,
      "learning_rate": 0.00048367192562245756,
      "loss": 5.1198,
      "mean_token_accuracy": 0.19349722415208817,
      "num_tokens": 28733859.0,
      "step": 16560
    },
    {
      "entropy": 5.52505316734314,
      "epoch": 1.28881540556312,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048366155855233067,
      "loss": 5.0158,
      "mean_token_accuracy": 0.19758519530296326,
      "num_tokens": 28741941.0,
      "step": 16565
    },
    {
      "entropy": 5.5179039478302006,
      "epoch": 1.2892044349348377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004836511883160968,
      "loss": 4.9459,
      "mean_token_accuracy": 0.20259023755788802,
      "num_tokens": 28750296.0,
      "step": 16570
    },
    {
      "entropy": 5.4994813919067385,
      "epoch": 1.2895934643065552,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004836408149139133,
      "loss": 4.9406,
      "mean_token_accuracy": 0.2073293909430504,
      "num_tokens": 28758359.0,
      "step": 16575
    },
    {
      "entropy": 5.61463098526001,
      "epoch": 1.2899824936782727,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048363043834593743,
      "loss": 5.14,
      "mean_token_accuracy": 0.1906037822365761,
      "num_tokens": 28766944.0,
      "step": 16580
    },
    {
      "entropy": 5.593300628662109,
      "epoch": 1.2903715230499904,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004836200586123268,
      "loss": 5.1483,
      "mean_token_accuracy": 0.1891775071620941,
      "num_tokens": 28775897.0,
      "step": 16585
    },
    {
      "entropy": 5.550483894348145,
      "epoch": 1.2907605524217078,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048360967571323875,
      "loss": 5.0644,
      "mean_token_accuracy": 0.19618291705846785,
      "num_tokens": 28784509.0,
      "step": 16590
    },
    {
      "entropy": 5.515319347381592,
      "epoch": 1.2911495817934253,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048359928964883094,
      "loss": 4.9466,
      "mean_token_accuracy": 0.20244549214839935,
      "num_tokens": 28793347.0,
      "step": 16595
    },
    {
      "entropy": 5.5574531078338625,
      "epoch": 1.291538611165143,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000483588900419261,
      "loss": 5.0429,
      "mean_token_accuracy": 0.19651865363121032,
      "num_tokens": 28801100.0,
      "step": 16600
    },
    {
      "entropy": 5.504043054580689,
      "epoch": 1.2919276405368605,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004835785080246864,
      "loss": 5.0202,
      "mean_token_accuracy": 0.19907681792974471,
      "num_tokens": 28809508.0,
      "step": 16605
    },
    {
      "entropy": 5.556125164031982,
      "epoch": 1.292316669908578,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000483568112465265,
      "loss": 5.1015,
      "mean_token_accuracy": 0.1911289930343628,
      "num_tokens": 28819263.0,
      "step": 16610
    },
    {
      "entropy": 5.6806353569030765,
      "epoch": 1.2927056992802957,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048355771374115436,
      "loss": 5.188,
      "mean_token_accuracy": 0.1793963521718979,
      "num_tokens": 28827316.0,
      "step": 16615
    },
    {
      "entropy": 5.634837627410889,
      "epoch": 1.2930947286520134,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004835473118525125,
      "loss": 5.1937,
      "mean_token_accuracy": 0.18826982378959656,
      "num_tokens": 28836272.0,
      "step": 16620
    },
    {
      "entropy": 5.567659711837768,
      "epoch": 1.2934837580237308,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004835369067994971,
      "loss": 5.0773,
      "mean_token_accuracy": 0.19229266941547393,
      "num_tokens": 28845180.0,
      "step": 16625
    },
    {
      "entropy": 5.560893630981445,
      "epoch": 1.2938727873954483,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000483526498582266,
      "loss": 5.0406,
      "mean_token_accuracy": 0.1899822846055031,
      "num_tokens": 28853903.0,
      "step": 16630
    },
    {
      "entropy": 5.557149267196655,
      "epoch": 1.294261816767166,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048351608720097715,
      "loss": 5.0591,
      "mean_token_accuracy": 0.1918584644794464,
      "num_tokens": 28863071.0,
      "step": 16635
    },
    {
      "entropy": 5.6259753704071045,
      "epoch": 1.2946508461388835,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048350567265578867,
      "loss": 5.1436,
      "mean_token_accuracy": 0.18397480845451356,
      "num_tokens": 28871620.0,
      "step": 16640
    },
    {
      "entropy": 5.669278907775879,
      "epoch": 1.295039875510601,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004834952549468584,
      "loss": 5.1861,
      "mean_token_accuracy": 0.18329119980335234,
      "num_tokens": 28879940.0,
      "step": 16645
    },
    {
      "entropy": 5.663364696502685,
      "epoch": 1.2954289048823187,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004834848340743446,
      "loss": 5.188,
      "mean_token_accuracy": 0.18708661198616028,
      "num_tokens": 28889012.0,
      "step": 16650
    },
    {
      "entropy": 5.593817567825317,
      "epoch": 1.2958179342540361,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004834744100384052,
      "loss": 5.0649,
      "mean_token_accuracy": 0.19851250797510148,
      "num_tokens": 28897312.0,
      "step": 16655
    },
    {
      "entropy": 5.596896457672119,
      "epoch": 1.2962069636257538,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004834639828391984,
      "loss": 5.0228,
      "mean_token_accuracy": 0.20130644738674164,
      "num_tokens": 28905977.0,
      "step": 16660
    },
    {
      "entropy": 5.5150003910064695,
      "epoch": 1.2965959929974713,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048345355247688256,
      "loss": 5.0315,
      "mean_token_accuracy": 0.19771711081266402,
      "num_tokens": 28913979.0,
      "step": 16665
    },
    {
      "entropy": 5.565609741210937,
      "epoch": 1.296985022369189,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004834431189516158,
      "loss": 5.0589,
      "mean_token_accuracy": 0.19748270362615586,
      "num_tokens": 28922090.0,
      "step": 16670
    },
    {
      "entropy": 5.644658708572388,
      "epoch": 1.2973740517409065,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004834326822635565,
      "loss": 5.196,
      "mean_token_accuracy": 0.1837651640176773,
      "num_tokens": 28930816.0,
      "step": 16675
    },
    {
      "entropy": 5.634592819213867,
      "epoch": 1.297763081112624,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048342224241286296,
      "loss": 5.1681,
      "mean_token_accuracy": 0.18510053902864457,
      "num_tokens": 28939125.0,
      "step": 16680
    },
    {
      "entropy": 5.5525157928466795,
      "epoch": 1.2981521104843416,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004834117993996937,
      "loss": 5.0497,
      "mean_token_accuracy": 0.19919029921293258,
      "num_tokens": 28947783.0,
      "step": 16685
    },
    {
      "entropy": 5.677631807327271,
      "epoch": 1.2985411398560591,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004834013532242071,
      "loss": 5.1768,
      "mean_token_accuracy": 0.19040046632289886,
      "num_tokens": 28957305.0,
      "step": 16690
    },
    {
      "entropy": 5.6526202201843265,
      "epoch": 1.2989301692277766,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004833909038865616,
      "loss": 5.1309,
      "mean_token_accuracy": 0.18405349999666215,
      "num_tokens": 28965757.0,
      "step": 16695
    },
    {
      "entropy": 5.632620620727539,
      "epoch": 1.2993191985994943,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004833804513869159,
      "loss": 5.1242,
      "mean_token_accuracy": 0.19095647931098939,
      "num_tokens": 28974846.0,
      "step": 16700
    },
    {
      "entropy": 5.501182699203492,
      "epoch": 1.2997082279712118,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004833699957254284,
      "loss": 4.9758,
      "mean_token_accuracy": 0.19298425763845445,
      "num_tokens": 28984060.0,
      "step": 16705
    },
    {
      "entropy": 5.580200910568237,
      "epoch": 1.3000972573429295,
      "grad_norm": 1.0625,
      "learning_rate": 0.000483359536902258,
      "loss": 5.0239,
      "mean_token_accuracy": 0.19201190173625945,
      "num_tokens": 28992705.0,
      "step": 16710
    },
    {
      "entropy": 5.635155057907104,
      "epoch": 1.300486286714647,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004833490749175632,
      "loss": 5.1393,
      "mean_token_accuracy": 0.19078140705823898,
      "num_tokens": 29001306.0,
      "step": 16715
    },
    {
      "entropy": 5.588790416717529,
      "epoch": 1.3008753160863646,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048333860977150286,
      "loss": 5.143,
      "mean_token_accuracy": 0.19001102298498154,
      "num_tokens": 29010371.0,
      "step": 16720
    },
    {
      "entropy": 5.6419358253479,
      "epoch": 1.3012643454580821,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048332814146423566,
      "loss": 5.1004,
      "mean_token_accuracy": 0.19558511972427367,
      "num_tokens": 29019940.0,
      "step": 16725
    },
    {
      "entropy": 5.660441589355469,
      "epoch": 1.3016533748297996,
      "grad_norm": 1.125,
      "learning_rate": 0.0004833176699959206,
      "loss": 5.1284,
      "mean_token_accuracy": 0.19217051714658737,
      "num_tokens": 29027644.0,
      "step": 16730
    },
    {
      "entropy": 5.57693076133728,
      "epoch": 1.3020424042015173,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048330719536671633,
      "loss": 5.0715,
      "mean_token_accuracy": 0.19291132539510727,
      "num_tokens": 29035166.0,
      "step": 16735
    },
    {
      "entropy": 5.517680549621582,
      "epoch": 1.3024314335732348,
      "grad_norm": 1.0,
      "learning_rate": 0.000483296717576782,
      "loss": 5.0236,
      "mean_token_accuracy": 0.19569295942783355,
      "num_tokens": 29044539.0,
      "step": 16740
    },
    {
      "entropy": 5.588887453079224,
      "epoch": 1.3028204629449522,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004832862366262765,
      "loss": 5.0743,
      "mean_token_accuracy": 0.2067807838320732,
      "num_tokens": 29052849.0,
      "step": 16745
    },
    {
      "entropy": 5.552817726135254,
      "epoch": 1.30320949231667,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048327575251535886,
      "loss": 5.0801,
      "mean_token_accuracy": 0.18806463479995728,
      "num_tokens": 29061881.0,
      "step": 16750
    },
    {
      "entropy": 5.706825923919678,
      "epoch": 1.3035985216883874,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004832652652441883,
      "loss": 5.1506,
      "mean_token_accuracy": 0.1846309334039688,
      "num_tokens": 29070460.0,
      "step": 16755
    },
    {
      "entropy": 5.688552808761597,
      "epoch": 1.303987551060105,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048325477481292375,
      "loss": 5.1726,
      "mean_token_accuracy": 0.1844995841383934,
      "num_tokens": 29079061.0,
      "step": 16760
    },
    {
      "entropy": 5.5268206119537355,
      "epoch": 1.3043765804318226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004832442812217244,
      "loss": 5.0562,
      "mean_token_accuracy": 0.19239771664142608,
      "num_tokens": 29087525.0,
      "step": 16765
    },
    {
      "entropy": 5.56028995513916,
      "epoch": 1.3047656098035403,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004832337844707496,
      "loss": 5.0332,
      "mean_token_accuracy": 0.19624444991350173,
      "num_tokens": 29096206.0,
      "step": 16770
    },
    {
      "entropy": 5.653877019882202,
      "epoch": 1.3051546391752578,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048322328456015855,
      "loss": 5.2253,
      "mean_token_accuracy": 0.19010618031024934,
      "num_tokens": 29105025.0,
      "step": 16775
    },
    {
      "entropy": 5.6057463645935055,
      "epoch": 1.3055436685469752,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048321278149011053,
      "loss": 5.1835,
      "mean_token_accuracy": 0.18936053216457366,
      "num_tokens": 29113468.0,
      "step": 16780
    },
    {
      "entropy": 5.5992865562438965,
      "epoch": 1.305932697918693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048320227526076504,
      "loss": 5.0798,
      "mean_token_accuracy": 0.19523320645093917,
      "num_tokens": 29122904.0,
      "step": 16785
    },
    {
      "entropy": 5.607943916320801,
      "epoch": 1.3063217272904104,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004831917658722814,
      "loss": 5.0633,
      "mean_token_accuracy": 0.20157826095819473,
      "num_tokens": 29132408.0,
      "step": 16790
    },
    {
      "entropy": 5.648335695266724,
      "epoch": 1.3067107566621279,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048318125332481903,
      "loss": 5.2157,
      "mean_token_accuracy": 0.18695988208055497,
      "num_tokens": 29141840.0,
      "step": 16795
    },
    {
      "entropy": 5.718431091308593,
      "epoch": 1.3070997860338456,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048317073761853764,
      "loss": 5.1752,
      "mean_token_accuracy": 0.18881198167800903,
      "num_tokens": 29150759.0,
      "step": 16800
    },
    {
      "entropy": 5.544278478622436,
      "epoch": 1.307488815405563,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004831602187535965,
      "loss": 4.9853,
      "mean_token_accuracy": 0.2031836599111557,
      "num_tokens": 29159489.0,
      "step": 16805
    },
    {
      "entropy": 5.56814866065979,
      "epoch": 1.3078778447772808,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004831496967301554,
      "loss": 5.0213,
      "mean_token_accuracy": 0.1931745857000351,
      "num_tokens": 29168160.0,
      "step": 16810
    },
    {
      "entropy": 5.628138732910156,
      "epoch": 1.3082668741489982,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048313917154837386,
      "loss": 5.1058,
      "mean_token_accuracy": 0.19329580068588256,
      "num_tokens": 29176534.0,
      "step": 16815
    },
    {
      "entropy": 5.542742204666138,
      "epoch": 1.308655903520716,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004831286432084118,
      "loss": 5.0077,
      "mean_token_accuracy": 0.203941248357296,
      "num_tokens": 29185461.0,
      "step": 16820
    },
    {
      "entropy": 5.604961585998535,
      "epoch": 1.3090449328924334,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004831181117104289,
      "loss": 5.144,
      "mean_token_accuracy": 0.18596413284540175,
      "num_tokens": 29194218.0,
      "step": 16825
    },
    {
      "entropy": 5.577546691894531,
      "epoch": 1.3094339622641509,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048310757705458476,
      "loss": 5.1079,
      "mean_token_accuracy": 0.19188106060028076,
      "num_tokens": 29203720.0,
      "step": 16830
    },
    {
      "entropy": 5.613342523574829,
      "epoch": 1.3098229916358686,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048309703924103944,
      "loss": 5.1674,
      "mean_token_accuracy": 0.1892984539270401,
      "num_tokens": 29212588.0,
      "step": 16835
    },
    {
      "entropy": 5.6453112125396725,
      "epoch": 1.310212021007586,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048308649826995283,
      "loss": 5.169,
      "mean_token_accuracy": 0.1883926883339882,
      "num_tokens": 29220718.0,
      "step": 16840
    },
    {
      "entropy": 5.6234375,
      "epoch": 1.3106010503793035,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048307595414148475,
      "loss": 5.0402,
      "mean_token_accuracy": 0.19771301746368408,
      "num_tokens": 29229386.0,
      "step": 16845
    },
    {
      "entropy": 5.604703712463379,
      "epoch": 1.3109900797510212,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004830654068557952,
      "loss": 5.0924,
      "mean_token_accuracy": 0.19491372704505922,
      "num_tokens": 29238390.0,
      "step": 16850
    },
    {
      "entropy": 5.643201112747192,
      "epoch": 1.3113791091227387,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004830548564130444,
      "loss": 5.183,
      "mean_token_accuracy": 0.18927546441555024,
      "num_tokens": 29246502.0,
      "step": 16855
    },
    {
      "entropy": 5.582104253768921,
      "epoch": 1.3117681384944564,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048304430281339216,
      "loss": 5.1069,
      "mean_token_accuracy": 0.1855737254023552,
      "num_tokens": 29254954.0,
      "step": 16860
    },
    {
      "entropy": 5.606703567504883,
      "epoch": 1.3121571678661739,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004830337460569989,
      "loss": 5.0788,
      "mean_token_accuracy": 0.2005214884877205,
      "num_tokens": 29263294.0,
      "step": 16865
    },
    {
      "entropy": 5.584879732131958,
      "epoch": 1.3125461972378916,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004830231861440246,
      "loss": 5.0548,
      "mean_token_accuracy": 0.19693197309970856,
      "num_tokens": 29271737.0,
      "step": 16870
    },
    {
      "entropy": 5.69629430770874,
      "epoch": 1.312935226609609,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004830126230746295,
      "loss": 5.1933,
      "mean_token_accuracy": 0.18574800491333007,
      "num_tokens": 29279473.0,
      "step": 16875
    },
    {
      "entropy": 5.60352578163147,
      "epoch": 1.3133242559813265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048300205684897405,
      "loss": 5.1038,
      "mean_token_accuracy": 0.20338490307331086,
      "num_tokens": 29287650.0,
      "step": 16880
    },
    {
      "entropy": 5.600238370895386,
      "epoch": 1.3137132853530442,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004829914874672184,
      "loss": 5.1287,
      "mean_token_accuracy": 0.18724319487810134,
      "num_tokens": 29296914.0,
      "step": 16885
    },
    {
      "entropy": 5.666216325759888,
      "epoch": 1.3141023147247617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048298091492952297,
      "loss": 5.1136,
      "mean_token_accuracy": 0.18820077627897264,
      "num_tokens": 29305409.0,
      "step": 16890
    },
    {
      "entropy": 5.69935245513916,
      "epoch": 1.3144913440964792,
      "grad_norm": 1.0,
      "learning_rate": 0.0004829703392360482,
      "loss": 5.135,
      "mean_token_accuracy": 0.18850217014551163,
      "num_tokens": 29314333.0,
      "step": 16895
    },
    {
      "entropy": 5.510940837860107,
      "epoch": 1.3148803734681969,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048295976038695455,
      "loss": 4.9864,
      "mean_token_accuracy": 0.19643135219812394,
      "num_tokens": 29323764.0,
      "step": 16900
    },
    {
      "entropy": 5.571589994430542,
      "epoch": 1.3152694028399143,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004829491783824025,
      "loss": 5.0723,
      "mean_token_accuracy": 0.19625774621963502,
      "num_tokens": 29331734.0,
      "step": 16905
    },
    {
      "entropy": 5.523209047317505,
      "epoch": 1.315658432211632,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048293859322255276,
      "loss": 4.9797,
      "mean_token_accuracy": 0.20637786090373994,
      "num_tokens": 29339791.0,
      "step": 16910
    },
    {
      "entropy": 5.538656234741211,
      "epoch": 1.3160474615833495,
      "grad_norm": 1.0,
      "learning_rate": 0.0004829280049075657,
      "loss": 5.1159,
      "mean_token_accuracy": 0.1856507495045662,
      "num_tokens": 29348548.0,
      "step": 16915
    },
    {
      "entropy": 5.570025873184204,
      "epoch": 1.3164364909550672,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048291741343760216,
      "loss": 5.0421,
      "mean_token_accuracy": 0.19357602000236512,
      "num_tokens": 29357151.0,
      "step": 16920
    },
    {
      "entropy": 5.672178745269775,
      "epoch": 1.3168255203267847,
      "grad_norm": 1.125,
      "learning_rate": 0.0004829068188128229,
      "loss": 5.0379,
      "mean_token_accuracy": 0.20119286328554153,
      "num_tokens": 29365643.0,
      "step": 16925
    },
    {
      "entropy": 5.581224250793457,
      "epoch": 1.3172145496985022,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004828962210333885,
      "loss": 5.1326,
      "mean_token_accuracy": 0.18838583528995514,
      "num_tokens": 29374215.0,
      "step": 16930
    },
    {
      "entropy": 5.569288730621338,
      "epoch": 1.3176035790702199,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004828856200994598,
      "loss": 5.0447,
      "mean_token_accuracy": 0.19315528720617295,
      "num_tokens": 29382586.0,
      "step": 16935
    },
    {
      "entropy": 5.535900545120239,
      "epoch": 1.3179926084419373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004828750160111978,
      "loss": 5.1007,
      "mean_token_accuracy": 0.18802523463964463,
      "num_tokens": 29391922.0,
      "step": 16940
    },
    {
      "entropy": 5.609203863143921,
      "epoch": 1.3183816378136548,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048286440876876325,
      "loss": 5.0571,
      "mean_token_accuracy": 0.19533028900623323,
      "num_tokens": 29400408.0,
      "step": 16945
    },
    {
      "entropy": 5.56855616569519,
      "epoch": 1.3187706671853725,
      "grad_norm": 1.125,
      "learning_rate": 0.00048285379837231714,
      "loss": 4.9909,
      "mean_token_accuracy": 0.20062803030014037,
      "num_tokens": 29408130.0,
      "step": 16950
    },
    {
      "entropy": 5.617013835906983,
      "epoch": 1.31915969655709,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004828431848220205,
      "loss": 5.0769,
      "mean_token_accuracy": 0.19123691767454148,
      "num_tokens": 29417446.0,
      "step": 16955
    },
    {
      "entropy": 5.649213027954102,
      "epoch": 1.3195487259288077,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004828325681180343,
      "loss": 5.1492,
      "mean_token_accuracy": 0.19348373264074326,
      "num_tokens": 29426023.0,
      "step": 16960
    },
    {
      "entropy": 5.621719026565552,
      "epoch": 1.3199377553005252,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004828219482605197,
      "loss": 5.0658,
      "mean_token_accuracy": 0.194810451567173,
      "num_tokens": 29434737.0,
      "step": 16965
    },
    {
      "entropy": 5.554741144180298,
      "epoch": 1.3203267846722428,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048281132524963786,
      "loss": 5.0852,
      "mean_token_accuracy": 0.18866194933652877,
      "num_tokens": 29443926.0,
      "step": 16970
    },
    {
      "entropy": 5.5818909168243405,
      "epoch": 1.3207158140439603,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004828006990855499,
      "loss": 5.082,
      "mean_token_accuracy": 0.19597212225198746,
      "num_tokens": 29452495.0,
      "step": 16975
    },
    {
      "entropy": 5.557660484313965,
      "epoch": 1.3211048434156778,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048279006976841704,
      "loss": 5.0463,
      "mean_token_accuracy": 0.19332642257213592,
      "num_tokens": 29460661.0,
      "step": 16980
    },
    {
      "entropy": 5.584459733963013,
      "epoch": 1.3214938727873955,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004827794372984007,
      "loss": 5.0413,
      "mean_token_accuracy": 0.19895520955324172,
      "num_tokens": 29469155.0,
      "step": 16985
    },
    {
      "entropy": 5.636346054077149,
      "epoch": 1.321882902159113,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000482768801675662,
      "loss": 5.1516,
      "mean_token_accuracy": 0.19496320486068724,
      "num_tokens": 29478220.0,
      "step": 16990
    },
    {
      "entropy": 5.606045913696289,
      "epoch": 1.3222719315308304,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004827581629003625,
      "loss": 5.1037,
      "mean_token_accuracy": 0.19156274050474167,
      "num_tokens": 29486336.0,
      "step": 16995
    },
    {
      "entropy": 5.654403591156006,
      "epoch": 1.3226609609025481,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048274752097266356,
      "loss": 5.2119,
      "mean_token_accuracy": 0.18937937766313553,
      "num_tokens": 29495150.0,
      "step": 17000
    },
    {
      "entropy": 5.626471328735351,
      "epoch": 1.3230499902742658,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004827368758927266,
      "loss": 5.0915,
      "mean_token_accuracy": 0.1930074691772461,
      "num_tokens": 29503508.0,
      "step": 17005
    },
    {
      "entropy": 5.571884632110596,
      "epoch": 1.3234390196459833,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004827262276607132,
      "loss": 5.0703,
      "mean_token_accuracy": 0.19201896339654922,
      "num_tokens": 29513038.0,
      "step": 17010
    },
    {
      "entropy": 5.560047960281372,
      "epoch": 1.3238280490177008,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000482715576276785,
      "loss": 5.0374,
      "mean_token_accuracy": 0.19725186377763748,
      "num_tokens": 29521344.0,
      "step": 17015
    },
    {
      "entropy": 5.5897095680236815,
      "epoch": 1.3242170783894185,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004827049217411035,
      "loss": 5.1529,
      "mean_token_accuracy": 0.18654258400201798,
      "num_tokens": 29530109.0,
      "step": 17020
    },
    {
      "entropy": 5.615104150772095,
      "epoch": 1.324606107761136,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048269426405383043,
      "loss": 5.0741,
      "mean_token_accuracy": 0.18548837006092073,
      "num_tokens": 29539114.0,
      "step": 17025
    },
    {
      "entropy": 5.609760284423828,
      "epoch": 1.3249951371328534,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048268360321512745,
      "loss": 5.1188,
      "mean_token_accuracy": 0.19076967090368271,
      "num_tokens": 29548746.0,
      "step": 17030
    },
    {
      "entropy": 5.63235182762146,
      "epoch": 1.3253841665045711,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004826729392251564,
      "loss": 5.1562,
      "mean_token_accuracy": 0.18577359467744828,
      "num_tokens": 29557681.0,
      "step": 17035
    },
    {
      "entropy": 5.6283042430877686,
      "epoch": 1.3257731958762886,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000482662272084079,
      "loss": 5.1407,
      "mean_token_accuracy": 0.19394930452108383,
      "num_tokens": 29566151.0,
      "step": 17040
    },
    {
      "entropy": 5.558325672149659,
      "epoch": 1.326162225248006,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004826516017920571,
      "loss": 4.9518,
      "mean_token_accuracy": 0.20908219814300538,
      "num_tokens": 29574853.0,
      "step": 17045
    },
    {
      "entropy": 5.56689190864563,
      "epoch": 1.3265512546197238,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004826409283492528,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19533458203077317,
      "num_tokens": 29583319.0,
      "step": 17050
    },
    {
      "entropy": 5.5869800567626955,
      "epoch": 1.3269402839914415,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048263025175582783,
      "loss": 5.1499,
      "mean_token_accuracy": 0.1880865439772606,
      "num_tokens": 29591949.0,
      "step": 17055
    },
    {
      "entropy": 5.6449895858764645,
      "epoch": 1.327329313363159,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004826195720119442,
      "loss": 5.1545,
      "mean_token_accuracy": 0.18147476613521576,
      "num_tokens": 29600293.0,
      "step": 17060
    },
    {
      "entropy": 5.657760190963745,
      "epoch": 1.3277183427348764,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004826088891177641,
      "loss": 5.1738,
      "mean_token_accuracy": 0.1958606019616127,
      "num_tokens": 29608602.0,
      "step": 17065
    },
    {
      "entropy": 5.581202125549316,
      "epoch": 1.3281073721065941,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048259820307344954,
      "loss": 5.134,
      "mean_token_accuracy": 0.19184042662382125,
      "num_tokens": 29617464.0,
      "step": 17070
    },
    {
      "entropy": 5.619830656051636,
      "epoch": 1.3284964014783116,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048258751387916267,
      "loss": 5.0386,
      "mean_token_accuracy": 0.20158812552690505,
      "num_tokens": 29626297.0,
      "step": 17075
    },
    {
      "entropy": 5.554988861083984,
      "epoch": 1.328885430850029,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048257682153506564,
      "loss": 5.0658,
      "mean_token_accuracy": 0.19179969131946564,
      "num_tokens": 29635085.0,
      "step": 17080
    },
    {
      "entropy": 5.552317905426025,
      "epoch": 1.3292744602217468,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004825661260413208,
      "loss": 5.051,
      "mean_token_accuracy": 0.1966339558362961,
      "num_tokens": 29643528.0,
      "step": 17085
    },
    {
      "entropy": 5.560385084152221,
      "epoch": 1.3296634895934643,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048255542739809035,
      "loss": 5.1091,
      "mean_token_accuracy": 0.1915844276547432,
      "num_tokens": 29651883.0,
      "step": 17090
    },
    {
      "entropy": 5.554691791534424,
      "epoch": 1.330052518965182,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048254472560553665,
      "loss": 5.0973,
      "mean_token_accuracy": 0.18957026898860932,
      "num_tokens": 29661092.0,
      "step": 17095
    },
    {
      "entropy": 5.601039552688599,
      "epoch": 1.3304415483368994,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048253402066382207,
      "loss": 5.1307,
      "mean_token_accuracy": 0.1981351912021637,
      "num_tokens": 29670285.0,
      "step": 17100
    },
    {
      "entropy": 5.637584638595581,
      "epoch": 1.3308305777086171,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004825233125731091,
      "loss": 5.1747,
      "mean_token_accuracy": 0.18887474089860917,
      "num_tokens": 29679249.0,
      "step": 17105
    },
    {
      "entropy": 5.6272870063781735,
      "epoch": 1.3312196070803346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048251260133356014,
      "loss": 5.1446,
      "mean_token_accuracy": 0.1861952617764473,
      "num_tokens": 29687978.0,
      "step": 17110
    },
    {
      "entropy": 5.59926118850708,
      "epoch": 1.331608636452052,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048250188694533774,
      "loss": 5.162,
      "mean_token_accuracy": 0.18740419149398804,
      "num_tokens": 29695972.0,
      "step": 17115
    },
    {
      "entropy": 5.626380062103271,
      "epoch": 1.3319976658237698,
      "grad_norm": 1.0,
      "learning_rate": 0.0004824911694086044,
      "loss": 4.9934,
      "mean_token_accuracy": 0.1966044545173645,
      "num_tokens": 29704997.0,
      "step": 17120
    },
    {
      "entropy": 5.707925271987915,
      "epoch": 1.3323866951954872,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000482480448723523,
      "loss": 5.2219,
      "mean_token_accuracy": 0.19110133349895478,
      "num_tokens": 29713957.0,
      "step": 17125
    },
    {
      "entropy": 5.640682506561279,
      "epoch": 1.3327757245672047,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048246972489025586,
      "loss": 5.115,
      "mean_token_accuracy": 0.19240255206823348,
      "num_tokens": 29721989.0,
      "step": 17130
    },
    {
      "entropy": 5.591029119491577,
      "epoch": 1.3331647539389224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048245899790896597,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19880155175924302,
      "num_tokens": 29730032.0,
      "step": 17135
    },
    {
      "entropy": 5.581392478942871,
      "epoch": 1.33355378331064,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004824482677798159,
      "loss": 5.0597,
      "mean_token_accuracy": 0.19574938714504242,
      "num_tokens": 29738545.0,
      "step": 17140
    },
    {
      "entropy": 5.626249551773071,
      "epoch": 1.3339428126823576,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048243753450296863,
      "loss": 5.0789,
      "mean_token_accuracy": 0.1984911873936653,
      "num_tokens": 29747371.0,
      "step": 17145
    },
    {
      "entropy": 5.592475748062133,
      "epoch": 1.334331842054075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048242679807858693,
      "loss": 5.0233,
      "mean_token_accuracy": 0.19690910279750823,
      "num_tokens": 29756565.0,
      "step": 17150
    },
    {
      "entropy": 5.540616273880005,
      "epoch": 1.3347208714257928,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048241605850683365,
      "loss": 5.0405,
      "mean_token_accuracy": 0.1935047596693039,
      "num_tokens": 29765846.0,
      "step": 17155
    },
    {
      "entropy": 5.601731586456299,
      "epoch": 1.3351099007975102,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048240531578787197,
      "loss": 5.0869,
      "mean_token_accuracy": 0.19569143950939177,
      "num_tokens": 29774222.0,
      "step": 17160
    },
    {
      "entropy": 5.565546274185181,
      "epoch": 1.3354989301692277,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004823945699218647,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20037296116352082,
      "num_tokens": 29782366.0,
      "step": 17165
    },
    {
      "entropy": 5.6020973205566404,
      "epoch": 1.3358879595409454,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004823838209089749,
      "loss": 5.1308,
      "mean_token_accuracy": 0.18899450004100798,
      "num_tokens": 29792147.0,
      "step": 17170
    },
    {
      "entropy": 5.615554237365723,
      "epoch": 1.3362769889126629,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048237306874936565,
      "loss": 5.0882,
      "mean_token_accuracy": 0.19848169833421708,
      "num_tokens": 29801325.0,
      "step": 17175
    },
    {
      "entropy": 5.579449605941773,
      "epoch": 1.3366660182843804,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004823623134432001,
      "loss": 5.0531,
      "mean_token_accuracy": 0.19223212003707885,
      "num_tokens": 29809353.0,
      "step": 17180
    },
    {
      "entropy": 5.566813325881958,
      "epoch": 1.337055047656098,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048235155499064166,
      "loss": 5.0997,
      "mean_token_accuracy": 0.19648808240890503,
      "num_tokens": 29817746.0,
      "step": 17185
    },
    {
      "entropy": 5.562612056732178,
      "epoch": 1.3374440770278155,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048234079339185335,
      "loss": 5.118,
      "mean_token_accuracy": 0.19193292707204818,
      "num_tokens": 29826468.0,
      "step": 17190
    },
    {
      "entropy": 5.651003646850586,
      "epoch": 1.3378331063995332,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048233002864699856,
      "loss": 5.1846,
      "mean_token_accuracy": 0.18184355199337005,
      "num_tokens": 29835051.0,
      "step": 17195
    },
    {
      "entropy": 5.632854032516479,
      "epoch": 1.3382221357712507,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004823192607562405,
      "loss": 5.081,
      "mean_token_accuracy": 0.19087070524692534,
      "num_tokens": 29843514.0,
      "step": 17200
    },
    {
      "entropy": 5.591190910339355,
      "epoch": 1.3386111651429684,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048230848971974265,
      "loss": 5.01,
      "mean_token_accuracy": 0.19248565584421157,
      "num_tokens": 29851861.0,
      "step": 17205
    },
    {
      "entropy": 5.5409363269805905,
      "epoch": 1.3390001945146859,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004822977155376684,
      "loss": 5.0213,
      "mean_token_accuracy": 0.19842875599861146,
      "num_tokens": 29860639.0,
      "step": 17210
    },
    {
      "entropy": 5.550060033798218,
      "epoch": 1.3393892238864034,
      "grad_norm": 1.125,
      "learning_rate": 0.0004822869382101814,
      "loss": 5.1883,
      "mean_token_accuracy": 0.18755810260772704,
      "num_tokens": 29869931.0,
      "step": 17215
    },
    {
      "entropy": 5.63261775970459,
      "epoch": 1.339778253258121,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004822761577374449,
      "loss": 5.1042,
      "mean_token_accuracy": 0.19078442007303237,
      "num_tokens": 29878933.0,
      "step": 17220
    },
    {
      "entropy": 5.607994461059571,
      "epoch": 1.3401672826298385,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004822653741196227,
      "loss": 5.0956,
      "mean_token_accuracy": 0.1942954868078232,
      "num_tokens": 29888031.0,
      "step": 17225
    },
    {
      "entropy": 5.574734401702881,
      "epoch": 1.340556312001556,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004822545873568783,
      "loss": 5.0562,
      "mean_token_accuracy": 0.19396908581256866,
      "num_tokens": 29896792.0,
      "step": 17230
    },
    {
      "entropy": 5.551525640487671,
      "epoch": 1.3409453413732737,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048224379744937545,
      "loss": 5.0441,
      "mean_token_accuracy": 0.19622675478458404,
      "num_tokens": 29905185.0,
      "step": 17235
    },
    {
      "entropy": 5.652295637130737,
      "epoch": 1.3413343707449912,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048223300439727783,
      "loss": 5.1544,
      "mean_token_accuracy": 0.18267314434051513,
      "num_tokens": 29913289.0,
      "step": 17240
    },
    {
      "entropy": 5.583216190338135,
      "epoch": 1.3417234001167089,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048222220820074926,
      "loss": 5.1015,
      "mean_token_accuracy": 0.1957106590270996,
      "num_tokens": 29922112.0,
      "step": 17245
    },
    {
      "entropy": 5.626125860214233,
      "epoch": 1.3421124294884264,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048221140885995346,
      "loss": 5.0849,
      "mean_token_accuracy": 0.1889687806367874,
      "num_tokens": 29930944.0,
      "step": 17250
    },
    {
      "entropy": 5.665236186981201,
      "epoch": 1.342501458860144,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004822006063750543,
      "loss": 5.1933,
      "mean_token_accuracy": 0.18412038832902908,
      "num_tokens": 29938576.0,
      "step": 17255
    },
    {
      "entropy": 5.588354682922363,
      "epoch": 1.3428904882318615,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048218980074621575,
      "loss": 5.0862,
      "mean_token_accuracy": 0.19173820912837983,
      "num_tokens": 29946771.0,
      "step": 17260
    },
    {
      "entropy": 5.6391833305358885,
      "epoch": 1.343279517603579,
      "grad_norm": 1.125,
      "learning_rate": 0.00048217899197360183,
      "loss": 5.1065,
      "mean_token_accuracy": 0.19124213308095933,
      "num_tokens": 29955098.0,
      "step": 17265
    },
    {
      "entropy": 5.576250219345093,
      "epoch": 1.3436685469752967,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004821681800573764,
      "loss": 5.1183,
      "mean_token_accuracy": 0.18985211104154587,
      "num_tokens": 29964538.0,
      "step": 17270
    },
    {
      "entropy": 5.586406278610229,
      "epoch": 1.3440575763470142,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004821573649977036,
      "loss": 5.0529,
      "mean_token_accuracy": 0.19566859155893326,
      "num_tokens": 29973965.0,
      "step": 17275
    },
    {
      "entropy": 5.5091344833374025,
      "epoch": 1.3444466057187316,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048214654679474753,
      "loss": 4.9762,
      "mean_token_accuracy": 0.19796336591243743,
      "num_tokens": 29982742.0,
      "step": 17280
    },
    {
      "entropy": 5.527531623840332,
      "epoch": 1.3448356350904493,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048213572544867224,
      "loss": 5.0483,
      "mean_token_accuracy": 0.19206763654947281,
      "num_tokens": 29991846.0,
      "step": 17285
    },
    {
      "entropy": 5.631241130828857,
      "epoch": 1.3452246644621668,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048212490095964213,
      "loss": 5.0025,
      "mean_token_accuracy": 0.20127080231904984,
      "num_tokens": 29999784.0,
      "step": 17290
    },
    {
      "entropy": 5.573941898345947,
      "epoch": 1.3456136938338845,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004821140733278212,
      "loss": 5.1206,
      "mean_token_accuracy": 0.192512209713459,
      "num_tokens": 30008139.0,
      "step": 17295
    },
    {
      "entropy": 5.596026515960693,
      "epoch": 1.346002723205602,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004821032425533739,
      "loss": 5.1149,
      "mean_token_accuracy": 0.1973307266831398,
      "num_tokens": 30016891.0,
      "step": 17300
    },
    {
      "entropy": 5.5896257877349855,
      "epoch": 1.3463917525773197,
      "grad_norm": 1.125,
      "learning_rate": 0.0004820924086364646,
      "loss": 5.0976,
      "mean_token_accuracy": 0.1946762517094612,
      "num_tokens": 30025472.0,
      "step": 17305
    },
    {
      "entropy": 5.624167728424072,
      "epoch": 1.3467807819490372,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048208157157725756,
      "loss": 5.0983,
      "mean_token_accuracy": 0.19798196107149124,
      "num_tokens": 30034088.0,
      "step": 17310
    },
    {
      "entropy": 5.5352325439453125,
      "epoch": 1.3471698113207546,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048207073137591726,
      "loss": 5.0417,
      "mean_token_accuracy": 0.192068450152874,
      "num_tokens": 30042651.0,
      "step": 17315
    },
    {
      "entropy": 5.580952596664429,
      "epoch": 1.3475588406924723,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048205988803260824,
      "loss": 5.0011,
      "mean_token_accuracy": 0.2002151608467102,
      "num_tokens": 30050898.0,
      "step": 17320
    },
    {
      "entropy": 5.677296829223633,
      "epoch": 1.3479478700641898,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048204904154749496,
      "loss": 5.1725,
      "mean_token_accuracy": 0.1894117623567581,
      "num_tokens": 30059357.0,
      "step": 17325
    },
    {
      "entropy": 5.60735297203064,
      "epoch": 1.3483368994359073,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048203819192074206,
      "loss": 5.0754,
      "mean_token_accuracy": 0.1892850935459137,
      "num_tokens": 30068137.0,
      "step": 17330
    },
    {
      "entropy": 5.600775814056396,
      "epoch": 1.348725928807625,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048202733915251407,
      "loss": 5.1299,
      "mean_token_accuracy": 0.195503132045269,
      "num_tokens": 30077139.0,
      "step": 17335
    },
    {
      "entropy": 5.633728885650635,
      "epoch": 1.3491149581793425,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048201648324297573,
      "loss": 5.1307,
      "mean_token_accuracy": 0.20009224861860275,
      "num_tokens": 30085267.0,
      "step": 17340
    },
    {
      "entropy": 5.570550918579102,
      "epoch": 1.3495039875510602,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048200562419229185,
      "loss": 5.0517,
      "mean_token_accuracy": 0.20046866983175277,
      "num_tokens": 30093804.0,
      "step": 17345
    },
    {
      "entropy": 5.5981920719146725,
      "epoch": 1.3498930169227776,
      "grad_norm": 1.046875,
      "learning_rate": 0.000481994762000627,
      "loss": 5.0327,
      "mean_token_accuracy": 0.20512161552906036,
      "num_tokens": 30102653.0,
      "step": 17350
    },
    {
      "entropy": 5.534904336929321,
      "epoch": 1.3502820462944953,
      "grad_norm": 1.125,
      "learning_rate": 0.00048198389666814616,
      "loss": 5.0145,
      "mean_token_accuracy": 0.19998365193605422,
      "num_tokens": 30111978.0,
      "step": 17355
    },
    {
      "entropy": 5.623388576507568,
      "epoch": 1.3506710756662128,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048197302819501416,
      "loss": 5.2542,
      "mean_token_accuracy": 0.18305663168430328,
      "num_tokens": 30120387.0,
      "step": 17360
    },
    {
      "entropy": 5.601443147659301,
      "epoch": 1.3510601050379303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048196215658139586,
      "loss": 5.0984,
      "mean_token_accuracy": 0.19018661379814147,
      "num_tokens": 30129069.0,
      "step": 17365
    },
    {
      "entropy": 5.5916832447052,
      "epoch": 1.351449134409648,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004819512818274562,
      "loss": 5.0703,
      "mean_token_accuracy": 0.19169774949550628,
      "num_tokens": 30138077.0,
      "step": 17370
    },
    {
      "entropy": 5.554825639724731,
      "epoch": 1.3518381637813655,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004819404039333603,
      "loss": 5.0164,
      "mean_token_accuracy": 0.19625434875488282,
      "num_tokens": 30146189.0,
      "step": 17375
    },
    {
      "entropy": 5.596820497512818,
      "epoch": 1.352227193153083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004819295228992731,
      "loss": 5.1441,
      "mean_token_accuracy": 0.18551240712404252,
      "num_tokens": 30155664.0,
      "step": 17380
    },
    {
      "entropy": 5.621693563461304,
      "epoch": 1.3526162225248006,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048191863872535984,
      "loss": 5.0982,
      "mean_token_accuracy": 0.19399722069501876,
      "num_tokens": 30164701.0,
      "step": 17385
    },
    {
      "entropy": 5.6402098655700685,
      "epoch": 1.3530052518965183,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004819077514117855,
      "loss": 5.1395,
      "mean_token_accuracy": 0.19072800129652023,
      "num_tokens": 30173916.0,
      "step": 17390
    },
    {
      "entropy": 5.588237047195435,
      "epoch": 1.3533942812682358,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048189686095871545,
      "loss": 5.0708,
      "mean_token_accuracy": 0.19412756562232972,
      "num_tokens": 30182803.0,
      "step": 17395
    },
    {
      "entropy": 5.5683369636535645,
      "epoch": 1.3537833106399533,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048188596736631475,
      "loss": 5.0786,
      "mean_token_accuracy": 0.20161097794771193,
      "num_tokens": 30191589.0,
      "step": 17400
    },
    {
      "entropy": 5.605751371383667,
      "epoch": 1.354172340011671,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048187507063474885,
      "loss": 5.1785,
      "mean_token_accuracy": 0.1880600258708,
      "num_tokens": 30200170.0,
      "step": 17405
    },
    {
      "entropy": 5.626162481307984,
      "epoch": 1.3545613693833884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048186417076418294,
      "loss": 5.0419,
      "mean_token_accuracy": 0.19382455348968505,
      "num_tokens": 30208683.0,
      "step": 17410
    },
    {
      "entropy": 5.589068412780762,
      "epoch": 1.354950398755106,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048185326775478256,
      "loss": 5.0985,
      "mean_token_accuracy": 0.1832721158862114,
      "num_tokens": 30217141.0,
      "step": 17415
    },
    {
      "entropy": 5.582963228225708,
      "epoch": 1.3553394281268236,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048184236160671306,
      "loss": 5.0901,
      "mean_token_accuracy": 0.18514712750911713,
      "num_tokens": 30226759.0,
      "step": 17420
    },
    {
      "entropy": 5.616628122329712,
      "epoch": 1.355728457498541,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048183145232013995,
      "loss": 5.1576,
      "mean_token_accuracy": 0.19223590493202208,
      "num_tokens": 30235945.0,
      "step": 17425
    },
    {
      "entropy": 5.620598363876343,
      "epoch": 1.3561174868702586,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048182053989522883,
      "loss": 5.1247,
      "mean_token_accuracy": 0.18232114762067794,
      "num_tokens": 30244912.0,
      "step": 17430
    },
    {
      "entropy": 5.6058262348175045,
      "epoch": 1.3565065162419763,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048180962433214515,
      "loss": 5.0715,
      "mean_token_accuracy": 0.19382267892360688,
      "num_tokens": 30253807.0,
      "step": 17435
    },
    {
      "entropy": 5.615492296218872,
      "epoch": 1.356895545613694,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048179870563105453,
      "loss": 5.172,
      "mean_token_accuracy": 0.1914404660463333,
      "num_tokens": 30262380.0,
      "step": 17440
    },
    {
      "entropy": 5.655902481079101,
      "epoch": 1.3572845749854114,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004817877837921228,
      "loss": 5.1003,
      "mean_token_accuracy": 0.19246790558099747,
      "num_tokens": 30271374.0,
      "step": 17445
    },
    {
      "entropy": 5.601302671432495,
      "epoch": 1.357673604357129,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004817768588155155,
      "loss": 5.1328,
      "mean_token_accuracy": 0.19475583285093306,
      "num_tokens": 30280037.0,
      "step": 17450
    },
    {
      "entropy": 5.531799364089966,
      "epoch": 1.3580626337288466,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048176593070139865,
      "loss": 4.9348,
      "mean_token_accuracy": 0.21244002878665924,
      "num_tokens": 30288335.0,
      "step": 17455
    },
    {
      "entropy": 5.519866943359375,
      "epoch": 1.358451663100564,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048175499944993767,
      "loss": 5.0368,
      "mean_token_accuracy": 0.19410879909992218,
      "num_tokens": 30297109.0,
      "step": 17460
    },
    {
      "entropy": 5.580335521697998,
      "epoch": 1.3588406924722816,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048174406506129884,
      "loss": 5.0509,
      "mean_token_accuracy": 0.19508031755685806,
      "num_tokens": 30305594.0,
      "step": 17465
    },
    {
      "entropy": 5.542119836807251,
      "epoch": 1.3592297218439993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048173312753564787,
      "loss": 4.9816,
      "mean_token_accuracy": 0.20457783639431,
      "num_tokens": 30313106.0,
      "step": 17470
    },
    {
      "entropy": 5.576837539672852,
      "epoch": 1.3596187512157167,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004817221868731506,
      "loss": 5.17,
      "mean_token_accuracy": 0.1869833990931511,
      "num_tokens": 30322194.0,
      "step": 17475
    },
    {
      "entropy": 5.53590497970581,
      "epoch": 1.3600077805874344,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048171124307397334,
      "loss": 5.0174,
      "mean_token_accuracy": 0.19708539694547653,
      "num_tokens": 30330247.0,
      "step": 17480
    },
    {
      "entropy": 5.619177913665771,
      "epoch": 1.360396809959152,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004817002961382819,
      "loss": 5.1134,
      "mean_token_accuracy": 0.19212924540042878,
      "num_tokens": 30338738.0,
      "step": 17485
    },
    {
      "entropy": 5.576988649368286,
      "epoch": 1.3607858393308696,
      "grad_norm": 1.125,
      "learning_rate": 0.00048168934606624255,
      "loss": 5.0908,
      "mean_token_accuracy": 0.19094870686531068,
      "num_tokens": 30347729.0,
      "step": 17490
    },
    {
      "entropy": 5.62130708694458,
      "epoch": 1.361174868702587,
      "grad_norm": 1.25,
      "learning_rate": 0.00048167839285802116,
      "loss": 5.1002,
      "mean_token_accuracy": 0.1885942116379738,
      "num_tokens": 30356408.0,
      "step": 17495
    },
    {
      "entropy": 5.5848783493042,
      "epoch": 1.3615638980743046,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004816674365137843,
      "loss": 5.0646,
      "mean_token_accuracy": 0.19244394302368165,
      "num_tokens": 30364910.0,
      "step": 17500
    },
    {
      "entropy": 5.557191944122314,
      "epoch": 1.3619529274460223,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000481656477033698,
      "loss": 5.0823,
      "mean_token_accuracy": 0.1955784320831299,
      "num_tokens": 30373882.0,
      "step": 17505
    },
    {
      "entropy": 5.643977117538452,
      "epoch": 1.3623419568177397,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004816455144179286,
      "loss": 5.1402,
      "mean_token_accuracy": 0.1846423327922821,
      "num_tokens": 30382884.0,
      "step": 17510
    },
    {
      "entropy": 5.592962169647217,
      "epoch": 1.3627309861894572,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004816345486666424,
      "loss": 5.0437,
      "mean_token_accuracy": 0.1940524697303772,
      "num_tokens": 30391798.0,
      "step": 17515
    },
    {
      "entropy": 5.583971691131592,
      "epoch": 1.363120015561175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004816235797800058,
      "loss": 5.0603,
      "mean_token_accuracy": 0.19570833891630174,
      "num_tokens": 30400194.0,
      "step": 17520
    },
    {
      "entropy": 5.569255828857422,
      "epoch": 1.3635090449328924,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004816126077581852,
      "loss": 4.9913,
      "mean_token_accuracy": 0.19892526268959046,
      "num_tokens": 30408653.0,
      "step": 17525
    },
    {
      "entropy": 5.550200891494751,
      "epoch": 1.36389807430461,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048160163260134724,
      "loss": 5.1267,
      "mean_token_accuracy": 0.19007736295461655,
      "num_tokens": 30417467.0,
      "step": 17530
    },
    {
      "entropy": 5.5903106212615965,
      "epoch": 1.3642871036763276,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004815906543096583,
      "loss": 5.0976,
      "mean_token_accuracy": 0.18592511117458344,
      "num_tokens": 30425993.0,
      "step": 17535
    },
    {
      "entropy": 5.634143924713134,
      "epoch": 1.3646761330480452,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000481579672883285,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19496258944272996,
      "num_tokens": 30434809.0,
      "step": 17540
    },
    {
      "entropy": 5.601830720901489,
      "epoch": 1.3650651624197627,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048156868832239396,
      "loss": 5.0998,
      "mean_token_accuracy": 0.1941066637635231,
      "num_tokens": 30443536.0,
      "step": 17545
    },
    {
      "entropy": 5.518461418151856,
      "epoch": 1.3654541917914802,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004815577006271519,
      "loss": 5.0835,
      "mean_token_accuracy": 0.19164232313632965,
      "num_tokens": 30452163.0,
      "step": 17550
    },
    {
      "entropy": 5.563957834243775,
      "epoch": 1.365843221163198,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048154670979772555,
      "loss": 5.0576,
      "mean_token_accuracy": 0.19976511895656585,
      "num_tokens": 30461234.0,
      "step": 17555
    },
    {
      "entropy": 5.725088453292846,
      "epoch": 1.3662322505349154,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004815357158342815,
      "loss": 5.2178,
      "mean_token_accuracy": 0.17882028073072434,
      "num_tokens": 30470438.0,
      "step": 17560
    },
    {
      "entropy": 5.639098882675171,
      "epoch": 1.3666212799066328,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004815247187369868,
      "loss": 5.0081,
      "mean_token_accuracy": 0.19483415633440018,
      "num_tokens": 30479454.0,
      "step": 17565
    },
    {
      "entropy": 5.632966947555542,
      "epoch": 1.3670103092783505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048151371850600814,
      "loss": 5.1187,
      "mean_token_accuracy": 0.189105848968029,
      "num_tokens": 30487968.0,
      "step": 17570
    },
    {
      "entropy": 5.595620012283325,
      "epoch": 1.367399338650068,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048150271514151255,
      "loss": 5.0474,
      "mean_token_accuracy": 0.1967533975839615,
      "num_tokens": 30495712.0,
      "step": 17575
    },
    {
      "entropy": 5.605516576766968,
      "epoch": 1.3677883680217857,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048149170864366693,
      "loss": 5.1447,
      "mean_token_accuracy": 0.18511847406625748,
      "num_tokens": 30505008.0,
      "step": 17580
    },
    {
      "entropy": 5.612831735610962,
      "epoch": 1.3681773973935032,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048148069901263836,
      "loss": 5.0421,
      "mean_token_accuracy": 0.19839342981576918,
      "num_tokens": 30514496.0,
      "step": 17585
    },
    {
      "entropy": 5.590772294998169,
      "epoch": 1.368566426765221,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048146968624859375,
      "loss": 5.0267,
      "mean_token_accuracy": 0.19972210973501206,
      "num_tokens": 30522942.0,
      "step": 17590
    },
    {
      "entropy": 5.658151483535766,
      "epoch": 1.3689554561369384,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004814586703517003,
      "loss": 5.1894,
      "mean_token_accuracy": 0.18800529688596726,
      "num_tokens": 30532124.0,
      "step": 17595
    },
    {
      "entropy": 5.671984100341797,
      "epoch": 1.3693444855086558,
      "grad_norm": 1.125,
      "learning_rate": 0.0004814476513221251,
      "loss": 5.2975,
      "mean_token_accuracy": 0.17737707495689392,
      "num_tokens": 30541063.0,
      "step": 17600
    },
    {
      "entropy": 5.664184284210205,
      "epoch": 1.3697335148803735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048143662916003545,
      "loss": 5.0155,
      "mean_token_accuracy": 0.19651644676923752,
      "num_tokens": 30549062.0,
      "step": 17605
    },
    {
      "entropy": 5.65010347366333,
      "epoch": 1.370122544252091,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004814256038655985,
      "loss": 5.1154,
      "mean_token_accuracy": 0.18934282064437866,
      "num_tokens": 30557211.0,
      "step": 17610
    },
    {
      "entropy": 5.613674211502075,
      "epoch": 1.3705115736238085,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004814145754389816,
      "loss": 5.1701,
      "mean_token_accuracy": 0.18557561188936234,
      "num_tokens": 30566603.0,
      "step": 17615
    },
    {
      "entropy": 5.574848222732544,
      "epoch": 1.3709006029955262,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048140354388035204,
      "loss": 5.0494,
      "mean_token_accuracy": 0.1934528112411499,
      "num_tokens": 30574903.0,
      "step": 17620
    },
    {
      "entropy": 5.661852407455444,
      "epoch": 1.3712896323672437,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004813925091898772,
      "loss": 5.1385,
      "mean_token_accuracy": 0.18710356801748276,
      "num_tokens": 30584925.0,
      "step": 17625
    },
    {
      "entropy": 5.688584709167481,
      "epoch": 1.3716786617389614,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004813814713677246,
      "loss": 5.1887,
      "mean_token_accuracy": 0.1886507675051689,
      "num_tokens": 30593376.0,
      "step": 17630
    },
    {
      "entropy": 5.611472225189209,
      "epoch": 1.3720676911106788,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004813704304140616,
      "loss": 5.1142,
      "mean_token_accuracy": 0.19458251148462297,
      "num_tokens": 30602306.0,
      "step": 17635
    },
    {
      "entropy": 5.556731748580932,
      "epoch": 1.3724567204823965,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004813593863290559,
      "loss": 5.0566,
      "mean_token_accuracy": 0.1932797759771347,
      "num_tokens": 30611340.0,
      "step": 17640
    },
    {
      "entropy": 5.617712688446045,
      "epoch": 1.372845749854114,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004813483391128748,
      "loss": 5.1591,
      "mean_token_accuracy": 0.19261574894189834,
      "num_tokens": 30619507.0,
      "step": 17645
    },
    {
      "entropy": 5.639880228042602,
      "epoch": 1.3732347792258315,
      "grad_norm": 1.0,
      "learning_rate": 0.0004813372887656862,
      "loss": 5.1521,
      "mean_token_accuracy": 0.1923792615532875,
      "num_tokens": 30629897.0,
      "step": 17650
    },
    {
      "entropy": 5.6188194274902346,
      "epoch": 1.3736238085975492,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004813262352876576,
      "loss": 5.0936,
      "mean_token_accuracy": 0.19418303966522216,
      "num_tokens": 30638517.0,
      "step": 17655
    },
    {
      "entropy": 5.646220588684082,
      "epoch": 1.3740128379692667,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048131517867895684,
      "loss": 5.1397,
      "mean_token_accuracy": 0.19555290937423705,
      "num_tokens": 30646816.0,
      "step": 17660
    },
    {
      "entropy": 5.628312063217163,
      "epoch": 1.3744018673409841,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048130411893975155,
      "loss": 5.0978,
      "mean_token_accuracy": 0.19647861868143082,
      "num_tokens": 30655199.0,
      "step": 17665
    },
    {
      "entropy": 5.557996082305908,
      "epoch": 1.3747908967127018,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004812930560702097,
      "loss": 4.9805,
      "mean_token_accuracy": 0.19906099885702133,
      "num_tokens": 30663562.0,
      "step": 17670
    },
    {
      "entropy": 5.60725769996643,
      "epoch": 1.3751799260844193,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048128199007049907,
      "loss": 5.1464,
      "mean_token_accuracy": 0.18604770302772522,
      "num_tokens": 30672611.0,
      "step": 17675
    },
    {
      "entropy": 5.591836309432983,
      "epoch": 1.375568955456137,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048127092094078756,
      "loss": 5.0758,
      "mean_token_accuracy": 0.19608640819787979,
      "num_tokens": 30680766.0,
      "step": 17680
    },
    {
      "entropy": 5.544107103347779,
      "epoch": 1.3759579848278545,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048125984868124317,
      "loss": 4.9433,
      "mean_token_accuracy": 0.19995449334383011,
      "num_tokens": 30689192.0,
      "step": 17685
    },
    {
      "entropy": 5.576938152313232,
      "epoch": 1.3763470141995722,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004812487732920338,
      "loss": 5.0601,
      "mean_token_accuracy": 0.1913628935813904,
      "num_tokens": 30697889.0,
      "step": 17690
    },
    {
      "entropy": 5.656646203994751,
      "epoch": 1.3767360435712896,
      "grad_norm": 1.125,
      "learning_rate": 0.0004812376947733277,
      "loss": 5.1042,
      "mean_token_accuracy": 0.19301426857709886,
      "num_tokens": 30706285.0,
      "step": 17695
    },
    {
      "entropy": 5.59697995185852,
      "epoch": 1.3771250729430071,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048122661312529263,
      "loss": 5.0701,
      "mean_token_accuracy": 0.1972261354327202,
      "num_tokens": 30715159.0,
      "step": 17700
    },
    {
      "entropy": 5.544631671905518,
      "epoch": 1.3775141023147248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004812155283480971,
      "loss": 4.9839,
      "mean_token_accuracy": 0.19768782705068588,
      "num_tokens": 30723133.0,
      "step": 17705
    },
    {
      "entropy": 5.658510971069336,
      "epoch": 1.3779031316864423,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004812044404419091,
      "loss": 5.1674,
      "mean_token_accuracy": 0.19228937476873398,
      "num_tokens": 30732272.0,
      "step": 17710
    },
    {
      "entropy": 5.620146799087524,
      "epoch": 1.3782921610581598,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000481193349406897,
      "loss": 5.0317,
      "mean_token_accuracy": 0.19477264136075972,
      "num_tokens": 30740439.0,
      "step": 17715
    },
    {
      "entropy": 5.56381368637085,
      "epoch": 1.3786811904298775,
      "grad_norm": 1.0625,
      "learning_rate": 0.000481182255243229,
      "loss": 5.0931,
      "mean_token_accuracy": 0.19626167863607408,
      "num_tokens": 30749722.0,
      "step": 17720
    },
    {
      "entropy": 5.718609285354614,
      "epoch": 1.379070219801595,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048117115795107335,
      "loss": 5.292,
      "mean_token_accuracy": 0.18305347114801407,
      "num_tokens": 30759261.0,
      "step": 17725
    },
    {
      "entropy": 5.693091630935669,
      "epoch": 1.3794592491733126,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004811600575305987,
      "loss": 5.1203,
      "mean_token_accuracy": 0.1956807032227516,
      "num_tokens": 30769133.0,
      "step": 17730
    },
    {
      "entropy": 5.615317535400391,
      "epoch": 1.3798482785450301,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004811489539819731,
      "loss": 5.0459,
      "mean_token_accuracy": 0.19954855144023895,
      "num_tokens": 30777879.0,
      "step": 17735
    },
    {
      "entropy": 5.6220598220825195,
      "epoch": 1.3802373079167478,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048113784730536544,
      "loss": 5.1674,
      "mean_token_accuracy": 0.19754636138677598,
      "num_tokens": 30786307.0,
      "step": 17740
    },
    {
      "entropy": 5.6582808017730715,
      "epoch": 1.3806263372884653,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048112673750094396,
      "loss": 5.1837,
      "mean_token_accuracy": 0.18878302574157715,
      "num_tokens": 30796153.0,
      "step": 17745
    },
    {
      "entropy": 5.608172369003296,
      "epoch": 1.3810153666601828,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004811156245688773,
      "loss": 4.9604,
      "mean_token_accuracy": 0.19866713136434555,
      "num_tokens": 30805441.0,
      "step": 17750
    },
    {
      "entropy": 5.56573076248169,
      "epoch": 1.3814043960319005,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004811045085093342,
      "loss": 5.1187,
      "mean_token_accuracy": 0.18319412022829057,
      "num_tokens": 30814898.0,
      "step": 17755
    },
    {
      "entropy": 5.681590938568116,
      "epoch": 1.381793425403618,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004810933893224831,
      "loss": 5.2924,
      "mean_token_accuracy": 0.18218099623918532,
      "num_tokens": 30823995.0,
      "step": 17760
    },
    {
      "entropy": 5.655806541442871,
      "epoch": 1.3821824547753354,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048108226700849287,
      "loss": 5.1243,
      "mean_token_accuracy": 0.19118371307849885,
      "num_tokens": 30832963.0,
      "step": 17765
    },
    {
      "entropy": 5.640542078018188,
      "epoch": 1.382571484147053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004810711415675323,
      "loss": 5.1283,
      "mean_token_accuracy": 0.18644823580980302,
      "num_tokens": 30841559.0,
      "step": 17770
    },
    {
      "entropy": 5.6816521167755125,
      "epoch": 1.3829605135187706,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004810600129997702,
      "loss": 5.1765,
      "mean_token_accuracy": 0.18798786252737046,
      "num_tokens": 30850481.0,
      "step": 17775
    },
    {
      "entropy": 5.63182635307312,
      "epoch": 1.3833495428904883,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048104888130537534,
      "loss": 5.0154,
      "mean_token_accuracy": 0.19417383372783661,
      "num_tokens": 30859632.0,
      "step": 17780
    },
    {
      "entropy": 5.654698419570923,
      "epoch": 1.3837385722622058,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048103774648451664,
      "loss": 5.0895,
      "mean_token_accuracy": 0.19187875539064408,
      "num_tokens": 30867909.0,
      "step": 17785
    },
    {
      "entropy": 5.6158520698547365,
      "epoch": 1.3841276016339235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048102660853736314,
      "loss": 5.1031,
      "mean_token_accuracy": 0.19536942839622498,
      "num_tokens": 30876339.0,
      "step": 17790
    },
    {
      "entropy": 5.624637222290039,
      "epoch": 1.384516631005641,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004810154674640837,
      "loss": 5.0804,
      "mean_token_accuracy": 0.19113468378782272,
      "num_tokens": 30884998.0,
      "step": 17795
    },
    {
      "entropy": 5.649057626724243,
      "epoch": 1.3849056603773584,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048100432326484744,
      "loss": 5.1144,
      "mean_token_accuracy": 0.18475880324840546,
      "num_tokens": 30893554.0,
      "step": 17800
    },
    {
      "entropy": 5.5887950420379635,
      "epoch": 1.385294689749076,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004809931759398235,
      "loss": 5.1169,
      "mean_token_accuracy": 0.19248030930757523,
      "num_tokens": 30901954.0,
      "step": 17805
    },
    {
      "entropy": 5.607624244689942,
      "epoch": 1.3856837191207936,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048098202548918094,
      "loss": 5.0737,
      "mean_token_accuracy": 0.19664642661809922,
      "num_tokens": 30911145.0,
      "step": 17810
    },
    {
      "entropy": 5.5625138759613035,
      "epoch": 1.386072748492511,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048097087191308914,
      "loss": 4.9676,
      "mean_token_accuracy": 0.2015418902039528,
      "num_tokens": 30918924.0,
      "step": 17815
    },
    {
      "entropy": 5.591161823272705,
      "epoch": 1.3864617778642288,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004809597152117171,
      "loss": 5.1286,
      "mean_token_accuracy": 0.19008181989192963,
      "num_tokens": 30927620.0,
      "step": 17820
    },
    {
      "entropy": 5.593923854827881,
      "epoch": 1.3868508072359464,
      "grad_norm": 1.125,
      "learning_rate": 0.0004809485553852342,
      "loss": 5.0181,
      "mean_token_accuracy": 0.19180162400007247,
      "num_tokens": 30936140.0,
      "step": 17825
    },
    {
      "entropy": 5.700794744491577,
      "epoch": 1.387239836607664,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004809373924338098,
      "loss": 5.2234,
      "mean_token_accuracy": 0.18718539029359818,
      "num_tokens": 30944450.0,
      "step": 17830
    },
    {
      "entropy": 5.572172403335571,
      "epoch": 1.3876288659793814,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048092622635761316,
      "loss": 5.0533,
      "mean_token_accuracy": 0.1959550365805626,
      "num_tokens": 30953536.0,
      "step": 17835
    },
    {
      "entropy": 5.6090741634368895,
      "epoch": 1.388017895351099,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048091505715681395,
      "loss": 5.071,
      "mean_token_accuracy": 0.19303107261657715,
      "num_tokens": 30962556.0,
      "step": 17840
    },
    {
      "entropy": 5.589877653121948,
      "epoch": 1.3884069247228166,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004809038848315814,
      "loss": 5.0184,
      "mean_token_accuracy": 0.20263680517673494,
      "num_tokens": 30970476.0,
      "step": 17845
    },
    {
      "entropy": 5.664753007888794,
      "epoch": 1.388795954094534,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004808927093820851,
      "loss": 5.0633,
      "mean_token_accuracy": 0.18840179443359376,
      "num_tokens": 30978945.0,
      "step": 17850
    },
    {
      "entropy": 5.552944183349609,
      "epoch": 1.3891849834662517,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004808815308084947,
      "loss": 5.0265,
      "mean_token_accuracy": 0.19097656458616258,
      "num_tokens": 30988135.0,
      "step": 17855
    },
    {
      "entropy": 5.656019258499145,
      "epoch": 1.3895740128379692,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004808703491109798,
      "loss": 5.11,
      "mean_token_accuracy": 0.1987979531288147,
      "num_tokens": 30996408.0,
      "step": 17860
    },
    {
      "entropy": 5.594230604171753,
      "epoch": 1.3899630422096867,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004808591642897099,
      "loss": 5.1335,
      "mean_token_accuracy": 0.19308444410562514,
      "num_tokens": 31004156.0,
      "step": 17865
    },
    {
      "entropy": 5.575358629226685,
      "epoch": 1.3903520715814044,
      "grad_norm": 1.09375,
      "learning_rate": 0.000480847976344855,
      "loss": 5.0647,
      "mean_token_accuracy": 0.19762326776981354,
      "num_tokens": 31013518.0,
      "step": 17870
    },
    {
      "entropy": 5.555838012695313,
      "epoch": 1.390741100953122,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048083678527658465,
      "loss": 4.9586,
      "mean_token_accuracy": 0.20326709598302842,
      "num_tokens": 31022116.0,
      "step": 17875
    },
    {
      "entropy": 5.635913276672364,
      "epoch": 1.3911301303248396,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004808255910850687,
      "loss": 5.1227,
      "mean_token_accuracy": 0.19242368191480635,
      "num_tokens": 31030320.0,
      "step": 17880
    },
    {
      "entropy": 5.63309416770935,
      "epoch": 1.391519159696557,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004808143937704769,
      "loss": 5.0846,
      "mean_token_accuracy": 0.1992114245891571,
      "num_tokens": 31038410.0,
      "step": 17885
    },
    {
      "entropy": 5.582927322387695,
      "epoch": 1.3919081890682747,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048080319333297937,
      "loss": 5.0019,
      "mean_token_accuracy": 0.19571396857500076,
      "num_tokens": 31047719.0,
      "step": 17890
    },
    {
      "entropy": 5.588042974472046,
      "epoch": 1.3922972184399922,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048079198977274597,
      "loss": 5.138,
      "mean_token_accuracy": 0.19519376158714294,
      "num_tokens": 31057536.0,
      "step": 17895
    },
    {
      "entropy": 5.6246764183044435,
      "epoch": 1.3926862478117097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048078078308994666,
      "loss": 5.0877,
      "mean_token_accuracy": 0.19471651166677476,
      "num_tokens": 31066975.0,
      "step": 17900
    },
    {
      "entropy": 5.617932558059692,
      "epoch": 1.3930752771834274,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004807695732847515,
      "loss": 5.0789,
      "mean_token_accuracy": 0.20295270383358002,
      "num_tokens": 31075573.0,
      "step": 17905
    },
    {
      "entropy": 5.56302843093872,
      "epoch": 1.3934643065551449,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048075836035733053,
      "loss": 5.0057,
      "mean_token_accuracy": 0.20600516051054002,
      "num_tokens": 31084486.0,
      "step": 17910
    },
    {
      "entropy": 5.588654088973999,
      "epoch": 1.3938533359268626,
      "grad_norm": 1.109375,
      "learning_rate": 0.000480747144307854,
      "loss": 5.0568,
      "mean_token_accuracy": 0.1980026915669441,
      "num_tokens": 31093045.0,
      "step": 17915
    },
    {
      "entropy": 5.6169164180755615,
      "epoch": 1.39424236529858,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048073592513649203,
      "loss": 5.0863,
      "mean_token_accuracy": 0.20203913152217864,
      "num_tokens": 31102024.0,
      "step": 17920
    },
    {
      "entropy": 5.603633403778076,
      "epoch": 1.3946313946702977,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004807247028434148,
      "loss": 5.1737,
      "mean_token_accuracy": 0.18329787105321885,
      "num_tokens": 31110656.0,
      "step": 17925
    },
    {
      "entropy": 5.626287126541138,
      "epoch": 1.3950204240420152,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004807134774287927,
      "loss": 5.0325,
      "mean_token_accuracy": 0.20022621005773544,
      "num_tokens": 31119199.0,
      "step": 17930
    },
    {
      "entropy": 5.59880633354187,
      "epoch": 1.3954094534137327,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048070224889279603,
      "loss": 5.0341,
      "mean_token_accuracy": 0.20225781351327896,
      "num_tokens": 31127301.0,
      "step": 17935
    },
    {
      "entropy": 5.591383457183838,
      "epoch": 1.3957984827854504,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048069101723559505,
      "loss": 5.1173,
      "mean_token_accuracy": 0.19614229649305343,
      "num_tokens": 31136714.0,
      "step": 17940
    },
    {
      "entropy": 5.635069084167481,
      "epoch": 1.3961875121571679,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004806797824573603,
      "loss": 5.1176,
      "mean_token_accuracy": 0.18883515149354935,
      "num_tokens": 31145663.0,
      "step": 17945
    },
    {
      "entropy": 5.516148710250855,
      "epoch": 1.3965765415288853,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004806685445582624,
      "loss": 4.8749,
      "mean_token_accuracy": 0.20871867388486862,
      "num_tokens": 31154335.0,
      "step": 17950
    },
    {
      "entropy": 5.572521686553955,
      "epoch": 1.396965570900603,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048065730353847143,
      "loss": 5.0792,
      "mean_token_accuracy": 0.19308810383081437,
      "num_tokens": 31163683.0,
      "step": 17955
    },
    {
      "entropy": 5.6317041397094725,
      "epoch": 1.3973546002723205,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048064605939815837,
      "loss": 5.1616,
      "mean_token_accuracy": 0.19347446858882905,
      "num_tokens": 31172361.0,
      "step": 17960
    },
    {
      "entropy": 5.602168416976928,
      "epoch": 1.3977436296440382,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004806348121374936,
      "loss": 5.0034,
      "mean_token_accuracy": 0.20054301917552947,
      "num_tokens": 31180572.0,
      "step": 17965
    },
    {
      "entropy": 5.522147560119629,
      "epoch": 1.3981326590157557,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004806235617566479,
      "loss": 5.0134,
      "mean_token_accuracy": 0.19778368771076202,
      "num_tokens": 31188751.0,
      "step": 17970
    },
    {
      "entropy": 5.653079319000244,
      "epoch": 1.3985216883874734,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000480612308255792,
      "loss": 5.2122,
      "mean_token_accuracy": 0.18623577207326888,
      "num_tokens": 31197762.0,
      "step": 17975
    },
    {
      "entropy": 5.67055115699768,
      "epoch": 1.3989107177591908,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048060105163509647,
      "loss": 5.1405,
      "mean_token_accuracy": 0.19143532663583757,
      "num_tokens": 31207203.0,
      "step": 17980
    },
    {
      "entropy": 5.684044122695923,
      "epoch": 1.3992997471309083,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004805897918947323,
      "loss": 5.1061,
      "mean_token_accuracy": 0.1947082608938217,
      "num_tokens": 31215711.0,
      "step": 17985
    },
    {
      "entropy": 5.602418899536133,
      "epoch": 1.399688776502626,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004805785290348702,
      "loss": 5.0652,
      "mean_token_accuracy": 0.19409611374139785,
      "num_tokens": 31224564.0,
      "step": 17990
    },
    {
      "entropy": 5.5717569351196286,
      "epoch": 1.4000778058743435,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048056726305568124,
      "loss": 5.043,
      "mean_token_accuracy": 0.1928005561232567,
      "num_tokens": 31233252.0,
      "step": 17995
    },
    {
      "entropy": 5.622393369674683,
      "epoch": 1.400466835246061,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048055599395733617,
      "loss": 5.1404,
      "mean_token_accuracy": 0.1910122126340866,
      "num_tokens": 31242539.0,
      "step": 18000
    },
    {
      "epoch": 1.400466835246061,
      "eval_entropy": 5.429286362553469,
      "eval_loss": 5.153907775878906,
      "eval_mean_token_accuracy": 0.19979570700900845,
      "eval_num_tokens": 31242539.0,
      "eval_runtime": 21.6061,
      "eval_samples_per_second": 1923.435,
      "eval_steps_per_second": 240.441,
      "step": 18000
    },
    {
      "entropy": 5.65056209564209,
      "epoch": 1.4008558646177787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048054472174000615,
      "loss": 5.1001,
      "mean_token_accuracy": 0.1949731782078743,
      "num_tokens": 31250770.0,
      "step": 18005
    },
    {
      "entropy": 5.578313302993775,
      "epoch": 1.4012448939894961,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048053344640386207,
      "loss": 5.0429,
      "mean_token_accuracy": 0.20055692493915558,
      "num_tokens": 31259507.0,
      "step": 18010
    },
    {
      "entropy": 5.611670923233032,
      "epoch": 1.4016339233612138,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048052216794907505,
      "loss": 5.144,
      "mean_token_accuracy": 0.19261469095945358,
      "num_tokens": 31268106.0,
      "step": 18015
    },
    {
      "entropy": 5.507248783111573,
      "epoch": 1.4020229527329313,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048051088637581624,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19696250557899475,
      "num_tokens": 31276495.0,
      "step": 18020
    },
    {
      "entropy": 5.639213037490845,
      "epoch": 1.402411982104649,
      "grad_norm": 1.109375,
      "learning_rate": 0.000480499601684257,
      "loss": 5.1316,
      "mean_token_accuracy": 0.18737502992153168,
      "num_tokens": 31284962.0,
      "step": 18025
    },
    {
      "entropy": 5.6458573818206785,
      "epoch": 1.4028010114763665,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004804883138745682,
      "loss": 5.1704,
      "mean_token_accuracy": 0.18758513033390045,
      "num_tokens": 31294497.0,
      "step": 18030
    },
    {
      "entropy": 5.593687057495117,
      "epoch": 1.403190040848084,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004804770229469214,
      "loss": 5.0873,
      "mean_token_accuracy": 0.1968716114759445,
      "num_tokens": 31302607.0,
      "step": 18035
    },
    {
      "entropy": 5.661804866790772,
      "epoch": 1.4035790702198017,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004804657289014878,
      "loss": 5.1457,
      "mean_token_accuracy": 0.19137624949216842,
      "num_tokens": 31311989.0,
      "step": 18040
    },
    {
      "entropy": 5.676852560043335,
      "epoch": 1.4039680995915191,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048045443173843884,
      "loss": 5.0587,
      "mean_token_accuracy": 0.19819005131721495,
      "num_tokens": 31320681.0,
      "step": 18045
    },
    {
      "entropy": 5.6062075138092045,
      "epoch": 1.4043571289632366,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048044313145794587,
      "loss": 5.0871,
      "mean_token_accuracy": 0.19399428516626357,
      "num_tokens": 31328985.0,
      "step": 18050
    },
    {
      "entropy": 5.583251905441284,
      "epoch": 1.4047461583349543,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048043182806018034,
      "loss": 5.0359,
      "mean_token_accuracy": 0.19411151856184006,
      "num_tokens": 31337486.0,
      "step": 18055
    },
    {
      "entropy": 5.608684492111206,
      "epoch": 1.4051351877066718,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048042052154531384,
      "loss": 5.1453,
      "mean_token_accuracy": 0.1958534985780716,
      "num_tokens": 31345831.0,
      "step": 18060
    },
    {
      "entropy": 5.571682929992676,
      "epoch": 1.4055242170783895,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004804092119135179,
      "loss": 5.0368,
      "mean_token_accuracy": 0.20379883348941802,
      "num_tokens": 31354791.0,
      "step": 18065
    },
    {
      "entropy": 5.566554880142212,
      "epoch": 1.405913246450107,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004803978991649641,
      "loss": 4.9957,
      "mean_token_accuracy": 0.20051505863666536,
      "num_tokens": 31362652.0,
      "step": 18070
    },
    {
      "entropy": 5.559219646453857,
      "epoch": 1.4063022758218247,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048038658329982404,
      "loss": 5.0968,
      "mean_token_accuracy": 0.19080246835947037,
      "num_tokens": 31371439.0,
      "step": 18075
    },
    {
      "entropy": 5.701272916793823,
      "epoch": 1.4066913051935421,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004803752643182695,
      "loss": 5.106,
      "mean_token_accuracy": 0.1903628170490265,
      "num_tokens": 31380039.0,
      "step": 18080
    },
    {
      "entropy": 5.662473726272583,
      "epoch": 1.4070803345652596,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004803639422204723,
      "loss": 5.2068,
      "mean_token_accuracy": 0.18122496753931044,
      "num_tokens": 31389318.0,
      "step": 18085
    },
    {
      "entropy": 5.623923587799072,
      "epoch": 1.4074693639369773,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004803526170066041,
      "loss": 5.0892,
      "mean_token_accuracy": 0.19313371181488037,
      "num_tokens": 31398442.0,
      "step": 18090
    },
    {
      "entropy": 5.623306941986084,
      "epoch": 1.4078583933086948,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048034128867683674,
      "loss": 5.0897,
      "mean_token_accuracy": 0.19325269460678102,
      "num_tokens": 31406566.0,
      "step": 18095
    },
    {
      "entropy": 5.607749891281128,
      "epoch": 1.4082474226804123,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004803299572313422,
      "loss": 5.1523,
      "mean_token_accuracy": 0.18824519217014313,
      "num_tokens": 31415207.0,
      "step": 18100
    },
    {
      "entropy": 5.692129611968994,
      "epoch": 1.40863645205213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004803186226702924,
      "loss": 5.1372,
      "mean_token_accuracy": 0.18130026161670684,
      "num_tokens": 31424175.0,
      "step": 18105
    },
    {
      "entropy": 5.656840801239014,
      "epoch": 1.4090254814238474,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004803072849938592,
      "loss": 5.1649,
      "mean_token_accuracy": 0.18290869295597076,
      "num_tokens": 31434022.0,
      "step": 18110
    },
    {
      "entropy": 5.665083551406861,
      "epoch": 1.4094145107955651,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004802959442022149,
      "loss": 5.0415,
      "mean_token_accuracy": 0.19713061600923537,
      "num_tokens": 31442838.0,
      "step": 18115
    },
    {
      "entropy": 5.638256502151489,
      "epoch": 1.4098035401672826,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048028460029553126,
      "loss": 5.0545,
      "mean_token_accuracy": 0.1910913735628128,
      "num_tokens": 31451482.0,
      "step": 18120
    },
    {
      "entropy": 5.612159204483032,
      "epoch": 1.4101925695390003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048027325327398065,
      "loss": 5.0966,
      "mean_token_accuracy": 0.19379920065402984,
      "num_tokens": 31460234.0,
      "step": 18125
    },
    {
      "entropy": 5.603456258773804,
      "epoch": 1.4105815989107178,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004802619031377351,
      "loss": 5.046,
      "mean_token_accuracy": 0.20101485550403594,
      "num_tokens": 31470466.0,
      "step": 18130
    },
    {
      "entropy": 5.54808406829834,
      "epoch": 1.4109706282824352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048025054988696684,
      "loss": 5.1508,
      "mean_token_accuracy": 0.19007025063037872,
      "num_tokens": 31480901.0,
      "step": 18135
    },
    {
      "entropy": 5.616864156723023,
      "epoch": 1.411359657654153,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048023919352184827,
      "loss": 5.0591,
      "mean_token_accuracy": 0.18969772458076478,
      "num_tokens": 31490329.0,
      "step": 18140
    },
    {
      "entropy": 5.579936361312866,
      "epoch": 1.4117486870258704,
      "grad_norm": 1.0,
      "learning_rate": 0.00048022783404255156,
      "loss": 5.1002,
      "mean_token_accuracy": 0.1935671091079712,
      "num_tokens": 31499519.0,
      "step": 18145
    },
    {
      "entropy": 5.624794054031372,
      "epoch": 1.412137716397588,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004802164714492491,
      "loss": 5.1373,
      "mean_token_accuracy": 0.19408158510923385,
      "num_tokens": 31507899.0,
      "step": 18150
    },
    {
      "entropy": 5.700310182571411,
      "epoch": 1.4125267457693056,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048020510574211335,
      "loss": 5.2804,
      "mean_token_accuracy": 0.179886831343174,
      "num_tokens": 31517861.0,
      "step": 18155
    },
    {
      "entropy": 5.656043863296508,
      "epoch": 1.412915775141023,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048019373692131674,
      "loss": 5.1562,
      "mean_token_accuracy": 0.18432103246450424,
      "num_tokens": 31527273.0,
      "step": 18160
    },
    {
      "entropy": 5.660457563400269,
      "epoch": 1.4133048045127408,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048018236498703176,
      "loss": 5.0776,
      "mean_token_accuracy": 0.19688465148210527,
      "num_tokens": 31536221.0,
      "step": 18165
    },
    {
      "entropy": 5.589381694793701,
      "epoch": 1.4136938338844582,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048017098993943097,
      "loss": 5.0384,
      "mean_token_accuracy": 0.19620970338582994,
      "num_tokens": 31544329.0,
      "step": 18170
    },
    {
      "entropy": 5.6423393249511715,
      "epoch": 1.414082863256176,
      "grad_norm": 1.125,
      "learning_rate": 0.000480159611778687,
      "loss": 5.1152,
      "mean_token_accuracy": 0.19195820093154908,
      "num_tokens": 31552323.0,
      "step": 18175
    },
    {
      "entropy": 5.620124769210816,
      "epoch": 1.4144718926278934,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048014823050497243,
      "loss": 5.0736,
      "mean_token_accuracy": 0.19446938186883928,
      "num_tokens": 31560432.0,
      "step": 18180
    },
    {
      "entropy": 5.5212007522583,
      "epoch": 1.4148609219996109,
      "grad_norm": 1.125,
      "learning_rate": 0.00048013684611846,
      "loss": 5.0513,
      "mean_token_accuracy": 0.19666382521390915,
      "num_tokens": 31569294.0,
      "step": 18185
    },
    {
      "entropy": 5.59936637878418,
      "epoch": 1.4152499513713286,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048012545861932245,
      "loss": 5.0672,
      "mean_token_accuracy": 0.19645187854766846,
      "num_tokens": 31577860.0,
      "step": 18190
    },
    {
      "entropy": 5.582728719711303,
      "epoch": 1.415638980743046,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004801140680077325,
      "loss": 5.0185,
      "mean_token_accuracy": 0.19730323255062104,
      "num_tokens": 31586034.0,
      "step": 18195
    },
    {
      "entropy": 5.606848239898682,
      "epoch": 1.4160280101147635,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004801026742838631,
      "loss": 5.0575,
      "mean_token_accuracy": 0.19499358236789704,
      "num_tokens": 31594911.0,
      "step": 18200
    },
    {
      "entropy": 5.622370338439941,
      "epoch": 1.4164170394864812,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004800912774478871,
      "loss": 5.1377,
      "mean_token_accuracy": 0.18979110568761826,
      "num_tokens": 31603477.0,
      "step": 18205
    },
    {
      "entropy": 5.669853544235229,
      "epoch": 1.416806068858199,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004800798774999773,
      "loss": 5.1092,
      "mean_token_accuracy": 0.1946551099419594,
      "num_tokens": 31612288.0,
      "step": 18210
    },
    {
      "entropy": 5.617135381698608,
      "epoch": 1.4171950982299164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048006847444030686,
      "loss": 5.0374,
      "mean_token_accuracy": 0.20270806401968003,
      "num_tokens": 31621580.0,
      "step": 18215
    },
    {
      "entropy": 5.555125713348389,
      "epoch": 1.4175841276016339,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004800570682690487,
      "loss": 5.0428,
      "mean_token_accuracy": 0.18795739561319352,
      "num_tokens": 31629860.0,
      "step": 18220
    },
    {
      "entropy": 5.686068058013916,
      "epoch": 1.4179731569733516,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004800456589863759,
      "loss": 5.1157,
      "mean_token_accuracy": 0.1867443472146988,
      "num_tokens": 31639012.0,
      "step": 18225
    },
    {
      "entropy": 5.632424545288086,
      "epoch": 1.418362186345069,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004800342465924616,
      "loss": 5.1263,
      "mean_token_accuracy": 0.19317781180143356,
      "num_tokens": 31648675.0,
      "step": 18230
    },
    {
      "entropy": 5.636453104019165,
      "epoch": 1.4187512157167865,
      "grad_norm": 1.15625,
      "learning_rate": 0.000480022831087479,
      "loss": 5.0783,
      "mean_token_accuracy": 0.204926760494709,
      "num_tokens": 31656570.0,
      "step": 18235
    },
    {
      "entropy": 5.5677026271820065,
      "epoch": 1.4191402450885042,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004800114124716012,
      "loss": 5.021,
      "mean_token_accuracy": 0.1986503854393959,
      "num_tokens": 31665636.0,
      "step": 18240
    },
    {
      "entropy": 5.652220916748047,
      "epoch": 1.4195292744602217,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004799999907450015,
      "loss": 5.1445,
      "mean_token_accuracy": 0.18698307126760483,
      "num_tokens": 31673638.0,
      "step": 18245
    },
    {
      "entropy": 5.647220087051392,
      "epoch": 1.4199183038319392,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004799885659078533,
      "loss": 5.1086,
      "mean_token_accuracy": 0.19938829094171523,
      "num_tokens": 31681886.0,
      "step": 18250
    },
    {
      "entropy": 5.696392822265625,
      "epoch": 1.4203073332036569,
      "grad_norm": 1.125,
      "learning_rate": 0.0004799771379603299,
      "loss": 5.2431,
      "mean_token_accuracy": 0.18172856718301772,
      "num_tokens": 31690349.0,
      "step": 18255
    },
    {
      "entropy": 5.596795272827149,
      "epoch": 1.4206963625753746,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004799657069026046,
      "loss": 5.0932,
      "mean_token_accuracy": 0.18580224364995956,
      "num_tokens": 31699183.0,
      "step": 18260
    },
    {
      "entropy": 5.589864921569824,
      "epoch": 1.421085391947092,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047995427273485097,
      "loss": 5.0729,
      "mean_token_accuracy": 0.19426646828651428,
      "num_tokens": 31707759.0,
      "step": 18265
    },
    {
      "entropy": 5.68631796836853,
      "epoch": 1.4214744213188095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047994283545724247,
      "loss": 5.1959,
      "mean_token_accuracy": 0.1878586396574974,
      "num_tokens": 31716383.0,
      "step": 18270
    },
    {
      "entropy": 5.6335736274719235,
      "epoch": 1.4218634506905272,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047993139506995273,
      "loss": 5.1149,
      "mean_token_accuracy": 0.18795016705989837,
      "num_tokens": 31724617.0,
      "step": 18275
    },
    {
      "entropy": 5.6211004734039305,
      "epoch": 1.4222524800622447,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004799199515731551,
      "loss": 5.0447,
      "mean_token_accuracy": 0.19736108928918839,
      "num_tokens": 31732027.0,
      "step": 18280
    },
    {
      "entropy": 5.556650590896607,
      "epoch": 1.4226415094339622,
      "grad_norm": 1.125,
      "learning_rate": 0.00047990850496702346,
      "loss": 5.0165,
      "mean_token_accuracy": 0.19981116205453872,
      "num_tokens": 31740707.0,
      "step": 18285
    },
    {
      "entropy": 5.614840316772461,
      "epoch": 1.4230305388056799,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004798970552517314,
      "loss": 5.0466,
      "mean_token_accuracy": 0.19524014443159105,
      "num_tokens": 31748864.0,
      "step": 18290
    },
    {
      "entropy": 5.67936749458313,
      "epoch": 1.4234195681773973,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047988560242745264,
      "loss": 5.2248,
      "mean_token_accuracy": 0.18662955462932587,
      "num_tokens": 31758356.0,
      "step": 18295
    },
    {
      "entropy": 5.652460098266602,
      "epoch": 1.423808597549115,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000479874146494361,
      "loss": 5.0928,
      "mean_token_accuracy": 0.20081959068775176,
      "num_tokens": 31766911.0,
      "step": 18300
    },
    {
      "entropy": 5.628424596786499,
      "epoch": 1.4241976269208325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004798626874526302,
      "loss": 5.1094,
      "mean_token_accuracy": 0.1921548694372177,
      "num_tokens": 31775551.0,
      "step": 18305
    },
    {
      "entropy": 5.655199813842773,
      "epoch": 1.4245866562925502,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047985122530243426,
      "loss": 5.2528,
      "mean_token_accuracy": 0.18085827976465224,
      "num_tokens": 31784064.0,
      "step": 18310
    },
    {
      "entropy": 5.674431228637696,
      "epoch": 1.4249756856642677,
      "grad_norm": 1.078125,
      "learning_rate": 0.000479839760043947,
      "loss": 5.1177,
      "mean_token_accuracy": 0.19235825836658477,
      "num_tokens": 31792574.0,
      "step": 18315
    },
    {
      "entropy": 5.617513179779053,
      "epoch": 1.4253647150359852,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047982829167734245,
      "loss": 4.9516,
      "mean_token_accuracy": 0.20302326679229737,
      "num_tokens": 31800331.0,
      "step": 18320
    },
    {
      "entropy": 5.56688084602356,
      "epoch": 1.4257537444077029,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047981682020279456,
      "loss": 5.0276,
      "mean_token_accuracy": 0.19550492763519287,
      "num_tokens": 31808474.0,
      "step": 18325
    },
    {
      "entropy": 5.634739637374878,
      "epoch": 1.4261427737794203,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004798053456204775,
      "loss": 5.1295,
      "mean_token_accuracy": 0.18659272640943528,
      "num_tokens": 31817026.0,
      "step": 18330
    },
    {
      "entropy": 5.552191543579101,
      "epoch": 1.4265318031511378,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004797938679305651,
      "loss": 4.9773,
      "mean_token_accuracy": 0.1975885584950447,
      "num_tokens": 31825096.0,
      "step": 18335
    },
    {
      "entropy": 5.586865663528442,
      "epoch": 1.4269208325228555,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047978238713323193,
      "loss": 5.1363,
      "mean_token_accuracy": 0.19227559268474578,
      "num_tokens": 31833650.0,
      "step": 18340
    },
    {
      "entropy": 5.621965312957764,
      "epoch": 1.427309861894573,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047977090322865183,
      "loss": 5.0668,
      "mean_token_accuracy": 0.19526671022176742,
      "num_tokens": 31841803.0,
      "step": 18345
    },
    {
      "entropy": 5.662142753601074,
      "epoch": 1.4276988912662907,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004797594162169993,
      "loss": 5.1272,
      "mean_token_accuracy": 0.19729547053575516,
      "num_tokens": 31850121.0,
      "step": 18350
    },
    {
      "entropy": 5.619376516342163,
      "epoch": 1.4280879206380082,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004797479260984485,
      "loss": 5.0872,
      "mean_token_accuracy": 0.192670339345932,
      "num_tokens": 31858585.0,
      "step": 18355
    },
    {
      "entropy": 5.587238359451294,
      "epoch": 1.4284769500097259,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047973643287317386,
      "loss": 5.0725,
      "mean_token_accuracy": 0.19097839891910554,
      "num_tokens": 31866199.0,
      "step": 18360
    },
    {
      "entropy": 5.58135027885437,
      "epoch": 1.4288659793814433,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047972493654134963,
      "loss": 5.0787,
      "mean_token_accuracy": 0.1959703803062439,
      "num_tokens": 31875423.0,
      "step": 18365
    },
    {
      "entropy": 5.576603269577026,
      "epoch": 1.4292550087531608,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047971343710315043,
      "loss": 4.9822,
      "mean_token_accuracy": 0.19986646622419357,
      "num_tokens": 31883110.0,
      "step": 18370
    },
    {
      "entropy": 5.550352907180786,
      "epoch": 1.4296440381248785,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004797019345587506,
      "loss": 5.1004,
      "mean_token_accuracy": 0.19365949779748917,
      "num_tokens": 31891588.0,
      "step": 18375
    },
    {
      "entropy": 5.580299806594849,
      "epoch": 1.430033067496596,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004796904289083248,
      "loss": 5.0493,
      "mean_token_accuracy": 0.19603380858898162,
      "num_tokens": 31900671.0,
      "step": 18380
    },
    {
      "entropy": 5.645295524597168,
      "epoch": 1.4304220968683135,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004796789201520474,
      "loss": 5.1061,
      "mean_token_accuracy": 0.1948881193995476,
      "num_tokens": 31909188.0,
      "step": 18385
    },
    {
      "entropy": 5.615600442886352,
      "epoch": 1.4308111262400312,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047966740829009333,
      "loss": 5.0601,
      "mean_token_accuracy": 0.20297318696975708,
      "num_tokens": 31917607.0,
      "step": 18390
    },
    {
      "entropy": 5.626970720291138,
      "epoch": 1.4312001556117486,
      "grad_norm": 1.09375,
      "learning_rate": 0.000479655893322637,
      "loss": 5.1179,
      "mean_token_accuracy": 0.18798398226499557,
      "num_tokens": 31926510.0,
      "step": 18395
    },
    {
      "entropy": 5.62557282447815,
      "epoch": 1.4315891849834663,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004796443752498532,
      "loss": 5.1973,
      "mean_token_accuracy": 0.18588899672031403,
      "num_tokens": 31935550.0,
      "step": 18400
    },
    {
      "entropy": 5.697294759750366,
      "epoch": 1.4319782143551838,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004796328540719169,
      "loss": 5.245,
      "mean_token_accuracy": 0.1828196555376053,
      "num_tokens": 31944499.0,
      "step": 18405
    },
    {
      "entropy": 5.682896375656128,
      "epoch": 1.4323672437269015,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004796213297890026,
      "loss": 4.9934,
      "mean_token_accuracy": 0.20257802605628966,
      "num_tokens": 31952622.0,
      "step": 18410
    },
    {
      "entropy": 5.576907920837402,
      "epoch": 1.432756273098619,
      "grad_norm": 1.125,
      "learning_rate": 0.0004796098024012853,
      "loss": 5.0463,
      "mean_token_accuracy": 0.19152364432811736,
      "num_tokens": 31961606.0,
      "step": 18415
    },
    {
      "entropy": 5.591465282440185,
      "epoch": 1.4331453024703364,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047959827190894,
      "loss": 5.1588,
      "mean_token_accuracy": 0.1911668822169304,
      "num_tokens": 31970312.0,
      "step": 18420
    },
    {
      "entropy": 5.638664960861206,
      "epoch": 1.4335343318420541,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047958673831214157,
      "loss": 5.1088,
      "mean_token_accuracy": 0.1924675703048706,
      "num_tokens": 31978718.0,
      "step": 18425
    },
    {
      "entropy": 5.6605628490447994,
      "epoch": 1.4339233612137716,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047957520161106496,
      "loss": 5.129,
      "mean_token_accuracy": 0.19112288504838942,
      "num_tokens": 31987451.0,
      "step": 18430
    },
    {
      "entropy": 5.630765008926391,
      "epoch": 1.434312390585489,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004795636618058853,
      "loss": 4.9444,
      "mean_token_accuracy": 0.20537668615579605,
      "num_tokens": 31994977.0,
      "step": 18435
    },
    {
      "entropy": 5.562106609344482,
      "epoch": 1.4347014199572068,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004795521188967777,
      "loss": 5.0107,
      "mean_token_accuracy": 0.205131596326828,
      "num_tokens": 32003804.0,
      "step": 18440
    },
    {
      "entropy": 5.65944094657898,
      "epoch": 1.4350904493289243,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004795405728839172,
      "loss": 5.157,
      "mean_token_accuracy": 0.186469466984272,
      "num_tokens": 32012090.0,
      "step": 18445
    },
    {
      "entropy": 5.625867509841919,
      "epoch": 1.435479478700642,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004795290237674792,
      "loss": 5.1,
      "mean_token_accuracy": 0.19061363786458968,
      "num_tokens": 32019664.0,
      "step": 18450
    },
    {
      "entropy": 5.639222145080566,
      "epoch": 1.4358685080723594,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047951747154763866,
      "loss": 5.1472,
      "mean_token_accuracy": 0.1914323627948761,
      "num_tokens": 32028457.0,
      "step": 18455
    },
    {
      "entropy": 5.5980840682983395,
      "epoch": 1.4362575374440771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047950591622457106,
      "loss": 4.9769,
      "mean_token_accuracy": 0.19838547110557556,
      "num_tokens": 32036730.0,
      "step": 18460
    },
    {
      "entropy": 5.667256832122803,
      "epoch": 1.4366465668157946,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047949435779845174,
      "loss": 5.157,
      "mean_token_accuracy": 0.19142668843269348,
      "num_tokens": 32045657.0,
      "step": 18465
    },
    {
      "entropy": 5.598685789108276,
      "epoch": 1.437035596187512,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047948279626945596,
      "loss": 5.0746,
      "mean_token_accuracy": 0.19695494174957276,
      "num_tokens": 32053973.0,
      "step": 18470
    },
    {
      "entropy": 5.700230741500855,
      "epoch": 1.4374246255592298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047947123163775924,
      "loss": 5.1867,
      "mean_token_accuracy": 0.18649926781654358,
      "num_tokens": 32062833.0,
      "step": 18475
    },
    {
      "entropy": 5.58428692817688,
      "epoch": 1.4378136549309473,
      "grad_norm": 1.0625,
      "learning_rate": 0.000479459663903537,
      "loss": 5.0426,
      "mean_token_accuracy": 0.20397589355707169,
      "num_tokens": 32071578.0,
      "step": 18480
    },
    {
      "entropy": 5.573904037475586,
      "epoch": 1.4382026843026647,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004794480930669649,
      "loss": 4.9845,
      "mean_token_accuracy": 0.19611933082342148,
      "num_tokens": 32079847.0,
      "step": 18485
    },
    {
      "entropy": 5.596569156646728,
      "epoch": 1.4385917136743824,
      "grad_norm": 1.125,
      "learning_rate": 0.0004794365191282183,
      "loss": 5.0594,
      "mean_token_accuracy": 0.1885624721646309,
      "num_tokens": 32089751.0,
      "step": 18490
    },
    {
      "entropy": 5.676380157470703,
      "epoch": 1.4389807430461,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047942494208747297,
      "loss": 5.2091,
      "mean_token_accuracy": 0.18499039858579636,
      "num_tokens": 32097791.0,
      "step": 18495
    },
    {
      "entropy": 5.6378261089324955,
      "epoch": 1.4393697724178176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004794133619449045,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20201625674962997,
      "num_tokens": 32106417.0,
      "step": 18500
    },
    {
      "entropy": 5.586153173446656,
      "epoch": 1.439758801789535,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004794017787006886,
      "loss": 5.1074,
      "mean_token_accuracy": 0.18926345109939574,
      "num_tokens": 32115035.0,
      "step": 18505
    },
    {
      "entropy": 5.579412221908569,
      "epoch": 1.4401478311612528,
      "grad_norm": 1.125,
      "learning_rate": 0.000479390192355001,
      "loss": 5.0476,
      "mean_token_accuracy": 0.19102127999067306,
      "num_tokens": 32123704.0,
      "step": 18510
    },
    {
      "entropy": 5.6527947902679445,
      "epoch": 1.4405368605329703,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004793786029080176,
      "loss": 5.1396,
      "mean_token_accuracy": 0.1928951308131218,
      "num_tokens": 32132672.0,
      "step": 18515
    },
    {
      "entropy": 5.603618097305298,
      "epoch": 1.4409258899046877,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004793670103599143,
      "loss": 5.0526,
      "mean_token_accuracy": 0.196324722468853,
      "num_tokens": 32142135.0,
      "step": 18520
    },
    {
      "entropy": 5.656747150421142,
      "epoch": 1.4413149192764054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047935541471086677,
      "loss": 5.1518,
      "mean_token_accuracy": 0.17748203575611116,
      "num_tokens": 32151261.0,
      "step": 18525
    },
    {
      "entropy": 5.587941455841064,
      "epoch": 1.441703948648123,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004793438159610511,
      "loss": 5.0467,
      "mean_token_accuracy": 0.1954227790236473,
      "num_tokens": 32159454.0,
      "step": 18530
    },
    {
      "entropy": 5.678868389129638,
      "epoch": 1.4420929780198404,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047933221411064334,
      "loss": 5.1374,
      "mean_token_accuracy": 0.18622193336486817,
      "num_tokens": 32168706.0,
      "step": 18535
    },
    {
      "entropy": 5.732566022872925,
      "epoch": 1.442482007391558,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004793206091598194,
      "loss": 5.1976,
      "mean_token_accuracy": 0.1847882091999054,
      "num_tokens": 32177892.0,
      "step": 18540
    },
    {
      "entropy": 5.631687307357788,
      "epoch": 1.4428710367632755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004793090011087554,
      "loss": 5.0835,
      "mean_token_accuracy": 0.19265912622213363,
      "num_tokens": 32186081.0,
      "step": 18545
    },
    {
      "entropy": 5.631122875213623,
      "epoch": 1.4432600661349932,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047929738995762755,
      "loss": 5.0564,
      "mean_token_accuracy": 0.1973811626434326,
      "num_tokens": 32193778.0,
      "step": 18550
    },
    {
      "entropy": 5.5404986381530765,
      "epoch": 1.4436490955067107,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000479285775706612,
      "loss": 5.0342,
      "mean_token_accuracy": 0.20107520520687103,
      "num_tokens": 32202773.0,
      "step": 18555
    },
    {
      "entropy": 5.617558765411377,
      "epoch": 1.4440381248784284,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047927415835588496,
      "loss": 5.0676,
      "mean_token_accuracy": 0.19942388385534288,
      "num_tokens": 32212105.0,
      "step": 18560
    },
    {
      "entropy": 5.570112943649292,
      "epoch": 1.444427154250146,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047926253790562265,
      "loss": 4.9492,
      "mean_token_accuracy": 0.20351517498493193,
      "num_tokens": 32220550.0,
      "step": 18565
    },
    {
      "entropy": 5.567720031738281,
      "epoch": 1.4448161836218634,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047925091435600147,
      "loss": 4.9395,
      "mean_token_accuracy": 0.19839542508125305,
      "num_tokens": 32228562.0,
      "step": 18570
    },
    {
      "entropy": 5.606336688995361,
      "epoch": 1.445205212993581,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047923928770719776,
      "loss": 5.1077,
      "mean_token_accuracy": 0.19797687083482743,
      "num_tokens": 32237172.0,
      "step": 18575
    },
    {
      "entropy": 5.594252729415894,
      "epoch": 1.4455942423652985,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047922765795938797,
      "loss": 5.0936,
      "mean_token_accuracy": 0.19104631394147872,
      "num_tokens": 32246211.0,
      "step": 18580
    },
    {
      "entropy": 5.534573268890381,
      "epoch": 1.445983271737016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004792160251127485,
      "loss": 5.0193,
      "mean_token_accuracy": 0.20064648985862732,
      "num_tokens": 32255366.0,
      "step": 18585
    },
    {
      "entropy": 5.708288764953613,
      "epoch": 1.4463723011087337,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047920438916745586,
      "loss": 5.1926,
      "mean_token_accuracy": 0.18635097593069078,
      "num_tokens": 32264342.0,
      "step": 18590
    },
    {
      "entropy": 5.649402761459351,
      "epoch": 1.4467613304804512,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004791927501236866,
      "loss": 5.1161,
      "mean_token_accuracy": 0.19162465631961823,
      "num_tokens": 32273184.0,
      "step": 18595
    },
    {
      "entropy": 5.640807628631592,
      "epoch": 1.4471503598521689,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047918110798161754,
      "loss": 5.1552,
      "mean_token_accuracy": 0.1887485519051552,
      "num_tokens": 32282303.0,
      "step": 18600
    },
    {
      "entropy": 5.624294090270996,
      "epoch": 1.4475393892238864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047916946274142503,
      "loss": 5.0258,
      "mean_token_accuracy": 0.20135403722524642,
      "num_tokens": 32291118.0,
      "step": 18605
    },
    {
      "entropy": 5.660922336578369,
      "epoch": 1.447928418595604,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047915781440328593,
      "loss": 5.0852,
      "mean_token_accuracy": 0.18727379739284516,
      "num_tokens": 32299746.0,
      "step": 18610
    },
    {
      "entropy": 5.605949115753174,
      "epoch": 1.4483174479673215,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004791461629673769,
      "loss": 5.0953,
      "mean_token_accuracy": 0.19682133048772812,
      "num_tokens": 32308515.0,
      "step": 18615
    },
    {
      "entropy": 5.65838851928711,
      "epoch": 1.448706477339039,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004791345084338748,
      "loss": 5.199,
      "mean_token_accuracy": 0.1900344505906105,
      "num_tokens": 32317607.0,
      "step": 18620
    },
    {
      "entropy": 5.669473266601562,
      "epoch": 1.4490955067107567,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004791228508029565,
      "loss": 5.1046,
      "mean_token_accuracy": 0.19572567492723464,
      "num_tokens": 32325822.0,
      "step": 18625
    },
    {
      "entropy": 5.672076082229614,
      "epoch": 1.4494845360824742,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047911119007479873,
      "loss": 5.1921,
      "mean_token_accuracy": 0.1872390791773796,
      "num_tokens": 32336297.0,
      "step": 18630
    },
    {
      "entropy": 5.614063358306884,
      "epoch": 1.4498735654541917,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047909952624957866,
      "loss": 5.0673,
      "mean_token_accuracy": 0.19189515411853791,
      "num_tokens": 32344306.0,
      "step": 18635
    },
    {
      "entropy": 5.6910542964935305,
      "epoch": 1.4502625948259094,
      "grad_norm": 1.0,
      "learning_rate": 0.000479087859327473,
      "loss": 5.182,
      "mean_token_accuracy": 0.1891653671860695,
      "num_tokens": 32353668.0,
      "step": 18640
    },
    {
      "entropy": 5.66265377998352,
      "epoch": 1.450651624197627,
      "grad_norm": 1.1875,
      "learning_rate": 0.000479076189308659,
      "loss": 5.0744,
      "mean_token_accuracy": 0.19263841062784196,
      "num_tokens": 32361563.0,
      "step": 18645
    },
    {
      "entropy": 5.682370519638061,
      "epoch": 1.4510406535693445,
      "grad_norm": 1.125,
      "learning_rate": 0.00047906451619331364,
      "loss": 5.1791,
      "mean_token_accuracy": 0.18816327899694443,
      "num_tokens": 32369612.0,
      "step": 18650
    },
    {
      "entropy": 5.604054594039917,
      "epoch": 1.451429682941062,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047905283998161404,
      "loss": 4.9846,
      "mean_token_accuracy": 0.2052964448928833,
      "num_tokens": 32379013.0,
      "step": 18655
    },
    {
      "entropy": 5.587451457977295,
      "epoch": 1.4518187123127797,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047904116067373743,
      "loss": 5.0885,
      "mean_token_accuracy": 0.19270721971988677,
      "num_tokens": 32387933.0,
      "step": 18660
    },
    {
      "entropy": 5.644082641601562,
      "epoch": 1.4522077416844972,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004790294782698609,
      "loss": 5.1323,
      "mean_token_accuracy": 0.19549304395914077,
      "num_tokens": 32395669.0,
      "step": 18665
    },
    {
      "entropy": 5.5467259883880615,
      "epoch": 1.4525967710562147,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004790177927701618,
      "loss": 4.9841,
      "mean_token_accuracy": 0.20273464620113374,
      "num_tokens": 32403676.0,
      "step": 18670
    },
    {
      "entropy": 5.637514305114746,
      "epoch": 1.4529858004279324,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004790061041748174,
      "loss": 5.064,
      "mean_token_accuracy": 0.19444344639778138,
      "num_tokens": 32412089.0,
      "step": 18675
    },
    {
      "entropy": 5.637416219711303,
      "epoch": 1.4533748297996498,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004789944124840051,
      "loss": 5.0345,
      "mean_token_accuracy": 0.20215693414211272,
      "num_tokens": 32420357.0,
      "step": 18680
    },
    {
      "entropy": 5.641674947738648,
      "epoch": 1.4537638591713673,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004789827176979022,
      "loss": 5.124,
      "mean_token_accuracy": 0.1940718859434128,
      "num_tokens": 32429118.0,
      "step": 18685
    },
    {
      "entropy": 5.666227865219116,
      "epoch": 1.454152888543085,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004789710198166864,
      "loss": 5.0413,
      "mean_token_accuracy": 0.19791026115417482,
      "num_tokens": 32437266.0,
      "step": 18690
    },
    {
      "entropy": 5.650438737869263,
      "epoch": 1.4545419179148027,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047895931884053497,
      "loss": 5.1451,
      "mean_token_accuracy": 0.18620180040597917,
      "num_tokens": 32446014.0,
      "step": 18695
    },
    {
      "entropy": 5.608058643341065,
      "epoch": 1.4549309472865202,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047894761476962547,
      "loss": 5.1222,
      "mean_token_accuracy": 0.19133968651294708,
      "num_tokens": 32454591.0,
      "step": 18700
    },
    {
      "entropy": 5.6327752590179445,
      "epoch": 1.4553199766582376,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047893590760413545,
      "loss": 5.0818,
      "mean_token_accuracy": 0.19855124056339263,
      "num_tokens": 32462748.0,
      "step": 18705
    },
    {
      "entropy": 5.619998121261597,
      "epoch": 1.4557090060299553,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047892419734424276,
      "loss": 5.0514,
      "mean_token_accuracy": 0.19774118363857268,
      "num_tokens": 32471792.0,
      "step": 18710
    },
    {
      "entropy": 5.629866409301758,
      "epoch": 1.4560980354016728,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047891248399012495,
      "loss": 5.1246,
      "mean_token_accuracy": 0.19463704377412797,
      "num_tokens": 32480804.0,
      "step": 18715
    },
    {
      "entropy": 5.662797212600708,
      "epoch": 1.4564870647733903,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004789007675419597,
      "loss": 5.0797,
      "mean_token_accuracy": 0.190041483938694,
      "num_tokens": 32489591.0,
      "step": 18720
    },
    {
      "entropy": 5.6002167701721195,
      "epoch": 1.456876094145108,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047888904799992486,
      "loss": 5.0887,
      "mean_token_accuracy": 0.19588097780942917,
      "num_tokens": 32498164.0,
      "step": 18725
    },
    {
      "entropy": 5.572938251495361,
      "epoch": 1.4572651235168255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047887732536419826,
      "loss": 4.9163,
      "mean_token_accuracy": 0.20508276373147966,
      "num_tokens": 32505909.0,
      "step": 18730
    },
    {
      "entropy": 5.56914005279541,
      "epoch": 1.4576541528885432,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004788655996349577,
      "loss": 5.0507,
      "mean_token_accuracy": 0.19836744666099548,
      "num_tokens": 32513990.0,
      "step": 18735
    },
    {
      "entropy": 5.54734354019165,
      "epoch": 1.4580431822602606,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047885387081238125,
      "loss": 5.0935,
      "mean_token_accuracy": 0.1966705337166786,
      "num_tokens": 32522932.0,
      "step": 18740
    },
    {
      "entropy": 5.586341333389282,
      "epoch": 1.4584322116319783,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004788421388966467,
      "loss": 5.1631,
      "mean_token_accuracy": 0.18542364984750748,
      "num_tokens": 32530953.0,
      "step": 18745
    },
    {
      "entropy": 5.598444080352783,
      "epoch": 1.4588212410036958,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047883040388793207,
      "loss": 5.0249,
      "mean_token_accuracy": 0.1909759059548378,
      "num_tokens": 32539812.0,
      "step": 18750
    },
    {
      "entropy": 5.590904426574707,
      "epoch": 1.4592102703754133,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047881866578641563,
      "loss": 5.004,
      "mean_token_accuracy": 0.19597628116607665,
      "num_tokens": 32548425.0,
      "step": 18755
    },
    {
      "entropy": 5.572804641723633,
      "epoch": 1.459599299747131,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004788069245922753,
      "loss": 5.0602,
      "mean_token_accuracy": 0.196938718855381,
      "num_tokens": 32557619.0,
      "step": 18760
    },
    {
      "entropy": 5.631060171127319,
      "epoch": 1.4599883291188485,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004787951803056893,
      "loss": 5.0974,
      "mean_token_accuracy": 0.1899582639336586,
      "num_tokens": 32565602.0,
      "step": 18765
    },
    {
      "entropy": 5.640539216995239,
      "epoch": 1.460377358490566,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004787834329268358,
      "loss": 5.0929,
      "mean_token_accuracy": 0.190221931040287,
      "num_tokens": 32573691.0,
      "step": 18770
    },
    {
      "entropy": 5.6178418636322025,
      "epoch": 1.4607663878622836,
      "grad_norm": 1.03125,
      "learning_rate": 0.000478771682455893,
      "loss": 5.1855,
      "mean_token_accuracy": 0.18590253591537476,
      "num_tokens": 32582642.0,
      "step": 18775
    },
    {
      "entropy": 5.6502768993377686,
      "epoch": 1.461155417234001,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004787599288930393,
      "loss": 5.0887,
      "mean_token_accuracy": 0.193520687520504,
      "num_tokens": 32591328.0,
      "step": 18780
    },
    {
      "entropy": 5.607243824005127,
      "epoch": 1.4615444466057188,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000478748172238453,
      "loss": 5.0334,
      "mean_token_accuracy": 0.1931418925523758,
      "num_tokens": 32599286.0,
      "step": 18785
    },
    {
      "entropy": 5.660870933532715,
      "epoch": 1.4619334759774363,
      "grad_norm": 1.125,
      "learning_rate": 0.0004787364124923125,
      "loss": 5.2085,
      "mean_token_accuracy": 0.18378716260194777,
      "num_tokens": 32608156.0,
      "step": 18790
    },
    {
      "entropy": 5.627944231033325,
      "epoch": 1.462322505349154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047872464965479625,
      "loss": 5.0326,
      "mean_token_accuracy": 0.1981569692492485,
      "num_tokens": 32616134.0,
      "step": 18795
    },
    {
      "entropy": 5.623494815826416,
      "epoch": 1.4627115347208715,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004787128837260826,
      "loss": 5.0649,
      "mean_token_accuracy": 0.19462388902902603,
      "num_tokens": 32624711.0,
      "step": 18800
    },
    {
      "entropy": 5.609109354019165,
      "epoch": 1.463100564092589,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004787011147063503,
      "loss": 5.0845,
      "mean_token_accuracy": 0.1947449803352356,
      "num_tokens": 32633049.0,
      "step": 18805
    },
    {
      "entropy": 5.604067325592041,
      "epoch": 1.4634895934643066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004786893425957777,
      "loss": 5.0366,
      "mean_token_accuracy": 0.19538322240114211,
      "num_tokens": 32641173.0,
      "step": 18810
    },
    {
      "entropy": 5.611535978317261,
      "epoch": 1.463878622836024,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004786775673945436,
      "loss": 5.0265,
      "mean_token_accuracy": 0.19927253425121308,
      "num_tokens": 32650275.0,
      "step": 18815
    },
    {
      "entropy": 5.623269271850586,
      "epoch": 1.4642676522077416,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047866578910282656,
      "loss": 5.0394,
      "mean_token_accuracy": 0.20161254853010177,
      "num_tokens": 32658675.0,
      "step": 18820
    },
    {
      "entropy": 5.5783905506134035,
      "epoch": 1.4646566815794593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047865400772080535,
      "loss": 5.0478,
      "mean_token_accuracy": 0.1950514480471611,
      "num_tokens": 32666643.0,
      "step": 18825
    },
    {
      "entropy": 5.569488525390625,
      "epoch": 1.4650457109511767,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047864222324865875,
      "loss": 4.964,
      "mean_token_accuracy": 0.1971247375011444,
      "num_tokens": 32674791.0,
      "step": 18830
    },
    {
      "entropy": 5.621426916122436,
      "epoch": 1.4654347403228944,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004786304356865655,
      "loss": 5.0667,
      "mean_token_accuracy": 0.20350729823112487,
      "num_tokens": 32682790.0,
      "step": 18835
    },
    {
      "entropy": 5.63403549194336,
      "epoch": 1.465823769694612,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047861864503470457,
      "loss": 5.1752,
      "mean_token_accuracy": 0.18855532109737397,
      "num_tokens": 32692701.0,
      "step": 18840
    },
    {
      "entropy": 5.666257238388061,
      "epoch": 1.4662127990663296,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004786068512932547,
      "loss": 5.1192,
      "mean_token_accuracy": 0.1941573441028595,
      "num_tokens": 32701297.0,
      "step": 18845
    },
    {
      "entropy": 5.622356748580932,
      "epoch": 1.466601828438047,
      "grad_norm": 1.046875,
      "learning_rate": 0.000478595054462395,
      "loss": 5.0296,
      "mean_token_accuracy": 0.19540116786956788,
      "num_tokens": 32710188.0,
      "step": 18850
    },
    {
      "entropy": 5.568635272979736,
      "epoch": 1.4669908578097646,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047858325454230437,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19875933676958085,
      "num_tokens": 32719255.0,
      "step": 18855
    },
    {
      "entropy": 5.635839128494263,
      "epoch": 1.4673798871814823,
      "grad_norm": 1.125,
      "learning_rate": 0.0004785714515331619,
      "loss": 5.1147,
      "mean_token_accuracy": 0.19851367622613908,
      "num_tokens": 32728133.0,
      "step": 18860
    },
    {
      "entropy": 5.679207706451416,
      "epoch": 1.4677689165531997,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047855964543514666,
      "loss": 5.1398,
      "mean_token_accuracy": 0.19351682662963868,
      "num_tokens": 32736333.0,
      "step": 18865
    },
    {
      "entropy": 5.54669885635376,
      "epoch": 1.4681579459249172,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047854783624843786,
      "loss": 5.0482,
      "mean_token_accuracy": 0.20009141862392427,
      "num_tokens": 32744954.0,
      "step": 18870
    },
    {
      "entropy": 5.565046787261963,
      "epoch": 1.468546975296635,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047853602397321453,
      "loss": 5.1133,
      "mean_token_accuracy": 0.18978338241577147,
      "num_tokens": 32753720.0,
      "step": 18875
    },
    {
      "entropy": 5.6289421081542965,
      "epoch": 1.4689360046683524,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047852420860965605,
      "loss": 5.0025,
      "mean_token_accuracy": 0.20046208798885345,
      "num_tokens": 32762903.0,
      "step": 18880
    },
    {
      "entropy": 5.704815912246704,
      "epoch": 1.46932503404007,
      "grad_norm": 1.125,
      "learning_rate": 0.00047851239015794166,
      "loss": 5.1908,
      "mean_token_accuracy": 0.19168568253517151,
      "num_tokens": 32771367.0,
      "step": 18885
    },
    {
      "entropy": 5.6372082233428955,
      "epoch": 1.4697140634117876,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047850056861825066,
      "loss": 5.1001,
      "mean_token_accuracy": 0.19459104984998704,
      "num_tokens": 32779782.0,
      "step": 18890
    },
    {
      "entropy": 5.512723922729492,
      "epoch": 1.4701030927835053,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047848874399076245,
      "loss": 5.0027,
      "mean_token_accuracy": 0.19342736452817916,
      "num_tokens": 32788157.0,
      "step": 18895
    },
    {
      "entropy": 5.506467819213867,
      "epoch": 1.4704921221552227,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004784769162756563,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20290024280548097,
      "num_tokens": 32796590.0,
      "step": 18900
    },
    {
      "entropy": 5.705487871170044,
      "epoch": 1.4708811515269402,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000478465085473112,
      "loss": 5.1989,
      "mean_token_accuracy": 0.1869395539164543,
      "num_tokens": 32806297.0,
      "step": 18905
    },
    {
      "entropy": 5.66557354927063,
      "epoch": 1.471270180898658,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004784532515833088,
      "loss": 5.0504,
      "mean_token_accuracy": 0.2005374625325203,
      "num_tokens": 32815370.0,
      "step": 18910
    },
    {
      "entropy": 5.544957304000855,
      "epoch": 1.4716592102703754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047844141460642636,
      "loss": 5.0886,
      "mean_token_accuracy": 0.1905941218137741,
      "num_tokens": 32824678.0,
      "step": 18915
    },
    {
      "entropy": 5.671629762649536,
      "epoch": 1.4720482396420929,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047842957454264425,
      "loss": 5.1525,
      "mean_token_accuracy": 0.18905438631772994,
      "num_tokens": 32834759.0,
      "step": 18920
    },
    {
      "entropy": 5.7063267707824705,
      "epoch": 1.4724372690138106,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047841773139214213,
      "loss": 5.1098,
      "mean_token_accuracy": 0.19695671796798705,
      "num_tokens": 32843152.0,
      "step": 18925
    },
    {
      "entropy": 5.69730920791626,
      "epoch": 1.472826298385528,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004784058851550997,
      "loss": 5.2789,
      "mean_token_accuracy": 0.17807060182094575,
      "num_tokens": 32852702.0,
      "step": 18930
    },
    {
      "entropy": 5.602971839904785,
      "epoch": 1.4732153277572457,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004783940358316967,
      "loss": 4.9783,
      "mean_token_accuracy": 0.20461677014827728,
      "num_tokens": 32860201.0,
      "step": 18935
    },
    {
      "entropy": 5.533371257781982,
      "epoch": 1.4736043571289632,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047838218342211296,
      "loss": 4.9526,
      "mean_token_accuracy": 0.20545047670602798,
      "num_tokens": 32868731.0,
      "step": 18940
    },
    {
      "entropy": 5.633207559585571,
      "epoch": 1.473993386500681,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047837032792652837,
      "loss": 5.1665,
      "mean_token_accuracy": 0.19108043164014815,
      "num_tokens": 32876960.0,
      "step": 18945
    },
    {
      "entropy": 5.64943528175354,
      "epoch": 1.4743824158723984,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004783584693451226,
      "loss": 5.1162,
      "mean_token_accuracy": 0.19863019585609437,
      "num_tokens": 32886446.0,
      "step": 18950
    },
    {
      "entropy": 5.683539390563965,
      "epoch": 1.4747714452441159,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004783466076780759,
      "loss": 5.1883,
      "mean_token_accuracy": 0.18699711412191392,
      "num_tokens": 32895141.0,
      "step": 18955
    },
    {
      "entropy": 5.632814931869507,
      "epoch": 1.4751604746158336,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004783347429255679,
      "loss": 5.0937,
      "mean_token_accuracy": 0.18989978581666947,
      "num_tokens": 32904191.0,
      "step": 18960
    },
    {
      "entropy": 5.643506288528442,
      "epoch": 1.475549503987551,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000478322875087779,
      "loss": 5.0869,
      "mean_token_accuracy": 0.19535375386476517,
      "num_tokens": 32912805.0,
      "step": 18965
    },
    {
      "entropy": 5.596161842346191,
      "epoch": 1.4759385333592685,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047831100416488906,
      "loss": 5.0408,
      "mean_token_accuracy": 0.19759644120931624,
      "num_tokens": 32921275.0,
      "step": 18970
    },
    {
      "entropy": 5.654126787185669,
      "epoch": 1.4763275627309862,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047829913015707816,
      "loss": 5.1255,
      "mean_token_accuracy": 0.19490253031253815,
      "num_tokens": 32929472.0,
      "step": 18975
    },
    {
      "entropy": 5.602848482131958,
      "epoch": 1.4767165921027037,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047828725306452657,
      "loss": 5.1017,
      "mean_token_accuracy": 0.1915998175740242,
      "num_tokens": 32937697.0,
      "step": 18980
    },
    {
      "entropy": 5.6230028629302975,
      "epoch": 1.4771056214744214,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047827537288741453,
      "loss": 5.0871,
      "mean_token_accuracy": 0.19737082123756408,
      "num_tokens": 32946516.0,
      "step": 18985
    },
    {
      "entropy": 5.602562999725341,
      "epoch": 1.4774946508461388,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004782634896259222,
      "loss": 4.9725,
      "mean_token_accuracy": 0.20284359455108641,
      "num_tokens": 32954841.0,
      "step": 18990
    },
    {
      "entropy": 5.624081468582153,
      "epoch": 1.4778836802178565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047825160328023,
      "loss": 5.1223,
      "mean_token_accuracy": 0.19007060378789903,
      "num_tokens": 32963678.0,
      "step": 18995
    },
    {
      "entropy": 5.547390031814575,
      "epoch": 1.478272709589574,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004782397138505181,
      "loss": 4.9276,
      "mean_token_accuracy": 0.20106931030750275,
      "num_tokens": 32973163.0,
      "step": 19000
    },
    {
      "entropy": 5.6826231479644775,
      "epoch": 1.4786617389612915,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047822782133696715,
      "loss": 5.1306,
      "mean_token_accuracy": 0.19287146031856536,
      "num_tokens": 32982473.0,
      "step": 19005
    },
    {
      "entropy": 5.630597114562988,
      "epoch": 1.4790507683330092,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004782159257397574,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21068858057260514,
      "num_tokens": 32990328.0,
      "step": 19010
    },
    {
      "entropy": 5.523217296600341,
      "epoch": 1.4794397977047267,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047820402705906953,
      "loss": 5.0562,
      "mean_token_accuracy": 0.19401719570159912,
      "num_tokens": 32998432.0,
      "step": 19015
    },
    {
      "entropy": 5.607138776779175,
      "epoch": 1.4798288270764441,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047819212529508394,
      "loss": 5.1349,
      "mean_token_accuracy": 0.19625232517719268,
      "num_tokens": 33007017.0,
      "step": 19020
    },
    {
      "entropy": 5.678186655044556,
      "epoch": 1.4802178564481618,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004781802204479812,
      "loss": 5.0574,
      "mean_token_accuracy": 0.2002095714211464,
      "num_tokens": 33015212.0,
      "step": 19025
    },
    {
      "entropy": 5.641040325164795,
      "epoch": 1.4806068858198795,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004781683125179421,
      "loss": 5.0957,
      "mean_token_accuracy": 0.19393496215343475,
      "num_tokens": 33023555.0,
      "step": 19030
    },
    {
      "entropy": 5.5690813064575195,
      "epoch": 1.480995915191597,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004781564015051472,
      "loss": 5.0908,
      "mean_token_accuracy": 0.1918323427438736,
      "num_tokens": 33033048.0,
      "step": 19035
    },
    {
      "entropy": 5.520502042770386,
      "epoch": 1.4813849445633145,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004781444874097772,
      "loss": 4.9688,
      "mean_token_accuracy": 0.19712026715278624,
      "num_tokens": 33041553.0,
      "step": 19040
    },
    {
      "entropy": 5.557957410812378,
      "epoch": 1.4817739739350322,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047813257023201307,
      "loss": 5.077,
      "mean_token_accuracy": 0.1975543677806854,
      "num_tokens": 33050864.0,
      "step": 19045
    },
    {
      "entropy": 5.562325382232666,
      "epoch": 1.4821630033067497,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004781206499720354,
      "loss": 4.9698,
      "mean_token_accuracy": 0.20305258631706238,
      "num_tokens": 33059700.0,
      "step": 19050
    },
    {
      "entropy": 5.51577582359314,
      "epoch": 1.4825520326784671,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047810872663002523,
      "loss": 4.9413,
      "mean_token_accuracy": 0.2056463822722435,
      "num_tokens": 33067784.0,
      "step": 19055
    },
    {
      "entropy": 5.6162621021270756,
      "epoch": 1.4829410620501848,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047809680020616333,
      "loss": 5.0959,
      "mean_token_accuracy": 0.18692475110292434,
      "num_tokens": 33077174.0,
      "step": 19060
    },
    {
      "entropy": 5.665803718566894,
      "epoch": 1.4833300914219023,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047808487070063083,
      "loss": 5.1253,
      "mean_token_accuracy": 0.19567935317754745,
      "num_tokens": 33085492.0,
      "step": 19065
    },
    {
      "entropy": 5.622681045532227,
      "epoch": 1.4837191207936198,
      "grad_norm": 1.125,
      "learning_rate": 0.0004780729381136086,
      "loss": 5.0582,
      "mean_token_accuracy": 0.19979335218667985,
      "num_tokens": 33094388.0,
      "step": 19070
    },
    {
      "entropy": 5.511441659927368,
      "epoch": 1.4841081501653375,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004780610024452778,
      "loss": 4.907,
      "mean_token_accuracy": 0.20075707733631135,
      "num_tokens": 33101996.0,
      "step": 19075
    },
    {
      "entropy": 5.577497148513794,
      "epoch": 1.4844971795370552,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047804906369581954,
      "loss": 5.1478,
      "mean_token_accuracy": 0.19141252636909484,
      "num_tokens": 33111139.0,
      "step": 19080
    },
    {
      "entropy": 5.589953374862671,
      "epoch": 1.4848862089087727,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004780371218654148,
      "loss": 4.9715,
      "mean_token_accuracy": 0.198114213347435,
      "num_tokens": 33119159.0,
      "step": 19085
    },
    {
      "entropy": 5.6928643703460695,
      "epoch": 1.4852752382804901,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047802517695424496,
      "loss": 5.1371,
      "mean_token_accuracy": 0.1929064154624939,
      "num_tokens": 33128317.0,
      "step": 19090
    },
    {
      "entropy": 5.583741903305054,
      "epoch": 1.4856642676522078,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004780132289624913,
      "loss": 5.0555,
      "mean_token_accuracy": 0.20420397222042083,
      "num_tokens": 33136936.0,
      "step": 19095
    },
    {
      "entropy": 5.670713901519775,
      "epoch": 1.4860532970239253,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004780012778903349,
      "loss": 5.1239,
      "mean_token_accuracy": 0.19669472724199294,
      "num_tokens": 33145059.0,
      "step": 19100
    },
    {
      "entropy": 5.5884113788604735,
      "epoch": 1.4864423263956428,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047798932373795724,
      "loss": 5.0171,
      "mean_token_accuracy": 0.18887748420238495,
      "num_tokens": 33153683.0,
      "step": 19105
    },
    {
      "entropy": 5.699123859405518,
      "epoch": 1.4868313557673605,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047797736650553977,
      "loss": 5.129,
      "mean_token_accuracy": 0.1959105670452118,
      "num_tokens": 33163179.0,
      "step": 19110
    },
    {
      "entropy": 5.607194423675537,
      "epoch": 1.487220385139078,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047796540619326366,
      "loss": 4.954,
      "mean_token_accuracy": 0.20326871126890184,
      "num_tokens": 33172029.0,
      "step": 19115
    },
    {
      "entropy": 5.655848169326783,
      "epoch": 1.4876094145107956,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004779534428013107,
      "loss": 5.1421,
      "mean_token_accuracy": 0.19332901984453202,
      "num_tokens": 33181168.0,
      "step": 19120
    },
    {
      "entropy": 5.642968559265137,
      "epoch": 1.4879984438825131,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047794147632986223,
      "loss": 5.1437,
      "mean_token_accuracy": 0.18727092891931535,
      "num_tokens": 33191320.0,
      "step": 19125
    },
    {
      "entropy": 5.556647300720215,
      "epoch": 1.4883874732542308,
      "grad_norm": 1.125,
      "learning_rate": 0.0004779295067790999,
      "loss": 5.0604,
      "mean_token_accuracy": 0.1952892005443573,
      "num_tokens": 33199625.0,
      "step": 19130
    },
    {
      "entropy": 5.62584342956543,
      "epoch": 1.4887765026259483,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004779175341492053,
      "loss": 5.0401,
      "mean_token_accuracy": 0.19846860468387603,
      "num_tokens": 33208145.0,
      "step": 19135
    },
    {
      "entropy": 5.664146137237549,
      "epoch": 1.4891655319976658,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004779055584403601,
      "loss": 5.0557,
      "mean_token_accuracy": 0.19516253471374512,
      "num_tokens": 33216441.0,
      "step": 19140
    },
    {
      "entropy": 5.667415475845337,
      "epoch": 1.4895545613693835,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000477893579652746,
      "loss": 5.0574,
      "mean_token_accuracy": 0.18994946032762527,
      "num_tokens": 33225612.0,
      "step": 19145
    },
    {
      "entropy": 5.591235065460205,
      "epoch": 1.489943590741101,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047788159778654475,
      "loss": 5.0459,
      "mean_token_accuracy": 0.19451239705085754,
      "num_tokens": 33234410.0,
      "step": 19150
    },
    {
      "entropy": 5.633623933792114,
      "epoch": 1.4903326201128184,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047786961284193813,
      "loss": 4.9892,
      "mean_token_accuracy": 0.2010344535112381,
      "num_tokens": 33242838.0,
      "step": 19155
    },
    {
      "entropy": 5.616855478286743,
      "epoch": 1.4907216494845361,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004778576248191081,
      "loss": 5.1354,
      "mean_token_accuracy": 0.1909354954957962,
      "num_tokens": 33251266.0,
      "step": 19160
    },
    {
      "entropy": 5.6047464370727536,
      "epoch": 1.4911106788562536,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004778456337182365,
      "loss": 5.0411,
      "mean_token_accuracy": 0.19840567260980607,
      "num_tokens": 33259644.0,
      "step": 19165
    },
    {
      "entropy": 5.662776708602905,
      "epoch": 1.4914997082279713,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004778336395395052,
      "loss": 5.0414,
      "mean_token_accuracy": 0.1946261316537857,
      "num_tokens": 33267547.0,
      "step": 19170
    },
    {
      "entropy": 5.63383264541626,
      "epoch": 1.4918887375996888,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047782164228309636,
      "loss": 5.0746,
      "mean_token_accuracy": 0.19716629683971404,
      "num_tokens": 33276808.0,
      "step": 19175
    },
    {
      "entropy": 5.614906454086304,
      "epoch": 1.4922777669714065,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047780964194919193,
      "loss": 5.1141,
      "mean_token_accuracy": 0.19483509808778762,
      "num_tokens": 33285569.0,
      "step": 19180
    },
    {
      "entropy": 5.6178529262542725,
      "epoch": 1.492666796343124,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047779763853797387,
      "loss": 5.0976,
      "mean_token_accuracy": 0.1935784101486206,
      "num_tokens": 33295214.0,
      "step": 19185
    },
    {
      "entropy": 5.658003234863282,
      "epoch": 1.4930558257148414,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004777856320496245,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20267871022224426,
      "num_tokens": 33304140.0,
      "step": 19190
    },
    {
      "entropy": 5.6380421161651615,
      "epoch": 1.493444855086559,
      "grad_norm": 1.125,
      "learning_rate": 0.000477773622484326,
      "loss": 5.1314,
      "mean_token_accuracy": 0.19123824387788774,
      "num_tokens": 33312879.0,
      "step": 19195
    },
    {
      "entropy": 5.594028568267822,
      "epoch": 1.4938338844582766,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004777616098422604,
      "loss": 5.0691,
      "mean_token_accuracy": 0.19961473047733308,
      "num_tokens": 33320769.0,
      "step": 19200
    },
    {
      "entropy": 5.551181507110596,
      "epoch": 1.494222913829994,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047774959412361014,
      "loss": 4.9382,
      "mean_token_accuracy": 0.2064416453242302,
      "num_tokens": 33328911.0,
      "step": 19205
    },
    {
      "entropy": 5.606557941436767,
      "epoch": 1.4946119432017118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004777375753285575,
      "loss": 5.0251,
      "mean_token_accuracy": 0.20039540082216262,
      "num_tokens": 33337607.0,
      "step": 19210
    },
    {
      "entropy": 5.620619678497315,
      "epoch": 1.4950009725734292,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047772555345728486,
      "loss": 5.0865,
      "mean_token_accuracy": 0.19120642244815828,
      "num_tokens": 33346663.0,
      "step": 19215
    },
    {
      "entropy": 5.726786136627197,
      "epoch": 1.495390001945147,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004777135285099746,
      "loss": 5.137,
      "mean_token_accuracy": 0.18543317168951035,
      "num_tokens": 33355501.0,
      "step": 19220
    },
    {
      "entropy": 5.636517143249511,
      "epoch": 1.4957790313168644,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004777015004868092,
      "loss": 5.0061,
      "mean_token_accuracy": 0.19206279516220093,
      "num_tokens": 33363497.0,
      "step": 19225
    },
    {
      "entropy": 5.575192832946778,
      "epoch": 1.496168060688582,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004776894693879712,
      "loss": 4.9634,
      "mean_token_accuracy": 0.20535971820354462,
      "num_tokens": 33371349.0,
      "step": 19230
    },
    {
      "entropy": 5.583275985717774,
      "epoch": 1.4965570900602996,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004776774352136431,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2051143616437912,
      "num_tokens": 33379889.0,
      "step": 19235
    },
    {
      "entropy": 5.647780275344848,
      "epoch": 1.496946119432017,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047766539796400756,
      "loss": 5.091,
      "mean_token_accuracy": 0.19515041410923004,
      "num_tokens": 33388452.0,
      "step": 19240
    },
    {
      "entropy": 5.561920404434204,
      "epoch": 1.4973351488037348,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004776533576392471,
      "loss": 5.0111,
      "mean_token_accuracy": 0.19974779337644577,
      "num_tokens": 33397598.0,
      "step": 19245
    },
    {
      "entropy": 5.650741195678711,
      "epoch": 1.4977241781754522,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004776413142395445,
      "loss": 5.077,
      "mean_token_accuracy": 0.19172862768173218,
      "num_tokens": 33406509.0,
      "step": 19250
    },
    {
      "entropy": 5.622381639480591,
      "epoch": 1.4981132075471697,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047762926776508244,
      "loss": 5.0578,
      "mean_token_accuracy": 0.19003280997276306,
      "num_tokens": 33415129.0,
      "step": 19255
    },
    {
      "entropy": 5.52165002822876,
      "epoch": 1.4985022369188874,
      "grad_norm": 1.125,
      "learning_rate": 0.00047761721821604387,
      "loss": 4.9222,
      "mean_token_accuracy": 0.1996190145611763,
      "num_tokens": 33423588.0,
      "step": 19260
    },
    {
      "entropy": 5.638728618621826,
      "epoch": 1.4988912662906049,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004776051655926115,
      "loss": 5.1293,
      "mean_token_accuracy": 0.19162203520536422,
      "num_tokens": 33431598.0,
      "step": 19265
    },
    {
      "entropy": 5.71886305809021,
      "epoch": 1.4992802956623226,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047759310989496813,
      "loss": 5.1785,
      "mean_token_accuracy": 0.1957082450389862,
      "num_tokens": 33441139.0,
      "step": 19270
    },
    {
      "entropy": 5.673923015594482,
      "epoch": 1.49966932503404,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004775810511232969,
      "loss": 5.035,
      "mean_token_accuracy": 0.19134389609098434,
      "num_tokens": 33449558.0,
      "step": 19275
    },
    {
      "entropy": 5.641165494918823,
      "epoch": 1.5000583544057577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047756898927778056,
      "loss": 5.1119,
      "mean_token_accuracy": 0.1928480863571167,
      "num_tokens": 33457429.0,
      "step": 19280
    },
    {
      "entropy": 5.558756589889526,
      "epoch": 1.5004473837774752,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047755692435860227,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20363960266113282,
      "num_tokens": 33465781.0,
      "step": 19285
    },
    {
      "entropy": 5.66274676322937,
      "epoch": 1.5008364131491927,
      "grad_norm": 1.125,
      "learning_rate": 0.000477544856365945,
      "loss": 5.0351,
      "mean_token_accuracy": 0.19682733565568925,
      "num_tokens": 33473865.0,
      "step": 19290
    },
    {
      "entropy": 5.622910451889038,
      "epoch": 1.5012254425209104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047753278529999205,
      "loss": 5.0924,
      "mean_token_accuracy": 0.1950071409344673,
      "num_tokens": 33482497.0,
      "step": 19295
    },
    {
      "entropy": 5.607402276992798,
      "epoch": 1.5016144718926279,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047752071116092637,
      "loss": 5.0988,
      "mean_token_accuracy": 0.19350376576185227,
      "num_tokens": 33490933.0,
      "step": 19300
    },
    {
      "entropy": 5.6274772644042965,
      "epoch": 1.5020035012643453,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004775086339489312,
      "loss": 5.1078,
      "mean_token_accuracy": 0.19557205885648726,
      "num_tokens": 33500192.0,
      "step": 19305
    },
    {
      "entropy": 5.633109998703003,
      "epoch": 1.502392530636063,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004774965536641899,
      "loss": 5.0701,
      "mean_token_accuracy": 0.19493254274129868,
      "num_tokens": 33508543.0,
      "step": 19310
    },
    {
      "entropy": 5.5880790710449215,
      "epoch": 1.5027815600077807,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047748447030688575,
      "loss": 5.0978,
      "mean_token_accuracy": 0.19160389602184297,
      "num_tokens": 33517310.0,
      "step": 19315
    },
    {
      "entropy": 5.650627994537354,
      "epoch": 1.503170589379498,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047747238387720194,
      "loss": 5.1335,
      "mean_token_accuracy": 0.18930622488260268,
      "num_tokens": 33526274.0,
      "step": 19320
    },
    {
      "entropy": 5.7084485530853275,
      "epoch": 1.5035596187512157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000477460294375322,
      "loss": 5.1364,
      "mean_token_accuracy": 0.19404075145721436,
      "num_tokens": 33534438.0,
      "step": 19325
    },
    {
      "entropy": 5.611669588088989,
      "epoch": 1.5039486481229334,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047744820180142935,
      "loss": 5.0547,
      "mean_token_accuracy": 0.19025860577821732,
      "num_tokens": 33542943.0,
      "step": 19330
    },
    {
      "entropy": 5.599581575393676,
      "epoch": 1.5043376774946509,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047743610615570746,
      "loss": 5.0614,
      "mean_token_accuracy": 0.19358940422534943,
      "num_tokens": 33550567.0,
      "step": 19335
    },
    {
      "entropy": 5.6271261215209964,
      "epoch": 1.5047267068663683,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047742400743833993,
      "loss": 5.1604,
      "mean_token_accuracy": 0.19190303087234498,
      "num_tokens": 33559851.0,
      "step": 19340
    },
    {
      "entropy": 5.669795417785645,
      "epoch": 1.505115736238086,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004774119056495102,
      "loss": 5.0715,
      "mean_token_accuracy": 0.1915639564394951,
      "num_tokens": 33568645.0,
      "step": 19345
    },
    {
      "entropy": 5.628902387619019,
      "epoch": 1.5055047656098035,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047739980078940206,
      "loss": 5.0424,
      "mean_token_accuracy": 0.2036150723695755,
      "num_tokens": 33577331.0,
      "step": 19350
    },
    {
      "entropy": 5.540964078903198,
      "epoch": 1.505893794981521,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047738769285819894,
      "loss": 4.9036,
      "mean_token_accuracy": 0.20337747782468796,
      "num_tokens": 33586196.0,
      "step": 19355
    },
    {
      "entropy": 5.588247871398925,
      "epoch": 1.5062828243532387,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004773755818560849,
      "loss": 5.1326,
      "mean_token_accuracy": 0.19261325299739837,
      "num_tokens": 33595200.0,
      "step": 19360
    },
    {
      "entropy": 5.60738172531128,
      "epoch": 1.5066718537249564,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004773634677832434,
      "loss": 5.013,
      "mean_token_accuracy": 0.20040953606367112,
      "num_tokens": 33604130.0,
      "step": 19365
    },
    {
      "entropy": 5.648499774932861,
      "epoch": 1.5070608830966739,
      "grad_norm": 1.125,
      "learning_rate": 0.0004773513506398584,
      "loss": 5.1968,
      "mean_token_accuracy": 0.18189657330513,
      "num_tokens": 33613254.0,
      "step": 19370
    },
    {
      "entropy": 5.6190064430236815,
      "epoch": 1.5074499124683913,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004773392304261137,
      "loss": 5.0537,
      "mean_token_accuracy": 0.20282718390226365,
      "num_tokens": 33621395.0,
      "step": 19375
    },
    {
      "entropy": 5.616506671905517,
      "epoch": 1.507838941840109,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004773271071421933,
      "loss": 5.0036,
      "mean_token_accuracy": 0.196007376909256,
      "num_tokens": 33630293.0,
      "step": 19380
    },
    {
      "entropy": 5.66860065460205,
      "epoch": 1.5082279712118265,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047731498078828105,
      "loss": 5.1109,
      "mean_token_accuracy": 0.19406146705150604,
      "num_tokens": 33638224.0,
      "step": 19385
    },
    {
      "entropy": 5.68125729560852,
      "epoch": 1.508617000583544,
      "grad_norm": 1.09375,
      "learning_rate": 0.000477302851364561,
      "loss": 5.1554,
      "mean_token_accuracy": 0.18304163515567778,
      "num_tokens": 33646974.0,
      "step": 19390
    },
    {
      "entropy": 5.640812683105469,
      "epoch": 1.5090060299552617,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047729071887121717,
      "loss": 5.1323,
      "mean_token_accuracy": 0.18878572136163713,
      "num_tokens": 33656171.0,
      "step": 19395
    },
    {
      "entropy": 5.622091245651245,
      "epoch": 1.5093950593269791,
      "grad_norm": 1.125,
      "learning_rate": 0.0004772785833084337,
      "loss": 5.1228,
      "mean_token_accuracy": 0.18826854079961777,
      "num_tokens": 33664605.0,
      "step": 19400
    },
    {
      "entropy": 5.652630090713501,
      "epoch": 1.5097840886986966,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004772664446763946,
      "loss": 5.1083,
      "mean_token_accuracy": 0.19310170710086821,
      "num_tokens": 33673128.0,
      "step": 19405
    },
    {
      "entropy": 5.594411039352417,
      "epoch": 1.5101731180704143,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004772543029752842,
      "loss": 5.0499,
      "mean_token_accuracy": 0.19972462952136993,
      "num_tokens": 33681358.0,
      "step": 19410
    },
    {
      "entropy": 5.556503772735596,
      "epoch": 1.510562147442132,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047724215820528666,
      "loss": 4.939,
      "mean_token_accuracy": 0.2080944821238518,
      "num_tokens": 33690282.0,
      "step": 19415
    },
    {
      "entropy": 5.549698352813721,
      "epoch": 1.5109511768138495,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004772300103665863,
      "loss": 5.0076,
      "mean_token_accuracy": 0.1941213235259056,
      "num_tokens": 33698294.0,
      "step": 19420
    },
    {
      "entropy": 5.5501477241516115,
      "epoch": 1.511340206185567,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047721785945936733,
      "loss": 4.9793,
      "mean_token_accuracy": 0.20674926787614822,
      "num_tokens": 33706552.0,
      "step": 19425
    },
    {
      "entropy": 5.619852876663208,
      "epoch": 1.5117292355572847,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004772057054838143,
      "loss": 5.0335,
      "mean_token_accuracy": 0.19712848514318465,
      "num_tokens": 33714704.0,
      "step": 19430
    },
    {
      "entropy": 5.642540121078492,
      "epoch": 1.5121182649290021,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047719354844011146,
      "loss": 5.094,
      "mean_token_accuracy": 0.19931830912828447,
      "num_tokens": 33723823.0,
      "step": 19435
    },
    {
      "entropy": 5.718218851089477,
      "epoch": 1.5125072943007196,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004771813883284434,
      "loss": 5.1549,
      "mean_token_accuracy": 0.19305465519428253,
      "num_tokens": 33732616.0,
      "step": 19440
    },
    {
      "entropy": 5.577765274047851,
      "epoch": 1.5128963236724373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047716922514899455,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20044719725847243,
      "num_tokens": 33741542.0,
      "step": 19445
    },
    {
      "entropy": 5.593866348266602,
      "epoch": 1.5132853530441548,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047715705890194953,
      "loss": 5.0813,
      "mean_token_accuracy": 0.1991390824317932,
      "num_tokens": 33750651.0,
      "step": 19450
    },
    {
      "entropy": 5.537260627746582,
      "epoch": 1.5136743824158723,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047714488958749285,
      "loss": 5.1175,
      "mean_token_accuracy": 0.19317656010389328,
      "num_tokens": 33760741.0,
      "step": 19455
    },
    {
      "entropy": 5.701646995544434,
      "epoch": 1.51406341178759,
      "grad_norm": 1.125,
      "learning_rate": 0.00047713271720580924,
      "loss": 5.1173,
      "mean_token_accuracy": 0.19490543603897095,
      "num_tokens": 33769541.0,
      "step": 19460
    },
    {
      "entropy": 5.644590234756469,
      "epoch": 1.5144524411593077,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004771205417570834,
      "loss": 5.1174,
      "mean_token_accuracy": 0.19758770763874053,
      "num_tokens": 33778684.0,
      "step": 19465
    },
    {
      "entropy": 5.656781387329102,
      "epoch": 1.5148414705310251,
      "grad_norm": 1.125,
      "learning_rate": 0.00047710836324150004,
      "loss": 5.0996,
      "mean_token_accuracy": 0.1960084393620491,
      "num_tokens": 33787782.0,
      "step": 19470
    },
    {
      "entropy": 5.638566589355468,
      "epoch": 1.5152304999027426,
      "grad_norm": 1.046875,
      "learning_rate": 0.000477096181659244,
      "loss": 5.054,
      "mean_token_accuracy": 0.19813997447490692,
      "num_tokens": 33796194.0,
      "step": 19475
    },
    {
      "entropy": 5.652682542800903,
      "epoch": 1.5156195292744603,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004770839970105,
      "loss": 5.0524,
      "mean_token_accuracy": 0.1946710765361786,
      "num_tokens": 33804615.0,
      "step": 19480
    },
    {
      "entropy": 5.653340196609497,
      "epoch": 1.5160085586461778,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047707180929545295,
      "loss": 5.1486,
      "mean_token_accuracy": 0.1954011395573616,
      "num_tokens": 33812937.0,
      "step": 19485
    },
    {
      "entropy": 5.601870775222778,
      "epoch": 1.5163975880178953,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047705961851428786,
      "loss": 5.0152,
      "mean_token_accuracy": 0.19929749220609666,
      "num_tokens": 33821454.0,
      "step": 19490
    },
    {
      "entropy": 5.665795755386353,
      "epoch": 1.516786617389613,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047704742466718973,
      "loss": 5.1156,
      "mean_token_accuracy": 0.19531653821468353,
      "num_tokens": 33829736.0,
      "step": 19495
    },
    {
      "entropy": 5.660660028457642,
      "epoch": 1.5171756467613304,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047703522775434354,
      "loss": 5.1024,
      "mean_token_accuracy": 0.19056737571954727,
      "num_tokens": 33838403.0,
      "step": 19500
    },
    {
      "entropy": 5.609117317199707,
      "epoch": 1.517564676133048,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047702302777593425,
      "loss": 5.0817,
      "mean_token_accuracy": 0.19499066323041916,
      "num_tokens": 33847149.0,
      "step": 19505
    },
    {
      "entropy": 5.630152463912964,
      "epoch": 1.5179537055047656,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047701082473214715,
      "loss": 5.0539,
      "mean_token_accuracy": 0.19630080610513687,
      "num_tokens": 33855924.0,
      "step": 19510
    },
    {
      "entropy": 5.605315256118774,
      "epoch": 1.5183427348764833,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047699861862316725,
      "loss": 5.0619,
      "mean_token_accuracy": 0.1962066501379013,
      "num_tokens": 33864314.0,
      "step": 19515
    },
    {
      "entropy": 5.6441854476928714,
      "epoch": 1.5187317642482008,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004769864094491798,
      "loss": 5.0128,
      "mean_token_accuracy": 0.20023605078458787,
      "num_tokens": 33873468.0,
      "step": 19520
    },
    {
      "entropy": 5.620679759979248,
      "epoch": 1.5191207936199183,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004769741972103702,
      "loss": 5.0722,
      "mean_token_accuracy": 0.19524154663085938,
      "num_tokens": 33881627.0,
      "step": 19525
    },
    {
      "entropy": 5.5914106369018555,
      "epoch": 1.519509822991636,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047696198190692353,
      "loss": 5.0683,
      "mean_token_accuracy": 0.1924902081489563,
      "num_tokens": 33889839.0,
      "step": 19530
    },
    {
      "entropy": 5.603225946426392,
      "epoch": 1.5198988523633534,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004769497635390253,
      "loss": 5.0035,
      "mean_token_accuracy": 0.1945620968937874,
      "num_tokens": 33898370.0,
      "step": 19535
    },
    {
      "entropy": 5.651098203659058,
      "epoch": 1.520287881735071,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004769375421068608,
      "loss": 5.1051,
      "mean_token_accuracy": 0.19593175649642944,
      "num_tokens": 33907419.0,
      "step": 19540
    },
    {
      "entropy": 5.646708917617798,
      "epoch": 1.5206769111067886,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047692531761061555,
      "loss": 5.0836,
      "mean_token_accuracy": 0.190091510117054,
      "num_tokens": 33915957.0,
      "step": 19545
    },
    {
      "entropy": 5.650356245040894,
      "epoch": 1.521065940478506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000476913090050475,
      "loss": 5.0468,
      "mean_token_accuracy": 0.19721494615077972,
      "num_tokens": 33924012.0,
      "step": 19550
    },
    {
      "entropy": 5.590018177032471,
      "epoch": 1.5214549698502235,
      "grad_norm": 1.375,
      "learning_rate": 0.00047690085942662464,
      "loss": 5.1079,
      "mean_token_accuracy": 0.18801727145910263,
      "num_tokens": 33932652.0,
      "step": 19555
    },
    {
      "entropy": 5.611227369308471,
      "epoch": 1.5218439992219412,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047688862573925015,
      "loss": 5.0231,
      "mean_token_accuracy": 0.1991322085261345,
      "num_tokens": 33941208.0,
      "step": 19560
    },
    {
      "entropy": 5.608739757537842,
      "epoch": 1.522233028593659,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047687638898853707,
      "loss": 5.0247,
      "mean_token_accuracy": 0.19637274295091628,
      "num_tokens": 33950435.0,
      "step": 19565
    },
    {
      "entropy": 5.60824146270752,
      "epoch": 1.5226220579653764,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004768641491746711,
      "loss": 5.03,
      "mean_token_accuracy": 0.2013051465153694,
      "num_tokens": 33959659.0,
      "step": 19570
    },
    {
      "entropy": 5.624347448348999,
      "epoch": 1.523011087337094,
      "grad_norm": 1.015625,
      "learning_rate": 0.000476851906297838,
      "loss": 5.0471,
      "mean_token_accuracy": 0.191919207572937,
      "num_tokens": 33969105.0,
      "step": 19575
    },
    {
      "entropy": 5.621804475784302,
      "epoch": 1.5234001167088116,
      "grad_norm": 1.125,
      "learning_rate": 0.00047683966035822346,
      "loss": 5.1487,
      "mean_token_accuracy": 0.1893953189253807,
      "num_tokens": 33977314.0,
      "step": 19580
    },
    {
      "entropy": 5.603160047531128,
      "epoch": 1.523789146080529,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047682741135601336,
      "loss": 5.0751,
      "mean_token_accuracy": 0.1930656224489212,
      "num_tokens": 33985696.0,
      "step": 19585
    },
    {
      "entropy": 5.506903409957886,
      "epoch": 1.5241781754522465,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047681515929139356,
      "loss": 4.847,
      "mean_token_accuracy": 0.21282618194818498,
      "num_tokens": 33994539.0,
      "step": 19590
    },
    {
      "entropy": 5.578264808654785,
      "epoch": 1.5245672048239642,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047680290416454995,
      "loss": 5.1177,
      "mean_token_accuracy": 0.18260177969932556,
      "num_tokens": 34003101.0,
      "step": 19595
    },
    {
      "entropy": 5.65296368598938,
      "epoch": 1.524956234195682,
      "grad_norm": 1.125,
      "learning_rate": 0.0004767906459756684,
      "loss": 5.0614,
      "mean_token_accuracy": 0.19860363900661468,
      "num_tokens": 34011344.0,
      "step": 19600
    },
    {
      "entropy": 5.6451750755310055,
      "epoch": 1.5253452635673992,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047677838472493504,
      "loss": 5.0775,
      "mean_token_accuracy": 0.19790596812963485,
      "num_tokens": 34019628.0,
      "step": 19605
    },
    {
      "entropy": 5.634744930267334,
      "epoch": 1.5257342929391169,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004767661204125358,
      "loss": 5.0498,
      "mean_token_accuracy": 0.19294812083244323,
      "num_tokens": 34027709.0,
      "step": 19610
    },
    {
      "entropy": 5.5801502704620365,
      "epoch": 1.5261233223108346,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004767538530386569,
      "loss": 5.1021,
      "mean_token_accuracy": 0.19398279637098312,
      "num_tokens": 34036412.0,
      "step": 19615
    },
    {
      "entropy": 5.618213891983032,
      "epoch": 1.526512351682552,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047674158260348437,
      "loss": 5.0417,
      "mean_token_accuracy": 0.20025135278701783,
      "num_tokens": 34044842.0,
      "step": 19620
    },
    {
      "entropy": 5.548794889450074,
      "epoch": 1.5269013810542695,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047672930910720436,
      "loss": 5.0051,
      "mean_token_accuracy": 0.19667597711086274,
      "num_tokens": 34053286.0,
      "step": 19625
    },
    {
      "entropy": 5.565608263015747,
      "epoch": 1.5272904104259872,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047671703255000326,
      "loss": 5.0255,
      "mean_token_accuracy": 0.2006692573428154,
      "num_tokens": 34062059.0,
      "step": 19630
    },
    {
      "entropy": 5.634258222579956,
      "epoch": 1.5276794397977047,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004767047529320673,
      "loss": 5.1078,
      "mean_token_accuracy": 0.1846057280898094,
      "num_tokens": 34070800.0,
      "step": 19635
    },
    {
      "entropy": 5.644186735153198,
      "epoch": 1.5280684691694222,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047669247025358257,
      "loss": 5.1418,
      "mean_token_accuracy": 0.19050773978233337,
      "num_tokens": 34079935.0,
      "step": 19640
    },
    {
      "entropy": 5.595399713516235,
      "epoch": 1.5284574985411399,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047668018451473584,
      "loss": 5.0139,
      "mean_token_accuracy": 0.19541258960962296,
      "num_tokens": 34088614.0,
      "step": 19645
    },
    {
      "entropy": 5.651969385147095,
      "epoch": 1.5288465279128576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004766678957157132,
      "loss": 5.0385,
      "mean_token_accuracy": 0.20503697097301482,
      "num_tokens": 34096899.0,
      "step": 19650
    },
    {
      "entropy": 5.656499862670898,
      "epoch": 1.5292355572845748,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004766556038567013,
      "loss": 5.1438,
      "mean_token_accuracy": 0.18763657212257384,
      "num_tokens": 34105936.0,
      "step": 19655
    },
    {
      "entropy": 5.592771768569946,
      "epoch": 1.5296245866562925,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004766433089378865,
      "loss": 5.0109,
      "mean_token_accuracy": 0.19464890658855438,
      "num_tokens": 34113675.0,
      "step": 19660
    },
    {
      "entropy": 5.600823307037354,
      "epoch": 1.5300136160280102,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047663101095945544,
      "loss": 4.9224,
      "mean_token_accuracy": 0.20701434314250947,
      "num_tokens": 34121944.0,
      "step": 19665
    },
    {
      "entropy": 5.530815601348877,
      "epoch": 1.5304026453997277,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047661870992159476,
      "loss": 5.0292,
      "mean_token_accuracy": 0.20334934294223786,
      "num_tokens": 34130403.0,
      "step": 19670
    },
    {
      "entropy": 5.6039036273956295,
      "epoch": 1.5307916747714452,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047660640582449106,
      "loss": 5.1219,
      "mean_token_accuracy": 0.20233838856220246,
      "num_tokens": 34139853.0,
      "step": 19675
    },
    {
      "entropy": 5.705723476409912,
      "epoch": 1.5311807041431629,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047659409866833104,
      "loss": 5.1529,
      "mean_token_accuracy": 0.20370273292064667,
      "num_tokens": 34147833.0,
      "step": 19680
    },
    {
      "entropy": 5.65820894241333,
      "epoch": 1.5315697335148803,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004765817884533014,
      "loss": 5.0617,
      "mean_token_accuracy": 0.2025201916694641,
      "num_tokens": 34156248.0,
      "step": 19685
    },
    {
      "entropy": 5.482030534744263,
      "epoch": 1.5319587628865978,
      "grad_norm": 1.140625,
      "learning_rate": 0.000476569475179589,
      "loss": 4.9385,
      "mean_token_accuracy": 0.20380509942770003,
      "num_tokens": 34164523.0,
      "step": 19690
    },
    {
      "entropy": 5.626768398284912,
      "epoch": 1.5323477922583155,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047655715884738074,
      "loss": 5.1945,
      "mean_token_accuracy": 0.19471436738967896,
      "num_tokens": 34174084.0,
      "step": 19695
    },
    {
      "entropy": 5.690047740936279,
      "epoch": 1.5327368216300332,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004765448394568632,
      "loss": 5.0932,
      "mean_token_accuracy": 0.19534295201301574,
      "num_tokens": 34182551.0,
      "step": 19700
    },
    {
      "entropy": 5.582829809188842,
      "epoch": 1.5331258510017505,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004765325170082237,
      "loss": 4.9426,
      "mean_token_accuracy": 0.20883558243513106,
      "num_tokens": 34190899.0,
      "step": 19705
    },
    {
      "entropy": 5.602930355072021,
      "epoch": 1.5335148803734682,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000476520191501649,
      "loss": 5.1503,
      "mean_token_accuracy": 0.1931648448109627,
      "num_tokens": 34200394.0,
      "step": 19710
    },
    {
      "entropy": 5.669780588150024,
      "epoch": 1.5339039097451859,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047650786293732607,
      "loss": 5.1346,
      "mean_token_accuracy": 0.1856587454676628,
      "num_tokens": 34209773.0,
      "step": 19715
    },
    {
      "entropy": 5.660230350494385,
      "epoch": 1.5342929391169033,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004764955313154421,
      "loss": 5.0693,
      "mean_token_accuracy": 0.1954827845096588,
      "num_tokens": 34218487.0,
      "step": 19720
    },
    {
      "entropy": 5.644893312454224,
      "epoch": 1.5346819684886208,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047648319663618415,
      "loss": 5.1257,
      "mean_token_accuracy": 0.20266671031713485,
      "num_tokens": 34227309.0,
      "step": 19725
    },
    {
      "entropy": 5.6277666091918945,
      "epoch": 1.5350709978603385,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047647085889973936,
      "loss": 5.0757,
      "mean_token_accuracy": 0.1967098072171211,
      "num_tokens": 34235877.0,
      "step": 19730
    },
    {
      "entropy": 5.627515888214111,
      "epoch": 1.535460027232056,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047645851810629503,
      "loss": 5.0395,
      "mean_token_accuracy": 0.19868775606155395,
      "num_tokens": 34244073.0,
      "step": 19735
    },
    {
      "entropy": 5.6074498176574705,
      "epoch": 1.5358490566037735,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047644617425603825,
      "loss": 5.0788,
      "mean_token_accuracy": 0.19275624603033065,
      "num_tokens": 34252255.0,
      "step": 19740
    },
    {
      "entropy": 5.6151293277740475,
      "epoch": 1.5362380859754912,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004764338273491565,
      "loss": 5.0686,
      "mean_token_accuracy": 0.1988752841949463,
      "num_tokens": 34261063.0,
      "step": 19745
    },
    {
      "entropy": 5.6376872062683105,
      "epoch": 1.5366271153472089,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047642147738583695,
      "loss": 5.1095,
      "mean_token_accuracy": 0.1929849550127983,
      "num_tokens": 34269072.0,
      "step": 19750
    },
    {
      "entropy": 5.6929536819458,
      "epoch": 1.5370161447189261,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004764091243662671,
      "loss": 5.0951,
      "mean_token_accuracy": 0.1925562933087349,
      "num_tokens": 34277845.0,
      "step": 19755
    },
    {
      "entropy": 5.50715217590332,
      "epoch": 1.5374051740906438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047639676829063437,
      "loss": 4.9282,
      "mean_token_accuracy": 0.20247926861047744,
      "num_tokens": 34285831.0,
      "step": 19760
    },
    {
      "entropy": 5.65914740562439,
      "epoch": 1.5377942034623615,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047638440915912623,
      "loss": 5.1262,
      "mean_token_accuracy": 0.1922749936580658,
      "num_tokens": 34294095.0,
      "step": 19765
    },
    {
      "entropy": 5.686305522918701,
      "epoch": 1.538183232834079,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004763720469719303,
      "loss": 5.0746,
      "mean_token_accuracy": 0.19617477506399156,
      "num_tokens": 34301570.0,
      "step": 19770
    },
    {
      "entropy": 5.622435522079468,
      "epoch": 1.5385722622057965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000476359681729234,
      "loss": 5.1067,
      "mean_token_accuracy": 0.19251677691936492,
      "num_tokens": 34310119.0,
      "step": 19775
    },
    {
      "entropy": 5.62856912612915,
      "epoch": 1.5389612915775142,
      "grad_norm": 1.15625,
      "learning_rate": 0.000476347313431225,
      "loss": 5.0238,
      "mean_token_accuracy": 0.2005590945482254,
      "num_tokens": 34319622.0,
      "step": 19780
    },
    {
      "entropy": 5.586194849014282,
      "epoch": 1.5393503209492316,
      "grad_norm": 1.0,
      "learning_rate": 0.00047633494207809096,
      "loss": 5.0356,
      "mean_token_accuracy": 0.19964371919631957,
      "num_tokens": 34328376.0,
      "step": 19785
    },
    {
      "entropy": 5.5850992679595945,
      "epoch": 1.539739350320949,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047632256767001977,
      "loss": 5.0821,
      "mean_token_accuracy": 0.19479367434978484,
      "num_tokens": 34337430.0,
      "step": 19790
    },
    {
      "entropy": 5.660489177703857,
      "epoch": 1.5401283796926668,
      "grad_norm": 1.09375,
      "learning_rate": 0.000476310190207199,
      "loss": 5.0811,
      "mean_token_accuracy": 0.19488630294799805,
      "num_tokens": 34345639.0,
      "step": 19795
    },
    {
      "entropy": 5.6912225723266605,
      "epoch": 1.5405174090643845,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047629780968981653,
      "loss": 5.0813,
      "mean_token_accuracy": 0.19781474620103837,
      "num_tokens": 34354083.0,
      "step": 19800
    },
    {
      "entropy": 5.585956335067749,
      "epoch": 1.540906438436102,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047628542611806007,
      "loss": 5.0535,
      "mean_token_accuracy": 0.19291966706514357,
      "num_tokens": 34363882.0,
      "step": 19805
    },
    {
      "entropy": 5.554418420791626,
      "epoch": 1.5412954678078195,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047627303949211773,
      "loss": 5.0023,
      "mean_token_accuracy": 0.19834552109241485,
      "num_tokens": 34372487.0,
      "step": 19810
    },
    {
      "entropy": 5.59479603767395,
      "epoch": 1.5416844971795371,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004762606498121774,
      "loss": 5.0322,
      "mean_token_accuracy": 0.19938748180866242,
      "num_tokens": 34380729.0,
      "step": 19815
    },
    {
      "entropy": 5.6450221061706545,
      "epoch": 1.5420735265512546,
      "grad_norm": 1.15625,
      "learning_rate": 0.000476248257078427,
      "loss": 5.0817,
      "mean_token_accuracy": 0.19415965676307678,
      "num_tokens": 34390121.0,
      "step": 19820
    },
    {
      "entropy": 5.60985803604126,
      "epoch": 1.542462555922972,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004762358612910547,
      "loss": 5.041,
      "mean_token_accuracy": 0.1984054520726204,
      "num_tokens": 34398460.0,
      "step": 19825
    },
    {
      "entropy": 5.607812738418579,
      "epoch": 1.5428515852946898,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004762234624502484,
      "loss": 5.0415,
      "mean_token_accuracy": 0.19524168968200684,
      "num_tokens": 34407406.0,
      "step": 19830
    },
    {
      "entropy": 5.6053698539733885,
      "epoch": 1.5432406146664073,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047621106055619644,
      "loss": 5.0016,
      "mean_token_accuracy": 0.20045041888952256,
      "num_tokens": 34416588.0,
      "step": 19835
    },
    {
      "entropy": 5.624873542785645,
      "epoch": 1.5436296440381247,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047619865560908687,
      "loss": 5.0621,
      "mean_token_accuracy": 0.1941896751523018,
      "num_tokens": 34425266.0,
      "step": 19840
    },
    {
      "entropy": 5.627095174789429,
      "epoch": 1.5440186734098424,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004761862476091079,
      "loss": 5.031,
      "mean_token_accuracy": 0.1932322636246681,
      "num_tokens": 34433790.0,
      "step": 19845
    },
    {
      "entropy": 5.685766983032226,
      "epoch": 1.5444077027815601,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047617383655644785,
      "loss": 5.1525,
      "mean_token_accuracy": 0.18489960730075836,
      "num_tokens": 34442733.0,
      "step": 19850
    },
    {
      "entropy": 5.693801641464233,
      "epoch": 1.5447967321532776,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004761614224512951,
      "loss": 5.0434,
      "mean_token_accuracy": 0.19998459666967391,
      "num_tokens": 34451756.0,
      "step": 19855
    },
    {
      "entropy": 5.616048765182495,
      "epoch": 1.545185761524995,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004761490052938379,
      "loss": 5.0738,
      "mean_token_accuracy": 0.19717212170362472,
      "num_tokens": 34460003.0,
      "step": 19860
    },
    {
      "entropy": 5.606335687637329,
      "epoch": 1.5455747908967128,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004761365850842646,
      "loss": 5.0699,
      "mean_token_accuracy": 0.19799195677042009,
      "num_tokens": 34468485.0,
      "step": 19865
    },
    {
      "entropy": 5.6451362609863285,
      "epoch": 1.5459638202684303,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004761241618227639,
      "loss": 5.0012,
      "mean_token_accuracy": 0.19553233534097672,
      "num_tokens": 34477221.0,
      "step": 19870
    },
    {
      "entropy": 5.65382604598999,
      "epoch": 1.5463528496401477,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047611173550952414,
      "loss": 5.1529,
      "mean_token_accuracy": 0.19217605739831925,
      "num_tokens": 34486339.0,
      "step": 19875
    },
    {
      "entropy": 5.635845947265625,
      "epoch": 1.5467418790118654,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047609930614473387,
      "loss": 5.0197,
      "mean_token_accuracy": 0.19935112446546555,
      "num_tokens": 34494595.0,
      "step": 19880
    },
    {
      "entropy": 5.58845887184143,
      "epoch": 1.547130908383583,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047608687372858175,
      "loss": 4.9813,
      "mean_token_accuracy": 0.19906204342842101,
      "num_tokens": 34502702.0,
      "step": 19885
    },
    {
      "entropy": 5.647160196304322,
      "epoch": 1.5475199377553004,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047607443826125633,
      "loss": 5.1251,
      "mean_token_accuracy": 0.19534933269023896,
      "num_tokens": 34510828.0,
      "step": 19890
    },
    {
      "entropy": 5.709363412857056,
      "epoch": 1.547908967127018,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047606199974294637,
      "loss": 5.1891,
      "mean_token_accuracy": 0.19509582072496415,
      "num_tokens": 34520288.0,
      "step": 19895
    },
    {
      "entropy": 5.6825488090515135,
      "epoch": 1.5482979964987358,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004760495581738406,
      "loss": 5.0566,
      "mean_token_accuracy": 0.20270530879497528,
      "num_tokens": 34528178.0,
      "step": 19900
    },
    {
      "entropy": 5.636096239089966,
      "epoch": 1.5486870258704533,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004760371135541278,
      "loss": 5.1029,
      "mean_token_accuracy": 0.18809671252965926,
      "num_tokens": 34536915.0,
      "step": 19905
    },
    {
      "entropy": 5.66641206741333,
      "epoch": 1.5490760552421707,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004760246658839967,
      "loss": 5.0548,
      "mean_token_accuracy": 0.19265463203191757,
      "num_tokens": 34545009.0,
      "step": 19910
    },
    {
      "entropy": 5.669793939590454,
      "epoch": 1.5494650846138884,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004760122151636364,
      "loss": 5.1462,
      "mean_token_accuracy": 0.1940530702471733,
      "num_tokens": 34553936.0,
      "step": 19915
    },
    {
      "entropy": 5.661856174468994,
      "epoch": 1.549854113985606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004759997613932356,
      "loss": 5.0243,
      "mean_token_accuracy": 0.2001774340867996,
      "num_tokens": 34563068.0,
      "step": 19920
    },
    {
      "entropy": 5.665142250061035,
      "epoch": 1.5502431433573234,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047598730457298335,
      "loss": 5.1666,
      "mean_token_accuracy": 0.18391934782266617,
      "num_tokens": 34571787.0,
      "step": 19925
    },
    {
      "entropy": 5.695776271820068,
      "epoch": 1.550632172729041,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047597484470306866,
      "loss": 5.0654,
      "mean_token_accuracy": 0.19397427886724472,
      "num_tokens": 34580375.0,
      "step": 19930
    },
    {
      "entropy": 5.688618230819702,
      "epoch": 1.5510212021007586,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004759623817836806,
      "loss": 5.1049,
      "mean_token_accuracy": 0.19783754348754884,
      "num_tokens": 34588762.0,
      "step": 19935
    },
    {
      "entropy": 5.59483494758606,
      "epoch": 1.551410231472476,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004759499158150082,
      "loss": 4.9973,
      "mean_token_accuracy": 0.2052197962999344,
      "num_tokens": 34596986.0,
      "step": 19940
    },
    {
      "entropy": 5.709632158279419,
      "epoch": 1.5517992608441937,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047593744679724076,
      "loss": 5.1209,
      "mean_token_accuracy": 0.19396817684173584,
      "num_tokens": 34605445.0,
      "step": 19945
    },
    {
      "entropy": 5.618798112869262,
      "epoch": 1.5521882902159114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047592497473056733,
      "loss": 5.1045,
      "mean_token_accuracy": 0.1942835971713066,
      "num_tokens": 34614793.0,
      "step": 19950
    },
    {
      "entropy": 5.7228367805480955,
      "epoch": 1.552577319587629,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047591249961517724,
      "loss": 5.2085,
      "mean_token_accuracy": 0.1784688040614128,
      "num_tokens": 34623790.0,
      "step": 19955
    },
    {
      "entropy": 5.611024665832519,
      "epoch": 1.5529663489593464,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004759000214512598,
      "loss": 5.0571,
      "mean_token_accuracy": 0.196479195356369,
      "num_tokens": 34632387.0,
      "step": 19960
    },
    {
      "entropy": 5.569279050827026,
      "epoch": 1.553355378331064,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004758875402390042,
      "loss": 5.0507,
      "mean_token_accuracy": 0.20331077873706818,
      "num_tokens": 34640097.0,
      "step": 19965
    },
    {
      "entropy": 5.607867193222046,
      "epoch": 1.5537444077027815,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047587505597859996,
      "loss": 5.0559,
      "mean_token_accuracy": 0.19027096182107925,
      "num_tokens": 34649525.0,
      "step": 19970
    },
    {
      "entropy": 5.591685009002686,
      "epoch": 1.554133437074499,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047586256867023646,
      "loss": 5.0012,
      "mean_token_accuracy": 0.2034752756357193,
      "num_tokens": 34658467.0,
      "step": 19975
    },
    {
      "entropy": 5.694282150268554,
      "epoch": 1.5545224664462167,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004758500783141032,
      "loss": 5.1483,
      "mean_token_accuracy": 0.1910340443253517,
      "num_tokens": 34666926.0,
      "step": 19980
    },
    {
      "entropy": 5.679008960723877,
      "epoch": 1.5549114958179344,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004758375849103897,
      "loss": 5.0726,
      "mean_token_accuracy": 0.19341611564159394,
      "num_tokens": 34675023.0,
      "step": 19985
    },
    {
      "entropy": 5.6026839256286625,
      "epoch": 1.5553005251896517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004758250884592855,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20133753567934037,
      "num_tokens": 34683191.0,
      "step": 19990
    },
    {
      "entropy": 5.616697645187378,
      "epoch": 1.5556895545613694,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047581258896098024,
      "loss": 5.0416,
      "mean_token_accuracy": 0.20284320861101152,
      "num_tokens": 34691899.0,
      "step": 19995
    },
    {
      "entropy": 5.645014047622681,
      "epoch": 1.556078583933087,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004758000864156636,
      "loss": 5.1262,
      "mean_token_accuracy": 0.19397966414690018,
      "num_tokens": 34700942.0,
      "step": 20000
    },
    {
      "entropy": 5.665093231201172,
      "epoch": 1.5564676133048045,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047578758082352527,
      "loss": 5.1599,
      "mean_token_accuracy": 0.19051943570375443,
      "num_tokens": 34710780.0,
      "step": 20005
    },
    {
      "entropy": 5.657600784301758,
      "epoch": 1.556856642676522,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047577507218475487,
      "loss": 5.0605,
      "mean_token_accuracy": 0.1937171623110771,
      "num_tokens": 34719332.0,
      "step": 20010
    },
    {
      "entropy": 5.676499223709106,
      "epoch": 1.5572456720482397,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047576256049954236,
      "loss": 5.1172,
      "mean_token_accuracy": 0.18435289561748505,
      "num_tokens": 34727890.0,
      "step": 20015
    },
    {
      "entropy": 5.643419790267944,
      "epoch": 1.5576347014199572,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004757500457680776,
      "loss": 5.099,
      "mean_token_accuracy": 0.19545732885599137,
      "num_tokens": 34736942.0,
      "step": 20020
    },
    {
      "entropy": 5.610087060928345,
      "epoch": 1.5580237307916747,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004757375279905504,
      "loss": 4.9548,
      "mean_token_accuracy": 0.20365367084741592,
      "num_tokens": 34745612.0,
      "step": 20025
    },
    {
      "entropy": 5.634346199035645,
      "epoch": 1.5584127601633924,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047572500716715075,
      "loss": 5.0568,
      "mean_token_accuracy": 0.19421056658029556,
      "num_tokens": 34754276.0,
      "step": 20030
    },
    {
      "entropy": 5.603091382980347,
      "epoch": 1.55880178953511,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047571248329806855,
      "loss": 4.9568,
      "mean_token_accuracy": 0.2041890576481819,
      "num_tokens": 34762042.0,
      "step": 20035
    },
    {
      "entropy": 5.564679431915283,
      "epoch": 1.5591908189068273,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047569995638349383,
      "loss": 4.9987,
      "mean_token_accuracy": 0.19739745259284974,
      "num_tokens": 34770172.0,
      "step": 20040
    },
    {
      "entropy": 5.593704557418823,
      "epoch": 1.559579848278545,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004756874264236168,
      "loss": 5.0106,
      "mean_token_accuracy": 0.20030488669872284,
      "num_tokens": 34778647.0,
      "step": 20045
    },
    {
      "entropy": 5.610961294174194,
      "epoch": 1.5599688776502627,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047567489341862753,
      "loss": 5.0453,
      "mean_token_accuracy": 0.1973257318139076,
      "num_tokens": 34787409.0,
      "step": 20050
    },
    {
      "entropy": 5.6935631275177006,
      "epoch": 1.5603579070219802,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047566235736871607,
      "loss": 5.1361,
      "mean_token_accuracy": 0.19000124782323838,
      "num_tokens": 34795634.0,
      "step": 20055
    },
    {
      "entropy": 5.627721214294434,
      "epoch": 1.5607469363936977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047564981827407277,
      "loss": 4.9943,
      "mean_token_accuracy": 0.2055493637919426,
      "num_tokens": 34803803.0,
      "step": 20060
    },
    {
      "entropy": 5.611069202423096,
      "epoch": 1.5611359657654154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047563727613488785,
      "loss": 5.0003,
      "mean_token_accuracy": 0.19775864779949187,
      "num_tokens": 34813202.0,
      "step": 20065
    },
    {
      "entropy": 5.627691268920898,
      "epoch": 1.5615249951371328,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047562473095135154,
      "loss": 5.107,
      "mean_token_accuracy": 0.19250735640525818,
      "num_tokens": 34821717.0,
      "step": 20070
    },
    {
      "entropy": 5.628086185455322,
      "epoch": 1.5619140245088503,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004756121827236544,
      "loss": 5.0173,
      "mean_token_accuracy": 0.19739003777503966,
      "num_tokens": 34830366.0,
      "step": 20075
    },
    {
      "entropy": 5.632188558578491,
      "epoch": 1.562303053880568,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004755996314519866,
      "loss": 5.0864,
      "mean_token_accuracy": 0.1988873600959778,
      "num_tokens": 34838734.0,
      "step": 20080
    },
    {
      "entropy": 5.562366199493408,
      "epoch": 1.5626920832522857,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004755870771365387,
      "loss": 5.001,
      "mean_token_accuracy": 0.20255364328622819,
      "num_tokens": 34846873.0,
      "step": 20085
    },
    {
      "entropy": 5.661673831939697,
      "epoch": 1.563081112624003,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047557451977750113,
      "loss": 5.0522,
      "mean_token_accuracy": 0.1934480294585228,
      "num_tokens": 34855542.0,
      "step": 20090
    },
    {
      "entropy": 5.614608907699585,
      "epoch": 1.5634701419957207,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004755619593750645,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20677870661020278,
      "num_tokens": 34864656.0,
      "step": 20095
    },
    {
      "entropy": 5.64049072265625,
      "epoch": 1.5638591713674383,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004755493959294194,
      "loss": 5.0992,
      "mean_token_accuracy": 0.19462711811065675,
      "num_tokens": 34873891.0,
      "step": 20100
    },
    {
      "entropy": 5.636171960830689,
      "epoch": 1.5642482007391558,
      "grad_norm": 1.125,
      "learning_rate": 0.0004755368294407564,
      "loss": 5.0955,
      "mean_token_accuracy": 0.19437652975320815,
      "num_tokens": 34883278.0,
      "step": 20105
    },
    {
      "entropy": 5.567042493820191,
      "epoch": 1.5646372301108733,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004755242599092662,
      "loss": 5.0096,
      "mean_token_accuracy": 0.20047936588525772,
      "num_tokens": 34892323.0,
      "step": 20110
    },
    {
      "entropy": 5.661008501052857,
      "epoch": 1.565026259482591,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047551168733513956,
      "loss": 5.0619,
      "mean_token_accuracy": 0.19229132384061814,
      "num_tokens": 34901774.0,
      "step": 20115
    },
    {
      "entropy": 5.717881345748902,
      "epoch": 1.5654152888543085,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047549911171856717,
      "loss": 5.0733,
      "mean_token_accuracy": 0.19608502089977264,
      "num_tokens": 34911082.0,
      "step": 20120
    },
    {
      "entropy": 5.655798101425171,
      "epoch": 1.565804318226026,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004754865330597398,
      "loss": 5.0725,
      "mean_token_accuracy": 0.1976458489894867,
      "num_tokens": 34919631.0,
      "step": 20125
    },
    {
      "entropy": 5.6539222240448,
      "epoch": 1.5661933475977436,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047547395135884854,
      "loss": 5.075,
      "mean_token_accuracy": 0.19253887832164765,
      "num_tokens": 34928258.0,
      "step": 20130
    },
    {
      "entropy": 5.629160833358765,
      "epoch": 1.5665823769694613,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004754613666160841,
      "loss": 5.0655,
      "mean_token_accuracy": 0.19675603806972503,
      "num_tokens": 34936590.0,
      "step": 20135
    },
    {
      "entropy": 5.582829284667969,
      "epoch": 1.5669714063411786,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047544877883163753,
      "loss": 5.0901,
      "mean_token_accuracy": 0.18784206956624985,
      "num_tokens": 34945468.0,
      "step": 20140
    },
    {
      "entropy": 5.588636589050293,
      "epoch": 1.5673604357128963,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047543618800569975,
      "loss": 5.0063,
      "mean_token_accuracy": 0.20052008628845214,
      "num_tokens": 34954005.0,
      "step": 20145
    },
    {
      "entropy": 5.568189287185669,
      "epoch": 1.567749465084614,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047542359413846184,
      "loss": 5.0066,
      "mean_token_accuracy": 0.19526299387216567,
      "num_tokens": 34962852.0,
      "step": 20150
    },
    {
      "entropy": 5.640872955322266,
      "epoch": 1.5681384944563315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004754109972301149,
      "loss": 5.0869,
      "mean_token_accuracy": 0.18892480283975602,
      "num_tokens": 34971718.0,
      "step": 20155
    },
    {
      "entropy": 5.669030952453613,
      "epoch": 1.568527523828049,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047539839728085007,
      "loss": 5.017,
      "mean_token_accuracy": 0.2027488976716995,
      "num_tokens": 34980155.0,
      "step": 20160
    },
    {
      "entropy": 5.698576593399048,
      "epoch": 1.5689165531997666,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004753857942908585,
      "loss": 5.2282,
      "mean_token_accuracy": 0.18437210470438004,
      "num_tokens": 34989971.0,
      "step": 20165
    },
    {
      "entropy": 5.598672199249267,
      "epoch": 1.5693055825714841,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004753731882603315,
      "loss": 5.0293,
      "mean_token_accuracy": 0.1893721714615822,
      "num_tokens": 34998760.0,
      "step": 20170
    },
    {
      "entropy": 5.630321455001831,
      "epoch": 1.5696946119432016,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047536057918946026,
      "loss": 5.0792,
      "mean_token_accuracy": 0.19391676783561707,
      "num_tokens": 35008020.0,
      "step": 20175
    },
    {
      "entropy": 5.556445217132568,
      "epoch": 1.5700836413149193,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004753479670784361,
      "loss": 4.9488,
      "mean_token_accuracy": 0.20516219288110732,
      "num_tokens": 35016357.0,
      "step": 20180
    },
    {
      "entropy": 5.586726331710816,
      "epoch": 1.570472670686637,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004753353519274505,
      "loss": 5.0611,
      "mean_token_accuracy": 0.19814965426921843,
      "num_tokens": 35025312.0,
      "step": 20185
    },
    {
      "entropy": 5.635017347335816,
      "epoch": 1.5708617000583545,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004753227337366948,
      "loss": 5.1322,
      "mean_token_accuracy": 0.1941147953271866,
      "num_tokens": 35033781.0,
      "step": 20190
    },
    {
      "entropy": 5.697467374801636,
      "epoch": 1.571250729430072,
      "grad_norm": 1.125,
      "learning_rate": 0.00047531011250636046,
      "loss": 5.1446,
      "mean_token_accuracy": 0.19301952570676803,
      "num_tokens": 35042884.0,
      "step": 20195
    },
    {
      "entropy": 5.628695440292359,
      "epoch": 1.5716397588017896,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047529748823663896,
      "loss": 4.9863,
      "mean_token_accuracy": 0.20543665885925294,
      "num_tokens": 35051789.0,
      "step": 20200
    },
    {
      "entropy": 5.6485589981079105,
      "epoch": 1.572028788173507,
      "grad_norm": 1.125,
      "learning_rate": 0.0004752848609277219,
      "loss": 5.0795,
      "mean_token_accuracy": 0.1973910391330719,
      "num_tokens": 35059958.0,
      "step": 20205
    },
    {
      "entropy": 5.632623386383057,
      "epoch": 1.5724178175452246,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047527223057980086,
      "loss": 5.0444,
      "mean_token_accuracy": 0.20189540088176727,
      "num_tokens": 35068384.0,
      "step": 20210
    },
    {
      "entropy": 5.685867166519165,
      "epoch": 1.5728068469169423,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047525959719306736,
      "loss": 5.114,
      "mean_token_accuracy": 0.18957228511571883,
      "num_tokens": 35076660.0,
      "step": 20215
    },
    {
      "entropy": 5.657887744903564,
      "epoch": 1.5731958762886598,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004752469607677134,
      "loss": 5.0324,
      "mean_token_accuracy": 0.20160100013017654,
      "num_tokens": 35085527.0,
      "step": 20220
    },
    {
      "entropy": 5.615867233276367,
      "epoch": 1.5735849056603772,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004752343213039305,
      "loss": 5.0224,
      "mean_token_accuracy": 0.1970536783337593,
      "num_tokens": 35094066.0,
      "step": 20225
    },
    {
      "entropy": 5.670680189132691,
      "epoch": 1.573973935032095,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004752216788019104,
      "loss": 5.1801,
      "mean_token_accuracy": 0.18883613646030425,
      "num_tokens": 35103219.0,
      "step": 20230
    },
    {
      "entropy": 5.684045076370239,
      "epoch": 1.5743629644038126,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004752090332618451,
      "loss": 5.0873,
      "mean_token_accuracy": 0.19407470673322677,
      "num_tokens": 35111336.0,
      "step": 20235
    },
    {
      "entropy": 5.662780809402466,
      "epoch": 1.57475199377553,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004751963846839263,
      "loss": 5.0485,
      "mean_token_accuracy": 0.20413943082094194,
      "num_tokens": 35119758.0,
      "step": 20240
    },
    {
      "entropy": 5.560201740264892,
      "epoch": 1.5751410231472476,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047518373306834605,
      "loss": 4.983,
      "mean_token_accuracy": 0.20027344226837157,
      "num_tokens": 35128535.0,
      "step": 20245
    },
    {
      "entropy": 5.625986528396607,
      "epoch": 1.5755300525189653,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047517107841529626,
      "loss": 5.134,
      "mean_token_accuracy": 0.189190836250782,
      "num_tokens": 35137443.0,
      "step": 20250
    },
    {
      "entropy": 5.561290884017945,
      "epoch": 1.5759190818906827,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000475158420724969,
      "loss": 4.9266,
      "mean_token_accuracy": 0.20940395146608354,
      "num_tokens": 35145801.0,
      "step": 20255
    },
    {
      "entropy": 5.694694566726684,
      "epoch": 1.5763081112624002,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047514575999755627,
      "loss": 5.1562,
      "mean_token_accuracy": 0.18767069578170775,
      "num_tokens": 35154661.0,
      "step": 20260
    },
    {
      "entropy": 5.66506061553955,
      "epoch": 1.576697140634118,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047513309623325024,
      "loss": 5.1608,
      "mean_token_accuracy": 0.18657090067863463,
      "num_tokens": 35163762.0,
      "step": 20265
    },
    {
      "entropy": 5.64682354927063,
      "epoch": 1.5770861700058354,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004751204294322429,
      "loss": 5.0665,
      "mean_token_accuracy": 0.19244659096002578,
      "num_tokens": 35172059.0,
      "step": 20270
    },
    {
      "entropy": 5.642732095718384,
      "epoch": 1.5774751993775529,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047510775959472675,
      "loss": 5.1283,
      "mean_token_accuracy": 0.1892454966902733,
      "num_tokens": 35179805.0,
      "step": 20275
    },
    {
      "entropy": 5.585176944732666,
      "epoch": 1.5778642287492706,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004750950867208937,
      "loss": 5.0817,
      "mean_token_accuracy": 0.19841177761554718,
      "num_tokens": 35188136.0,
      "step": 20280
    },
    {
      "entropy": 5.736662435531616,
      "epoch": 1.5782532581209883,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004750824108109362,
      "loss": 5.1645,
      "mean_token_accuracy": 0.1901683747768402,
      "num_tokens": 35196577.0,
      "step": 20285
    },
    {
      "entropy": 5.61679253578186,
      "epoch": 1.5786422874927057,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004750697318650466,
      "loss": 5.0736,
      "mean_token_accuracy": 0.19375281482934953,
      "num_tokens": 35205223.0,
      "step": 20290
    },
    {
      "entropy": 5.69085545539856,
      "epoch": 1.5790313168644232,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004750570498834173,
      "loss": 5.1416,
      "mean_token_accuracy": 0.187985922396183,
      "num_tokens": 35213776.0,
      "step": 20295
    },
    {
      "entropy": 5.618429803848267,
      "epoch": 1.579420346236141,
      "grad_norm": 1.125,
      "learning_rate": 0.0004750443648662407,
      "loss": 5.0696,
      "mean_token_accuracy": 0.19948683083057403,
      "num_tokens": 35222596.0,
      "step": 20300
    },
    {
      "entropy": 5.7082775115966795,
      "epoch": 1.5798093756078584,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047503167681370924,
      "loss": 5.2039,
      "mean_token_accuracy": 0.18726951330900193,
      "num_tokens": 35232402.0,
      "step": 20305
    },
    {
      "entropy": 5.639067220687866,
      "epoch": 1.5801984049795759,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004750189857260154,
      "loss": 4.9905,
      "mean_token_accuracy": 0.20895290076732637,
      "num_tokens": 35240651.0,
      "step": 20310
    },
    {
      "entropy": 5.581398582458496,
      "epoch": 1.5805874343512936,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004750062916033519,
      "loss": 5.027,
      "mean_token_accuracy": 0.19455200731754302,
      "num_tokens": 35249101.0,
      "step": 20315
    },
    {
      "entropy": 5.647253561019897,
      "epoch": 1.580976463723011,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004749935944459113,
      "loss": 5.1305,
      "mean_token_accuracy": 0.18680695444345474,
      "num_tokens": 35257909.0,
      "step": 20320
    },
    {
      "entropy": 5.706323909759521,
      "epoch": 1.5813654930947285,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004749808942538862,
      "loss": 5.1865,
      "mean_token_accuracy": 0.18164627701044084,
      "num_tokens": 35267128.0,
      "step": 20325
    },
    {
      "entropy": 5.670701122283935,
      "epoch": 1.5817545224664462,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004749681910274693,
      "loss": 5.0713,
      "mean_token_accuracy": 0.19438831955194474,
      "num_tokens": 35275751.0,
      "step": 20330
    },
    {
      "entropy": 5.604038524627685,
      "epoch": 1.582143551838164,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047495548476685334,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20060366541147232,
      "num_tokens": 35285359.0,
      "step": 20335
    },
    {
      "entropy": 5.577478981018066,
      "epoch": 1.5825325812098814,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047494277547223125,
      "loss": 4.9781,
      "mean_token_accuracy": 0.19926492124795914,
      "num_tokens": 35293691.0,
      "step": 20340
    },
    {
      "entropy": 5.582086610794067,
      "epoch": 1.5829216105815989,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047493006314379573,
      "loss": 5.0148,
      "mean_token_accuracy": 0.2020304471254349,
      "num_tokens": 35302751.0,
      "step": 20345
    },
    {
      "entropy": 5.70212755203247,
      "epoch": 1.5833106399533166,
      "grad_norm": 1.125,
      "learning_rate": 0.00047491734778173976,
      "loss": 5.093,
      "mean_token_accuracy": 0.19042124450206757,
      "num_tokens": 35311849.0,
      "step": 20350
    },
    {
      "entropy": 5.606876611709595,
      "epoch": 1.583699669325034,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047490462938625617,
      "loss": 5.0321,
      "mean_token_accuracy": 0.198320609331131,
      "num_tokens": 35320248.0,
      "step": 20355
    },
    {
      "entropy": 5.651252222061157,
      "epoch": 1.5840886986967515,
      "grad_norm": 1.125,
      "learning_rate": 0.00047489190795753806,
      "loss": 5.0366,
      "mean_token_accuracy": 0.20442617684602737,
      "num_tokens": 35328565.0,
      "step": 20360
    },
    {
      "entropy": 5.604735374450684,
      "epoch": 1.5844777280684692,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004748791834957784,
      "loss": 5.0352,
      "mean_token_accuracy": 0.20153813660144806,
      "num_tokens": 35337828.0,
      "step": 20365
    },
    {
      "entropy": 5.640591764450074,
      "epoch": 1.5848667574401867,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047486645600117037,
      "loss": 5.1281,
      "mean_token_accuracy": 0.18953794836997986,
      "num_tokens": 35347107.0,
      "step": 20370
    },
    {
      "entropy": 5.6811388492584225,
      "epoch": 1.5852557868119042,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004748537254739068,
      "loss": 5.0515,
      "mean_token_accuracy": 0.1988525688648224,
      "num_tokens": 35355925.0,
      "step": 20375
    },
    {
      "entropy": 5.623303079605103,
      "epoch": 1.5856448161836219,
      "grad_norm": 1.125,
      "learning_rate": 0.0004748409919141812,
      "loss": 5.0239,
      "mean_token_accuracy": 0.19985021501779557,
      "num_tokens": 35364319.0,
      "step": 20380
    },
    {
      "entropy": 5.546020793914795,
      "epoch": 1.5860338455553395,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004748282553221865,
      "loss": 4.9642,
      "mean_token_accuracy": 0.20473470389842988,
      "num_tokens": 35372358.0,
      "step": 20385
    },
    {
      "entropy": 5.639964818954468,
      "epoch": 1.586422874927057,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004748155156981162,
      "loss": 5.2054,
      "mean_token_accuracy": 0.18248383104801177,
      "num_tokens": 35381301.0,
      "step": 20390
    },
    {
      "entropy": 5.651611757278443,
      "epoch": 1.5868119042987745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047480277304216346,
      "loss": 5.0181,
      "mean_token_accuracy": 0.19694725126028062,
      "num_tokens": 35389778.0,
      "step": 20395
    },
    {
      "entropy": 5.6118851661682125,
      "epoch": 1.5872009336704922,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047479002735452164,
      "loss": 4.9859,
      "mean_token_accuracy": 0.19889788329601288,
      "num_tokens": 35397926.0,
      "step": 20400
    },
    {
      "entropy": 5.5962324142456055,
      "epoch": 1.5875899630422097,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047477727863538415,
      "loss": 5.0185,
      "mean_token_accuracy": 0.20125045776367187,
      "num_tokens": 35406462.0,
      "step": 20405
    },
    {
      "entropy": 5.679141616821289,
      "epoch": 1.5879789924139271,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047476452688494444,
      "loss": 5.0256,
      "mean_token_accuracy": 0.19560559540987016,
      "num_tokens": 35414989.0,
      "step": 20410
    },
    {
      "entropy": 5.667212677001953,
      "epoch": 1.5883680217856448,
      "grad_norm": 1.15625,
      "learning_rate": 0.000474751772103396,
      "loss": 5.0136,
      "mean_token_accuracy": 0.20355913341045379,
      "num_tokens": 35423738.0,
      "step": 20415
    },
    {
      "entropy": 5.67338376045227,
      "epoch": 1.5887570511573625,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004747390142909323,
      "loss": 5.0561,
      "mean_token_accuracy": 0.19209839254617692,
      "num_tokens": 35432377.0,
      "step": 20420
    },
    {
      "entropy": 5.589076089859009,
      "epoch": 1.5891460805290798,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047472625344774713,
      "loss": 5.0438,
      "mean_token_accuracy": 0.20113733559846877,
      "num_tokens": 35441188.0,
      "step": 20425
    },
    {
      "entropy": 5.6141499996185305,
      "epoch": 1.5895351099007975,
      "grad_norm": 1.125,
      "learning_rate": 0.00047471348957403374,
      "loss": 5.1025,
      "mean_token_accuracy": 0.1982101321220398,
      "num_tokens": 35449715.0,
      "step": 20430
    },
    {
      "entropy": 5.627524709701538,
      "epoch": 1.5899241392725152,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004747007226699862,
      "loss": 5.0656,
      "mean_token_accuracy": 0.1978027567267418,
      "num_tokens": 35458420.0,
      "step": 20435
    },
    {
      "entropy": 5.6718086242675785,
      "epoch": 1.5903131686442327,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047468795273579803,
      "loss": 4.9883,
      "mean_token_accuracy": 0.19835505038499832,
      "num_tokens": 35467072.0,
      "step": 20440
    },
    {
      "entropy": 5.638486337661743,
      "epoch": 1.5907021980159501,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004746751797716629,
      "loss": 5.0537,
      "mean_token_accuracy": 0.19619325697422027,
      "num_tokens": 35475539.0,
      "step": 20445
    },
    {
      "entropy": 5.636385822296143,
      "epoch": 1.5910912273876678,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004746624037777748,
      "loss": 5.0582,
      "mean_token_accuracy": 0.20286996513605118,
      "num_tokens": 35484260.0,
      "step": 20450
    },
    {
      "entropy": 5.665251350402832,
      "epoch": 1.5914802567593853,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047464962475432754,
      "loss": 4.9885,
      "mean_token_accuracy": 0.20174664109945298,
      "num_tokens": 35492539.0,
      "step": 20455
    },
    {
      "entropy": 5.718202590942383,
      "epoch": 1.5918692861311028,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000474636842701515,
      "loss": 5.1516,
      "mean_token_accuracy": 0.19236619174480438,
      "num_tokens": 35502008.0,
      "step": 20460
    },
    {
      "entropy": 5.6056407451629635,
      "epoch": 1.5922583155028205,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004746240576195311,
      "loss": 5.0198,
      "mean_token_accuracy": 0.19924120903015136,
      "num_tokens": 35510677.0,
      "step": 20465
    },
    {
      "entropy": 5.60391092300415,
      "epoch": 1.5926473448745382,
      "grad_norm": 1.125,
      "learning_rate": 0.00047461126950856987,
      "loss": 5.0622,
      "mean_token_accuracy": 0.19693288952112198,
      "num_tokens": 35519368.0,
      "step": 20470
    },
    {
      "entropy": 5.701827716827393,
      "epoch": 1.5930363742462554,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004745984783688253,
      "loss": 5.1815,
      "mean_token_accuracy": 0.1867993012070656,
      "num_tokens": 35527952.0,
      "step": 20475
    },
    {
      "entropy": 5.706687116622925,
      "epoch": 1.5934254036179731,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047458568420049153,
      "loss": 5.1105,
      "mean_token_accuracy": 0.1935567080974579,
      "num_tokens": 35537806.0,
      "step": 20480
    },
    {
      "entropy": 5.578287982940674,
      "epoch": 1.5938144329896908,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047457288700376274,
      "loss": 4.9572,
      "mean_token_accuracy": 0.1940494492650032,
      "num_tokens": 35546135.0,
      "step": 20485
    },
    {
      "entropy": 5.551790428161621,
      "epoch": 1.5942034623614083,
      "grad_norm": 1.125,
      "learning_rate": 0.00047456008677883304,
      "loss": 5.0313,
      "mean_token_accuracy": 0.19893161803483964,
      "num_tokens": 35555065.0,
      "step": 20490
    },
    {
      "entropy": 5.657136487960815,
      "epoch": 1.5945924917331258,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004745472835258966,
      "loss": 5.1352,
      "mean_token_accuracy": 0.19141838401556016,
      "num_tokens": 35564222.0,
      "step": 20495
    },
    {
      "entropy": 5.6803154945373535,
      "epoch": 1.5949815211048435,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047453447724514773,
      "loss": 5.099,
      "mean_token_accuracy": 0.1897246018052101,
      "num_tokens": 35573021.0,
      "step": 20500
    },
    {
      "entropy": 5.676893854141236,
      "epoch": 1.595370550476561,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004745216679367808,
      "loss": 5.1332,
      "mean_token_accuracy": 0.1959197148680687,
      "num_tokens": 35581105.0,
      "step": 20505
    },
    {
      "entropy": 5.692387914657592,
      "epoch": 1.5957595798482784,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004745088556009901,
      "loss": 5.1767,
      "mean_token_accuracy": 0.18537239879369735,
      "num_tokens": 35590169.0,
      "step": 20510
    },
    {
      "entropy": 5.643527221679688,
      "epoch": 1.5961486092199961,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004744960402379701,
      "loss": 5.0617,
      "mean_token_accuracy": 0.20321208089590073,
      "num_tokens": 35598760.0,
      "step": 20515
    },
    {
      "entropy": 5.690423107147216,
      "epoch": 1.5965376385917138,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047448322184791525,
      "loss": 5.1003,
      "mean_token_accuracy": 0.19415076971054077,
      "num_tokens": 35607014.0,
      "step": 20520
    },
    {
      "entropy": 5.689444875717163,
      "epoch": 1.596926667963431,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047447040043101994,
      "loss": 5.0687,
      "mean_token_accuracy": 0.19446302056312562,
      "num_tokens": 35615319.0,
      "step": 20525
    },
    {
      "entropy": 5.621392011642456,
      "epoch": 1.5973156973351488,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047445757598747886,
      "loss": 5.0904,
      "mean_token_accuracy": 0.1962117910385132,
      "num_tokens": 35623836.0,
      "step": 20530
    },
    {
      "entropy": 5.585716772079468,
      "epoch": 1.5977047267068665,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004744447485174864,
      "loss": 4.981,
      "mean_token_accuracy": 0.1976259395480156,
      "num_tokens": 35632601.0,
      "step": 20535
    },
    {
      "entropy": 5.734390735626221,
      "epoch": 1.598093756078584,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047443191802123746,
      "loss": 5.1334,
      "mean_token_accuracy": 0.19487304538488387,
      "num_tokens": 35642035.0,
      "step": 20540
    },
    {
      "entropy": 5.639154052734375,
      "epoch": 1.5984827854503014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047441908449892664,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20130744874477385,
      "num_tokens": 35650454.0,
      "step": 20545
    },
    {
      "entropy": 5.6871569633483885,
      "epoch": 1.5988718148220191,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047440624795074855,
      "loss": 5.0974,
      "mean_token_accuracy": 0.1976846382021904,
      "num_tokens": 35658945.0,
      "step": 20550
    },
    {
      "entropy": 5.674994468688965,
      "epoch": 1.5992608441937366,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047439340837689804,
      "loss": 5.0921,
      "mean_token_accuracy": 0.19574213027954102,
      "num_tokens": 35667193.0,
      "step": 20555
    },
    {
      "entropy": 5.605404663085937,
      "epoch": 1.599649873565454,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047438056577756986,
      "loss": 4.943,
      "mean_token_accuracy": 0.2057245135307312,
      "num_tokens": 35676100.0,
      "step": 20560
    },
    {
      "entropy": 5.647966051101685,
      "epoch": 1.6000389029371718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047436772015295903,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19314984381198883,
      "num_tokens": 35684352.0,
      "step": 20565
    },
    {
      "entropy": 5.673441696166992,
      "epoch": 1.6004279323088895,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047435487150326036,
      "loss": 5.0147,
      "mean_token_accuracy": 0.20399393886327744,
      "num_tokens": 35693044.0,
      "step": 20570
    },
    {
      "entropy": 5.591395711898803,
      "epoch": 1.6008169616806067,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004743420198286688,
      "loss": 5.0088,
      "mean_token_accuracy": 0.20347841084003448,
      "num_tokens": 35701172.0,
      "step": 20575
    },
    {
      "entropy": 5.62863564491272,
      "epoch": 1.6012059910523244,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047432916512937936,
      "loss": 5.1096,
      "mean_token_accuracy": 0.1904487982392311,
      "num_tokens": 35709905.0,
      "step": 20580
    },
    {
      "entropy": 5.622174978256226,
      "epoch": 1.6015950204240421,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004743163074055871,
      "loss": 5.0712,
      "mean_token_accuracy": 0.19702855348587037,
      "num_tokens": 35718404.0,
      "step": 20585
    },
    {
      "entropy": 5.677816534042359,
      "epoch": 1.6019840497957596,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004743034466574871,
      "loss": 5.0889,
      "mean_token_accuracy": 0.19194709211587907,
      "num_tokens": 35726475.0,
      "step": 20590
    },
    {
      "entropy": 5.654997396469116,
      "epoch": 1.602373079167477,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004742905828852746,
      "loss": 5.1117,
      "mean_token_accuracy": 0.1939750149846077,
      "num_tokens": 35734819.0,
      "step": 20595
    },
    {
      "entropy": 5.58772087097168,
      "epoch": 1.6027621085391948,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004742777160891447,
      "loss": 4.9955,
      "mean_token_accuracy": 0.200380939245224,
      "num_tokens": 35743098.0,
      "step": 20600
    },
    {
      "entropy": 5.66486496925354,
      "epoch": 1.6031511379109122,
      "grad_norm": 1.125,
      "learning_rate": 0.0004742648462692926,
      "loss": 5.1079,
      "mean_token_accuracy": 0.19390379041433334,
      "num_tokens": 35752502.0,
      "step": 20605
    },
    {
      "entropy": 5.675458478927612,
      "epoch": 1.6035401672826297,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047425197342591363,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20833075493574144,
      "num_tokens": 35761293.0,
      "step": 20610
    },
    {
      "entropy": 5.647287654876709,
      "epoch": 1.6039291966543474,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004742390975592031,
      "loss": 5.1622,
      "mean_token_accuracy": 0.1827400729060173,
      "num_tokens": 35771031.0,
      "step": 20615
    },
    {
      "entropy": 5.675501918792724,
      "epoch": 1.604318226026065,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047422621866935645,
      "loss": 5.0616,
      "mean_token_accuracy": 0.19546449929475784,
      "num_tokens": 35778894.0,
      "step": 20620
    },
    {
      "entropy": 5.735651063919067,
      "epoch": 1.6047072553977826,
      "grad_norm": 1.140625,
      "learning_rate": 0.000474213336756569,
      "loss": 5.1229,
      "mean_token_accuracy": 0.19083805829286576,
      "num_tokens": 35787159.0,
      "step": 20625
    },
    {
      "entropy": 5.60609540939331,
      "epoch": 1.6050962847695,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004742004518210362,
      "loss": 5.04,
      "mean_token_accuracy": 0.2049606204032898,
      "num_tokens": 35795858.0,
      "step": 20630
    },
    {
      "entropy": 5.604734611511231,
      "epoch": 1.6054853141412178,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004741875638629536,
      "loss": 5.0508,
      "mean_token_accuracy": 0.2021583303809166,
      "num_tokens": 35804754.0,
      "step": 20635
    },
    {
      "entropy": 5.707863759994507,
      "epoch": 1.6058743435129352,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004741746728825168,
      "loss": 5.1498,
      "mean_token_accuracy": 0.1861529216170311,
      "num_tokens": 35813622.0,
      "step": 20640
    },
    {
      "entropy": 5.638468790054321,
      "epoch": 1.6062633728846527,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004741617788799213,
      "loss": 5.0428,
      "mean_token_accuracy": 0.19861431419849396,
      "num_tokens": 35821345.0,
      "step": 20645
    },
    {
      "entropy": 5.608671092987061,
      "epoch": 1.6066524022563704,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047414888185536285,
      "loss": 5.0527,
      "mean_token_accuracy": 0.20123450756072997,
      "num_tokens": 35830058.0,
      "step": 20650
    },
    {
      "entropy": 5.674367952346802,
      "epoch": 1.6070414316280879,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004741359818090371,
      "loss": 5.0754,
      "mean_token_accuracy": 0.19666314125061035,
      "num_tokens": 35838951.0,
      "step": 20655
    },
    {
      "entropy": 5.6696014404296875,
      "epoch": 1.6074304609998054,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047412307874113976,
      "loss": 5.039,
      "mean_token_accuracy": 0.2006827637553215,
      "num_tokens": 35847245.0,
      "step": 20660
    },
    {
      "entropy": 5.678802108764648,
      "epoch": 1.607819490371523,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004741101726518667,
      "loss": 5.1442,
      "mean_token_accuracy": 0.1897593155503273,
      "num_tokens": 35856037.0,
      "step": 20665
    },
    {
      "entropy": 5.600672960281372,
      "epoch": 1.6082085197432407,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004740972635414136,
      "loss": 4.9946,
      "mean_token_accuracy": 0.20069389194250106,
      "num_tokens": 35864881.0,
      "step": 20670
    },
    {
      "entropy": 5.734729290008545,
      "epoch": 1.6085975491149582,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004740843514099765,
      "loss": 5.1038,
      "mean_token_accuracy": 0.19034747034311295,
      "num_tokens": 35872941.0,
      "step": 20675
    },
    {
      "entropy": 5.600311899185181,
      "epoch": 1.6089865784866757,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004740714362577512,
      "loss": 5.0818,
      "mean_token_accuracy": 0.19161975085735322,
      "num_tokens": 35882068.0,
      "step": 20680
    },
    {
      "entropy": 5.595686388015747,
      "epoch": 1.6093756078583934,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047405851808493364,
      "loss": 5.0161,
      "mean_token_accuracy": 0.19953440874814987,
      "num_tokens": 35890306.0,
      "step": 20685
    },
    {
      "entropy": 5.6738725185394285,
      "epoch": 1.6097646372301109,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047404559689172006,
      "loss": 5.0854,
      "mean_token_accuracy": 0.20426378399133682,
      "num_tokens": 35899311.0,
      "step": 20690
    },
    {
      "entropy": 5.660208320617675,
      "epoch": 1.6101536666018283,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047403267267830617,
      "loss": 5.139,
      "mean_token_accuracy": 0.1853990823030472,
      "num_tokens": 35908599.0,
      "step": 20695
    },
    {
      "entropy": 5.656005811691284,
      "epoch": 1.610542695973546,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047401974544488844,
      "loss": 5.0885,
      "mean_token_accuracy": 0.194300676882267,
      "num_tokens": 35917318.0,
      "step": 20700
    },
    {
      "entropy": 5.686577224731446,
      "epoch": 1.6109317253452635,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004740068151916628,
      "loss": 5.0472,
      "mean_token_accuracy": 0.20165519118309022,
      "num_tokens": 35926179.0,
      "step": 20705
    },
    {
      "entropy": 5.579198503494263,
      "epoch": 1.611320754716981,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004739938819188255,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21100088357925414,
      "num_tokens": 35934155.0,
      "step": 20710
    },
    {
      "entropy": 5.5677179336547855,
      "epoch": 1.6117097840886987,
      "grad_norm": 1.125,
      "learning_rate": 0.0004739809456265727,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20057405531406403,
      "num_tokens": 35942301.0,
      "step": 20715
    },
    {
      "entropy": 5.572567415237427,
      "epoch": 1.6120988134604164,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004739680063151008,
      "loss": 4.9897,
      "mean_token_accuracy": 0.20024237930774688,
      "num_tokens": 35951005.0,
      "step": 20720
    },
    {
      "entropy": 5.603487777709961,
      "epoch": 1.6124878428321339,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004739550639846061,
      "loss": 5.1029,
      "mean_token_accuracy": 0.19691314101219176,
      "num_tokens": 35959425.0,
      "step": 20725
    },
    {
      "entropy": 5.644115734100342,
      "epoch": 1.6128768722038513,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047394211863528496,
      "loss": 5.0507,
      "mean_token_accuracy": 0.20300205945968627,
      "num_tokens": 35967803.0,
      "step": 20730
    },
    {
      "entropy": 5.65472149848938,
      "epoch": 1.613265901575569,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004739291702673338,
      "loss": 5.1508,
      "mean_token_accuracy": 0.19188271313905716,
      "num_tokens": 35977313.0,
      "step": 20735
    },
    {
      "entropy": 5.754576921463013,
      "epoch": 1.6136549309472865,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004739162188809492,
      "loss": 5.1301,
      "mean_token_accuracy": 0.19150076657533646,
      "num_tokens": 35986404.0,
      "step": 20740
    },
    {
      "entropy": 5.693943977355957,
      "epoch": 1.614043960319004,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047390326447632747,
      "loss": 5.1136,
      "mean_token_accuracy": 0.1960122987627983,
      "num_tokens": 35995865.0,
      "step": 20745
    },
    {
      "entropy": 5.606211805343628,
      "epoch": 1.6144329896907217,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004738903070536654,
      "loss": 5.0028,
      "mean_token_accuracy": 0.2009105697274208,
      "num_tokens": 36005098.0,
      "step": 20750
    },
    {
      "entropy": 5.585506772994995,
      "epoch": 1.6148220190624392,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004738773466131594,
      "loss": 5.0468,
      "mean_token_accuracy": 0.19974856525659562,
      "num_tokens": 36013642.0,
      "step": 20755
    },
    {
      "entropy": 5.663722085952759,
      "epoch": 1.6152110484341566,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004738643831550063,
      "loss": 5.1496,
      "mean_token_accuracy": 0.19051231145858766,
      "num_tokens": 36022565.0,
      "step": 20760
    },
    {
      "entropy": 5.688801622390747,
      "epoch": 1.6156000778058743,
      "grad_norm": 1.125,
      "learning_rate": 0.0004738514166794026,
      "loss": 5.105,
      "mean_token_accuracy": 0.19139495342969895,
      "num_tokens": 36031406.0,
      "step": 20765
    },
    {
      "entropy": 5.672761058807373,
      "epoch": 1.615989107177592,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047383844718654525,
      "loss": 5.1122,
      "mean_token_accuracy": 0.19586308896541596,
      "num_tokens": 36040282.0,
      "step": 20770
    },
    {
      "entropy": 5.731324529647827,
      "epoch": 1.6163781365493095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004738254746766309,
      "loss": 5.1755,
      "mean_token_accuracy": 0.19230180233716965,
      "num_tokens": 36049647.0,
      "step": 20775
    },
    {
      "entropy": 5.612593412399292,
      "epoch": 1.616767165921027,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004738124991498565,
      "loss": 4.997,
      "mean_token_accuracy": 0.20100639760494232,
      "num_tokens": 36057992.0,
      "step": 20780
    },
    {
      "entropy": 5.637743711471558,
      "epoch": 1.6171561952927447,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047379952060641866,
      "loss": 5.0465,
      "mean_token_accuracy": 0.19832704663276673,
      "num_tokens": 36066145.0,
      "step": 20785
    },
    {
      "entropy": 5.610766649246216,
      "epoch": 1.6175452246644622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047378653904651475,
      "loss": 5.0538,
      "mean_token_accuracy": 0.1973472848534584,
      "num_tokens": 36074596.0,
      "step": 20790
    },
    {
      "entropy": 5.680673980712891,
      "epoch": 1.6179342540361796,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047377355447034136,
      "loss": 4.9744,
      "mean_token_accuracy": 0.20498497039079666,
      "num_tokens": 36083499.0,
      "step": 20795
    },
    {
      "entropy": 5.675352239608765,
      "epoch": 1.6183232834078973,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004737605668780958,
      "loss": 5.0828,
      "mean_token_accuracy": 0.19587368816137313,
      "num_tokens": 36092293.0,
      "step": 20800
    },
    {
      "entropy": 5.603817415237427,
      "epoch": 1.618712312779615,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047374757626997494,
      "loss": 5.0672,
      "mean_token_accuracy": 0.19137837141752242,
      "num_tokens": 36100926.0,
      "step": 20805
    },
    {
      "entropy": 5.71808557510376,
      "epoch": 1.6191013421513323,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004737345826461759,
      "loss": 5.117,
      "mean_token_accuracy": 0.1947830930352211,
      "num_tokens": 36109831.0,
      "step": 20810
    },
    {
      "entropy": 5.720207166671753,
      "epoch": 1.61949037152305,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004737215860068959,
      "loss": 5.1604,
      "mean_token_accuracy": 0.19615888744592666,
      "num_tokens": 36118759.0,
      "step": 20815
    },
    {
      "entropy": 5.680451774597168,
      "epoch": 1.6198794008947677,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047370858635233223,
      "loss": 5.1996,
      "mean_token_accuracy": 0.18841734826564788,
      "num_tokens": 36127854.0,
      "step": 20820
    },
    {
      "entropy": 5.649180889129639,
      "epoch": 1.6202684302664851,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047369558368268194,
      "loss": 5.1098,
      "mean_token_accuracy": 0.19646184742450715,
      "num_tokens": 36137686.0,
      "step": 20825
    },
    {
      "entropy": 5.7078650951385494,
      "epoch": 1.6206574596382026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004736825779981424,
      "loss": 5.0943,
      "mean_token_accuracy": 0.19082683622837066,
      "num_tokens": 36146352.0,
      "step": 20830
    },
    {
      "entropy": 5.636092805862427,
      "epoch": 1.6210464890099203,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004736695692989111,
      "loss": 5.0235,
      "mean_token_accuracy": 0.19839919209480286,
      "num_tokens": 36156106.0,
      "step": 20835
    },
    {
      "entropy": 5.604526233673096,
      "epoch": 1.6214355183816378,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004736565575851852,
      "loss": 4.9353,
      "mean_token_accuracy": 0.20191120654344558,
      "num_tokens": 36164169.0,
      "step": 20840
    },
    {
      "entropy": 5.657121229171753,
      "epoch": 1.6218245477533553,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004736435428571622,
      "loss": 5.0836,
      "mean_token_accuracy": 0.19656098037958145,
      "num_tokens": 36173368.0,
      "step": 20845
    },
    {
      "entropy": 5.669279909133911,
      "epoch": 1.622213577125073,
      "grad_norm": 1.125,
      "learning_rate": 0.0004736305251150397,
      "loss": 5.0564,
      "mean_token_accuracy": 0.19547295570373535,
      "num_tokens": 36182405.0,
      "step": 20850
    },
    {
      "entropy": 5.653002834320068,
      "epoch": 1.6226026064967907,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004736175043590151,
      "loss": 5.1261,
      "mean_token_accuracy": 0.19115351438522338,
      "num_tokens": 36191683.0,
      "step": 20855
    },
    {
      "entropy": 5.658615255355835,
      "epoch": 1.622991635868508,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000473604480589286,
      "loss": 5.0142,
      "mean_token_accuracy": 0.1963836133480072,
      "num_tokens": 36200173.0,
      "step": 20860
    },
    {
      "entropy": 5.559234714508056,
      "epoch": 1.6233806652402256,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047359145380605007,
      "loss": 5.0341,
      "mean_token_accuracy": 0.1956578016281128,
      "num_tokens": 36209057.0,
      "step": 20865
    },
    {
      "entropy": 5.611531209945679,
      "epoch": 1.6237696946119433,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004735784240095049,
      "loss": 5.0414,
      "mean_token_accuracy": 0.1999165654182434,
      "num_tokens": 36217645.0,
      "step": 20870
    },
    {
      "entropy": 5.714205408096314,
      "epoch": 1.6241587239836608,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047356539119984813,
      "loss": 5.1148,
      "mean_token_accuracy": 0.19280873239040375,
      "num_tokens": 36226282.0,
      "step": 20875
    },
    {
      "entropy": 5.635988759994507,
      "epoch": 1.6245477533553783,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004735523553772777,
      "loss": 5.1034,
      "mean_token_accuracy": 0.19602324962615966,
      "num_tokens": 36235322.0,
      "step": 20880
    },
    {
      "entropy": 5.6181481838226315,
      "epoch": 1.624936782727096,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004735393165419912,
      "loss": 5.0222,
      "mean_token_accuracy": 0.20810025185346603,
      "num_tokens": 36244212.0,
      "step": 20885
    },
    {
      "entropy": 5.6242388725280765,
      "epoch": 1.6253258120988134,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047352627469418666,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19684476107358934,
      "num_tokens": 36252837.0,
      "step": 20890
    },
    {
      "entropy": 5.70183482170105,
      "epoch": 1.625714841470531,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004735132298340619,
      "loss": 5.0918,
      "mean_token_accuracy": 0.188697150349617,
      "num_tokens": 36261492.0,
      "step": 20895
    },
    {
      "entropy": 5.670249605178833,
      "epoch": 1.6261038708422486,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004735001819618148,
      "loss": 5.0639,
      "mean_token_accuracy": 0.1967158392071724,
      "num_tokens": 36270195.0,
      "step": 20900
    },
    {
      "entropy": 5.6828728199005125,
      "epoch": 1.6264929002139663,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004734871310776434,
      "loss": 5.0073,
      "mean_token_accuracy": 0.21293383240699768,
      "num_tokens": 36278546.0,
      "step": 20905
    },
    {
      "entropy": 5.668334245681763,
      "epoch": 1.6268819295856836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004734740771817457,
      "loss": 5.1361,
      "mean_token_accuracy": 0.18799256533384323,
      "num_tokens": 36287433.0,
      "step": 20910
    },
    {
      "entropy": 5.582912302017212,
      "epoch": 1.6272709589574013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004734610202743198,
      "loss": 4.9889,
      "mean_token_accuracy": 0.18900973051786424,
      "num_tokens": 36296213.0,
      "step": 20915
    },
    {
      "entropy": 5.5907923698425295,
      "epoch": 1.627659988329119,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004734479603555638,
      "loss": 4.9535,
      "mean_token_accuracy": 0.2138149544596672,
      "num_tokens": 36303989.0,
      "step": 20920
    },
    {
      "entropy": 5.713953113555908,
      "epoch": 1.6280490177008364,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047343489742567593,
      "loss": 5.2562,
      "mean_token_accuracy": 0.18459749966859818,
      "num_tokens": 36312939.0,
      "step": 20925
    },
    {
      "entropy": 5.702159643173218,
      "epoch": 1.628438047072554,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047342183148485424,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20019447952508926,
      "num_tokens": 36322130.0,
      "step": 20930
    },
    {
      "entropy": 5.649999284744263,
      "epoch": 1.6288270764442716,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047340876253329706,
      "loss": 4.9951,
      "mean_token_accuracy": 0.19987224489450456,
      "num_tokens": 36330376.0,
      "step": 20935
    },
    {
      "entropy": 5.579406833648681,
      "epoch": 1.629216105815989,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047339569057120275,
      "loss": 5.004,
      "mean_token_accuracy": 0.20319799184799195,
      "num_tokens": 36339614.0,
      "step": 20940
    },
    {
      "entropy": 5.659418344497681,
      "epoch": 1.6296051351877066,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047338261559876966,
      "loss": 5.1035,
      "mean_token_accuracy": 0.19244890213012694,
      "num_tokens": 36349561.0,
      "step": 20945
    },
    {
      "entropy": 5.707289171218872,
      "epoch": 1.6299941645594243,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047336953761619604,
      "loss": 5.1455,
      "mean_token_accuracy": 0.189005284011364,
      "num_tokens": 36358277.0,
      "step": 20950
    },
    {
      "entropy": 5.719502544403076,
      "epoch": 1.630383193931142,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047335645662368046,
      "loss": 5.1167,
      "mean_token_accuracy": 0.19907114356756211,
      "num_tokens": 36366736.0,
      "step": 20955
    },
    {
      "entropy": 5.663247871398926,
      "epoch": 1.6307722233028592,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004733433726214214,
      "loss": 5.0661,
      "mean_token_accuracy": 0.20213507413864135,
      "num_tokens": 36374802.0,
      "step": 20960
    },
    {
      "entropy": 5.646998643875122,
      "epoch": 1.631161252674577,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047333028560961733,
      "loss": 5.0854,
      "mean_token_accuracy": 0.19363243728876114,
      "num_tokens": 36384195.0,
      "step": 20965
    },
    {
      "entropy": 5.631627368927002,
      "epoch": 1.6315502820462946,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047331719558846687,
      "loss": 5.0988,
      "mean_token_accuracy": 0.19158527851104737,
      "num_tokens": 36393001.0,
      "step": 20970
    },
    {
      "entropy": 5.679379653930664,
      "epoch": 1.631939311418012,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047330410255816863,
      "loss": 5.0502,
      "mean_token_accuracy": 0.18936222791671753,
      "num_tokens": 36401674.0,
      "step": 20975
    },
    {
      "entropy": 5.610768032073975,
      "epoch": 1.6323283407897295,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047329100651892123,
      "loss": 5.0255,
      "mean_token_accuracy": 0.20752585530281067,
      "num_tokens": 36409817.0,
      "step": 20980
    },
    {
      "entropy": 5.630240964889526,
      "epoch": 1.6327173701614472,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004732779074709234,
      "loss": 4.9905,
      "mean_token_accuracy": 0.20000280290842057,
      "num_tokens": 36417541.0,
      "step": 20985
    },
    {
      "entropy": 5.6237305164337155,
      "epoch": 1.6331063995331647,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000473264805414374,
      "loss": 5.1055,
      "mean_token_accuracy": 0.18834645599126815,
      "num_tokens": 36426613.0,
      "step": 20990
    },
    {
      "entropy": 5.651272487640381,
      "epoch": 1.6334954289048822,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047325170034947167,
      "loss": 5.0966,
      "mean_token_accuracy": 0.19901169538497926,
      "num_tokens": 36436052.0,
      "step": 20995
    },
    {
      "entropy": 5.612300539016724,
      "epoch": 1.6338844582766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047323859227641537,
      "loss": 5.0048,
      "mean_token_accuracy": 0.20065058767795563,
      "num_tokens": 36444386.0,
      "step": 21000
    },
    {
      "epoch": 1.6338844582766,
      "eval_entropy": 5.425801253846566,
      "eval_loss": 5.124657154083252,
      "eval_mean_token_accuracy": 0.20376745658757253,
      "eval_num_tokens": 36444386.0,
      "eval_runtime": 21.8112,
      "eval_samples_per_second": 1905.353,
      "eval_steps_per_second": 238.181,
      "step": 21000
    },
    {
      "entropy": 5.685236263275146,
      "epoch": 1.6342734876483176,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047322548119540397,
      "loss": 5.0627,
      "mean_token_accuracy": 0.1946759581565857,
      "num_tokens": 36452759.0,
      "step": 21005
    },
    {
      "entropy": 5.650661134719849,
      "epoch": 1.634662517020035,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047321236710663636,
      "loss": 5.024,
      "mean_token_accuracy": 0.20091758370399476,
      "num_tokens": 36460581.0,
      "step": 21010
    },
    {
      "entropy": 5.629543972015381,
      "epoch": 1.6350515463917525,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047319925001031165,
      "loss": 5.0451,
      "mean_token_accuracy": 0.20526855140924455,
      "num_tokens": 36468995.0,
      "step": 21015
    },
    {
      "entropy": 5.718970918655396,
      "epoch": 1.6354405757634702,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004731861299066287,
      "loss": 5.1196,
      "mean_token_accuracy": 0.19818554818630219,
      "num_tokens": 36477901.0,
      "step": 21020
    },
    {
      "entropy": 5.68755407333374,
      "epoch": 1.6358296051351877,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004731730067957867,
      "loss": 5.1119,
      "mean_token_accuracy": 0.19964181035757064,
      "num_tokens": 36486500.0,
      "step": 21025
    },
    {
      "entropy": 5.626431035995483,
      "epoch": 1.6362186345069052,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047315988067798476,
      "loss": 5.053,
      "mean_token_accuracy": 0.19768139570951462,
      "num_tokens": 36495247.0,
      "step": 21030
    },
    {
      "entropy": 5.626111078262329,
      "epoch": 1.6366076638786229,
      "grad_norm": 1.21875,
      "learning_rate": 0.000473146751553422,
      "loss": 5.0111,
      "mean_token_accuracy": 0.206063412129879,
      "num_tokens": 36503633.0,
      "step": 21035
    },
    {
      "entropy": 5.65632381439209,
      "epoch": 1.6369966932503404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004731336194222978,
      "loss": 5.0788,
      "mean_token_accuracy": 0.1980946645140648,
      "num_tokens": 36512138.0,
      "step": 21040
    },
    {
      "entropy": 5.6815392017364506,
      "epoch": 1.6373857226220578,
      "grad_norm": 1.125,
      "learning_rate": 0.00047312048428481114,
      "loss": 5.061,
      "mean_token_accuracy": 0.20013161301612853,
      "num_tokens": 36521029.0,
      "step": 21045
    },
    {
      "entropy": 5.608483123779297,
      "epoch": 1.6377747519937755,
      "grad_norm": 1.125,
      "learning_rate": 0.00047310734614116153,
      "loss": 4.9886,
      "mean_token_accuracy": 0.20001065731048584,
      "num_tokens": 36530658.0,
      "step": 21050
    },
    {
      "entropy": 5.618550252914429,
      "epoch": 1.6381637813654932,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004730942049915483,
      "loss": 4.9898,
      "mean_token_accuracy": 0.1970773383975029,
      "num_tokens": 36539037.0,
      "step": 21055
    },
    {
      "entropy": 5.646327257156372,
      "epoch": 1.6385528107372107,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004730810608361707,
      "loss": 5.0826,
      "mean_token_accuracy": 0.1998617023229599,
      "num_tokens": 36547863.0,
      "step": 21060
    },
    {
      "entropy": 5.684886646270752,
      "epoch": 1.6389418401089282,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047306791367522833,
      "loss": 5.1045,
      "mean_token_accuracy": 0.19899885803461076,
      "num_tokens": 36555896.0,
      "step": 21065
    },
    {
      "entropy": 5.595557403564453,
      "epoch": 1.6393308694806459,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004730547635089206,
      "loss": 5.0367,
      "mean_token_accuracy": 0.20867674499750138,
      "num_tokens": 36564088.0,
      "step": 21070
    },
    {
      "entropy": 5.656909227371216,
      "epoch": 1.6397198988523634,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047304161033744714,
      "loss": 4.9992,
      "mean_token_accuracy": 0.2010756775736809,
      "num_tokens": 36572490.0,
      "step": 21075
    },
    {
      "entropy": 5.686183261871338,
      "epoch": 1.6401089282240808,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004730284541610075,
      "loss": 5.0703,
      "mean_token_accuracy": 0.1874392732977867,
      "num_tokens": 36580834.0,
      "step": 21080
    },
    {
      "entropy": 5.637238359451294,
      "epoch": 1.6404979575957985,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004730152949798012,
      "loss": 5.0346,
      "mean_token_accuracy": 0.20410232841968537,
      "num_tokens": 36589420.0,
      "step": 21085
    },
    {
      "entropy": 5.579383277893067,
      "epoch": 1.640886986967516,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047300213279402787,
      "loss": 5.0534,
      "mean_token_accuracy": 0.19768901467323302,
      "num_tokens": 36598419.0,
      "step": 21090
    },
    {
      "entropy": 5.614743328094482,
      "epoch": 1.6412760163392335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047298896760388745,
      "loss": 5.0776,
      "mean_token_accuracy": 0.19908344149589538,
      "num_tokens": 36607367.0,
      "step": 21095
    },
    {
      "entropy": 5.709177398681641,
      "epoch": 1.6416650457109512,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004729757994095796,
      "loss": 5.1625,
      "mean_token_accuracy": 0.19179748445749284,
      "num_tokens": 36616197.0,
      "step": 21100
    },
    {
      "entropy": 5.733490371704102,
      "epoch": 1.6420540750826689,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047296262821130405,
      "loss": 5.1889,
      "mean_token_accuracy": 0.1905401736497879,
      "num_tokens": 36625452.0,
      "step": 21105
    },
    {
      "entropy": 5.6510701179504395,
      "epoch": 1.6424431044543863,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047294945400926065,
      "loss": 5.1163,
      "mean_token_accuracy": 0.19272997081279755,
      "num_tokens": 36634356.0,
      "step": 21110
    },
    {
      "entropy": 5.66380820274353,
      "epoch": 1.6428321338261038,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004729362768036494,
      "loss": 5.0771,
      "mean_token_accuracy": 0.19462316185235978,
      "num_tokens": 36642548.0,
      "step": 21115
    },
    {
      "entropy": 5.640546560287476,
      "epoch": 1.6432211631978215,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047292309659467024,
      "loss": 4.9804,
      "mean_token_accuracy": 0.20825111716985703,
      "num_tokens": 36650548.0,
      "step": 21120
    },
    {
      "entropy": 5.604476976394653,
      "epoch": 1.643610192569539,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004729099133825231,
      "loss": 5.0202,
      "mean_token_accuracy": 0.19761257469654084,
      "num_tokens": 36659322.0,
      "step": 21125
    },
    {
      "entropy": 5.616048955917359,
      "epoch": 1.6439992219412565,
      "grad_norm": 1.109375,
      "learning_rate": 0.000472896727167408,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20036645233631134,
      "num_tokens": 36667699.0,
      "step": 21130
    },
    {
      "entropy": 5.715201282501221,
      "epoch": 1.6443882513129742,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047288353794952505,
      "loss": 5.13,
      "mean_token_accuracy": 0.1948148638010025,
      "num_tokens": 36677842.0,
      "step": 21135
    },
    {
      "entropy": 5.608692312240601,
      "epoch": 1.6447772806846916,
      "grad_norm": 1.125,
      "learning_rate": 0.0004728703457290744,
      "loss": 4.9294,
      "mean_token_accuracy": 0.21180818676948548,
      "num_tokens": 36686488.0,
      "step": 21140
    },
    {
      "entropy": 5.652051067352295,
      "epoch": 1.6451663100564091,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004728571505062562,
      "loss": 5.1088,
      "mean_token_accuracy": 0.19624157398939132,
      "num_tokens": 36694955.0,
      "step": 21145
    },
    {
      "entropy": 5.557367324829102,
      "epoch": 1.6455553394281268,
      "grad_norm": 1.125,
      "learning_rate": 0.0004728439522812707,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2034481644630432,
      "num_tokens": 36703481.0,
      "step": 21150
    },
    {
      "entropy": 5.686665010452271,
      "epoch": 1.6459443687998445,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047283075105431806,
      "loss": 5.1937,
      "mean_token_accuracy": 0.19371117651462555,
      "num_tokens": 36711812.0,
      "step": 21155
    },
    {
      "entropy": 5.668038415908813,
      "epoch": 1.646333398171562,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004728175468255987,
      "loss": 5.0522,
      "mean_token_accuracy": 0.20208872854709625,
      "num_tokens": 36720496.0,
      "step": 21160
    },
    {
      "entropy": 5.6225639343261715,
      "epoch": 1.6467224275432795,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047280433959531287,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20318156331777573,
      "num_tokens": 36728959.0,
      "step": 21165
    },
    {
      "entropy": 5.585181140899659,
      "epoch": 1.6471114569149972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000472791129363661,
      "loss": 5.0361,
      "mean_token_accuracy": 0.19560608118772507,
      "num_tokens": 36737492.0,
      "step": 21170
    },
    {
      "entropy": 5.640459728240967,
      "epoch": 1.6475004862867146,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004727779161308436,
      "loss": 4.9933,
      "mean_token_accuracy": 0.2004745975136757,
      "num_tokens": 36745852.0,
      "step": 21175
    },
    {
      "entropy": 5.680353593826294,
      "epoch": 1.6478895156584321,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004727646998970612,
      "loss": 5.0735,
      "mean_token_accuracy": 0.19790082424879074,
      "num_tokens": 36753897.0,
      "step": 21180
    },
    {
      "entropy": 5.634651851654053,
      "epoch": 1.6482785450301498,
      "grad_norm": 1.125,
      "learning_rate": 0.00047275148066251417,
      "loss": 5.0246,
      "mean_token_accuracy": 0.19867342561483384,
      "num_tokens": 36762737.0,
      "step": 21185
    },
    {
      "entropy": 5.704489469528198,
      "epoch": 1.6486675744018673,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004727382584274032,
      "loss": 5.1112,
      "mean_token_accuracy": 0.19380357414484023,
      "num_tokens": 36770742.0,
      "step": 21190
    },
    {
      "entropy": 5.739238977432251,
      "epoch": 1.6490566037735848,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047272503319192887,
      "loss": 5.1698,
      "mean_token_accuracy": 0.1844904363155365,
      "num_tokens": 36779420.0,
      "step": 21195
    },
    {
      "entropy": 5.5963489532470705,
      "epoch": 1.6494456331453025,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004727118049562919,
      "loss": 5.0016,
      "mean_token_accuracy": 0.19979230910539628,
      "num_tokens": 36787888.0,
      "step": 21200
    },
    {
      "entropy": 5.57241997718811,
      "epoch": 1.6498346625170202,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000472698573720693,
      "loss": 4.9328,
      "mean_token_accuracy": 0.2079233855009079,
      "num_tokens": 36797131.0,
      "step": 21205
    },
    {
      "entropy": 5.616825723648072,
      "epoch": 1.6502236918887376,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004726853394853329,
      "loss": 5.1078,
      "mean_token_accuracy": 0.19077519625425338,
      "num_tokens": 36805429.0,
      "step": 21210
    },
    {
      "entropy": 5.645532846450806,
      "epoch": 1.650612721260455,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047267210225041234,
      "loss": 5.1497,
      "mean_token_accuracy": 0.18943463861942292,
      "num_tokens": 36814204.0,
      "step": 21215
    },
    {
      "entropy": 5.757275390625,
      "epoch": 1.6510017506321728,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047265886201613234,
      "loss": 5.1125,
      "mean_token_accuracy": 0.19360610097646713,
      "num_tokens": 36823229.0,
      "step": 21220
    },
    {
      "entropy": 5.729987573623657,
      "epoch": 1.6513907800038903,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047264561878269365,
      "loss": 5.18,
      "mean_token_accuracy": 0.18842648416757585,
      "num_tokens": 36832252.0,
      "step": 21225
    },
    {
      "entropy": 5.614921617507934,
      "epoch": 1.6517798093756078,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004726323725502973,
      "loss": 5.0472,
      "mean_token_accuracy": 0.19894642531871795,
      "num_tokens": 36841456.0,
      "step": 21230
    },
    {
      "entropy": 5.6571146011352536,
      "epoch": 1.6521688387473255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004726191233191443,
      "loss": 5.0228,
      "mean_token_accuracy": 0.2042941778898239,
      "num_tokens": 36849717.0,
      "step": 21235
    },
    {
      "entropy": 5.650038576126098,
      "epoch": 1.6525578681190431,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004726058710894357,
      "loss": 5.1076,
      "mean_token_accuracy": 0.1954307049512863,
      "num_tokens": 36858222.0,
      "step": 21240
    },
    {
      "entropy": 5.592621326446533,
      "epoch": 1.6529468974907604,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004725926158613724,
      "loss": 5.0881,
      "mean_token_accuracy": 0.1922290101647377,
      "num_tokens": 36866899.0,
      "step": 21245
    },
    {
      "entropy": 5.658283805847168,
      "epoch": 1.653335926862478,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004725793576351557,
      "loss": 5.0491,
      "mean_token_accuracy": 0.20789932161569596,
      "num_tokens": 36876124.0,
      "step": 21250
    },
    {
      "entropy": 5.691045713424683,
      "epoch": 1.6537249562341958,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004725660964109867,
      "loss": 5.1068,
      "mean_token_accuracy": 0.19472858160734177,
      "num_tokens": 36884689.0,
      "step": 21255
    },
    {
      "entropy": 5.6766516208648685,
      "epoch": 1.6541139856059133,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047255283218906673,
      "loss": 5.0806,
      "mean_token_accuracy": 0.1931871712207794,
      "num_tokens": 36893073.0,
      "step": 21260
    },
    {
      "entropy": 5.670859432220459,
      "epoch": 1.6545030149776307,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000472539564969597,
      "loss": 5.0887,
      "mean_token_accuracy": 0.19451346546411513,
      "num_tokens": 36902428.0,
      "step": 21265
    },
    {
      "entropy": 5.69565224647522,
      "epoch": 1.6548920443493484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047252629475277863,
      "loss": 5.1098,
      "mean_token_accuracy": 0.19970912784337996,
      "num_tokens": 36910297.0,
      "step": 21270
    },
    {
      "entropy": 5.641442012786865,
      "epoch": 1.655281073721066,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004725130215388132,
      "loss": 5.0534,
      "mean_token_accuracy": 0.19424191415309905,
      "num_tokens": 36919115.0,
      "step": 21275
    },
    {
      "entropy": 5.6860127449035645,
      "epoch": 1.6556701030927834,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004724997453279021,
      "loss": 5.0811,
      "mean_token_accuracy": 0.19655924141407013,
      "num_tokens": 36928419.0,
      "step": 21280
    },
    {
      "entropy": 5.6646819591522215,
      "epoch": 1.656059132464501,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004724864661202467,
      "loss": 5.0378,
      "mean_token_accuracy": 0.20006120949983597,
      "num_tokens": 36936511.0,
      "step": 21285
    },
    {
      "entropy": 5.672817659378052,
      "epoch": 1.6564481618362188,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047247318391604846,
      "loss": 5.1035,
      "mean_token_accuracy": 0.1965818926692009,
      "num_tokens": 36945316.0,
      "step": 21290
    },
    {
      "entropy": 5.634977912902832,
      "epoch": 1.656837191207936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047245989871550897,
      "loss": 5.1275,
      "mean_token_accuracy": 0.18893318325281144,
      "num_tokens": 36954140.0,
      "step": 21295
    },
    {
      "entropy": 5.628003311157227,
      "epoch": 1.6572262205796537,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004724466105188299,
      "loss": 5.0776,
      "mean_token_accuracy": 0.1945530191063881,
      "num_tokens": 36962852.0,
      "step": 21300
    },
    {
      "entropy": 5.7002171039581295,
      "epoch": 1.6576152499513714,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047243331932621263,
      "loss": 5.0906,
      "mean_token_accuracy": 0.19977089613676072,
      "num_tokens": 36971617.0,
      "step": 21305
    },
    {
      "entropy": 5.641374588012695,
      "epoch": 1.658004279323089,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004724200251378591,
      "loss": 5.1099,
      "mean_token_accuracy": 0.19899785071611403,
      "num_tokens": 36981078.0,
      "step": 21310
    },
    {
      "entropy": 5.644744968414306,
      "epoch": 1.6583933086948064,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004724067279539709,
      "loss": 5.0131,
      "mean_token_accuracy": 0.19661311954259872,
      "num_tokens": 36989860.0,
      "step": 21315
    },
    {
      "entropy": 5.652134370803833,
      "epoch": 1.658782338066524,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047239342777474975,
      "loss": 5.0463,
      "mean_token_accuracy": 0.2002684146165848,
      "num_tokens": 36998037.0,
      "step": 21320
    },
    {
      "entropy": 5.707924842834473,
      "epoch": 1.6591713674382416,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047238012460039765,
      "loss": 5.0732,
      "mean_token_accuracy": 0.19763096421957016,
      "num_tokens": 37006375.0,
      "step": 21325
    },
    {
      "entropy": 5.6436220645904545,
      "epoch": 1.659560396809959,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004723668184311162,
      "loss": 5.141,
      "mean_token_accuracy": 0.18975942879915236,
      "num_tokens": 37014650.0,
      "step": 21330
    },
    {
      "entropy": 5.637217855453491,
      "epoch": 1.6599494261816767,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004723535092671074,
      "loss": 5.1053,
      "mean_token_accuracy": 0.18916617184877396,
      "num_tokens": 37023147.0,
      "step": 21335
    },
    {
      "entropy": 5.749104070663452,
      "epoch": 1.6603384555533944,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047234019710857335,
      "loss": 5.1037,
      "mean_token_accuracy": 0.1955679774284363,
      "num_tokens": 37031461.0,
      "step": 21340
    },
    {
      "entropy": 5.726814270019531,
      "epoch": 1.6607274849251117,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004723268819557158,
      "loss": 5.1414,
      "mean_token_accuracy": 0.19410899430513381,
      "num_tokens": 37039924.0,
      "step": 21345
    },
    {
      "entropy": 5.630940341949463,
      "epoch": 1.6611165142968294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047231356380873687,
      "loss": 5.038,
      "mean_token_accuracy": 0.20254645347595215,
      "num_tokens": 37048631.0,
      "step": 21350
    },
    {
      "entropy": 5.637073755264282,
      "epoch": 1.661505543668547,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004723002426678388,
      "loss": 5.0486,
      "mean_token_accuracy": 0.2034415289759636,
      "num_tokens": 37057018.0,
      "step": 21355
    },
    {
      "entropy": 5.666430187225342,
      "epoch": 1.6618945730402646,
      "grad_norm": 1.25,
      "learning_rate": 0.0004722869185332235,
      "loss": 5.083,
      "mean_token_accuracy": 0.19698277115821838,
      "num_tokens": 37065762.0,
      "step": 21360
    },
    {
      "entropy": 5.638964700698852,
      "epoch": 1.662283602411982,
      "grad_norm": 1.125,
      "learning_rate": 0.00047227359140509324,
      "loss": 5.0778,
      "mean_token_accuracy": 0.19166834354400636,
      "num_tokens": 37073836.0,
      "step": 21365
    },
    {
      "entropy": 5.654311800003052,
      "epoch": 1.6626726317836997,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004722602612836501,
      "loss": 5.127,
      "mean_token_accuracy": 0.19017945677042009,
      "num_tokens": 37082788.0,
      "step": 21370
    },
    {
      "entropy": 5.716874313354492,
      "epoch": 1.6630616611554172,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047224692816909665,
      "loss": 5.1524,
      "mean_token_accuracy": 0.18403398394584655,
      "num_tokens": 37091228.0,
      "step": 21375
    },
    {
      "entropy": 5.668335390090943,
      "epoch": 1.6634506905271347,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004722335920616349,
      "loss": 5.0562,
      "mean_token_accuracy": 0.19864943325519563,
      "num_tokens": 37099819.0,
      "step": 21380
    },
    {
      "entropy": 5.638710927963257,
      "epoch": 1.6638397198988524,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047222025296146745,
      "loss": 5.0927,
      "mean_token_accuracy": 0.1881011739373207,
      "num_tokens": 37108129.0,
      "step": 21385
    },
    {
      "entropy": 5.591794013977051,
      "epoch": 1.66422874927057,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047220691086879643,
      "loss": 4.994,
      "mean_token_accuracy": 0.2084198921918869,
      "num_tokens": 37116093.0,
      "step": 21390
    },
    {
      "entropy": 5.622945880889892,
      "epoch": 1.6646177786422873,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004721935657838245,
      "loss": 5.0458,
      "mean_token_accuracy": 0.1923132747411728,
      "num_tokens": 37124098.0,
      "step": 21395
    },
    {
      "entropy": 5.719037961959839,
      "epoch": 1.665006808014005,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000472180217706754,
      "loss": 5.1232,
      "mean_token_accuracy": 0.18765225410461425,
      "num_tokens": 37132608.0,
      "step": 21400
    },
    {
      "entropy": 5.701768589019776,
      "epoch": 1.6653958373857227,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004721668666377876,
      "loss": 5.127,
      "mean_token_accuracy": 0.19519110321998595,
      "num_tokens": 37141365.0,
      "step": 21405
    },
    {
      "entropy": 5.592532682418823,
      "epoch": 1.6657848667574402,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047215351257712783,
      "loss": 5.0091,
      "mean_token_accuracy": 0.20509780645370485,
      "num_tokens": 37150146.0,
      "step": 21410
    },
    {
      "entropy": 5.720004463195801,
      "epoch": 1.6661738961291577,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047214015552497735,
      "loss": 5.1887,
      "mean_token_accuracy": 0.1878465473651886,
      "num_tokens": 37158887.0,
      "step": 21415
    },
    {
      "entropy": 5.709566068649292,
      "epoch": 1.6665629255008754,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047212679548153867,
      "loss": 5.0286,
      "mean_token_accuracy": 0.1976467952132225,
      "num_tokens": 37167902.0,
      "step": 21420
    },
    {
      "entropy": 5.7032022953033445,
      "epoch": 1.6669519548725928,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047211343244701475,
      "loss": 5.1605,
      "mean_token_accuracy": 0.19156619608402253,
      "num_tokens": 37175853.0,
      "step": 21425
    },
    {
      "entropy": 5.619184303283691,
      "epoch": 1.6673409842443103,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004721000664216081,
      "loss": 5.0397,
      "mean_token_accuracy": 0.19947168380022048,
      "num_tokens": 37184036.0,
      "step": 21430
    },
    {
      "entropy": 5.685922431945801,
      "epoch": 1.667730013616028,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004720866974055219,
      "loss": 5.1022,
      "mean_token_accuracy": 0.19761382937431335,
      "num_tokens": 37193277.0,
      "step": 21435
    },
    {
      "entropy": 5.701192426681518,
      "epoch": 1.6681190429877457,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004720733253989585,
      "loss": 5.0644,
      "mean_token_accuracy": 0.2003820925951004,
      "num_tokens": 37201925.0,
      "step": 21440
    },
    {
      "entropy": 5.6847645282745365,
      "epoch": 1.6685080723594632,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047205995040212114,
      "loss": 5.1555,
      "mean_token_accuracy": 0.1892498627305031,
      "num_tokens": 37210552.0,
      "step": 21445
    },
    {
      "entropy": 5.6613525390625,
      "epoch": 1.6688971017311807,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004720465724152127,
      "loss": 5.0784,
      "mean_token_accuracy": 0.1969710037112236,
      "num_tokens": 37219001.0,
      "step": 21450
    },
    {
      "entropy": 5.64767861366272,
      "epoch": 1.6692861311028984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047203319143843613,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20369212180376053,
      "num_tokens": 37227635.0,
      "step": 21455
    },
    {
      "entropy": 5.615681934356689,
      "epoch": 1.6696751604746158,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004720198074719946,
      "loss": 5.0596,
      "mean_token_accuracy": 0.19911978095769883,
      "num_tokens": 37236872.0,
      "step": 21460
    },
    {
      "entropy": 5.706751585006714,
      "epoch": 1.6700641898463333,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047200642051609094,
      "loss": 5.1304,
      "mean_token_accuracy": 0.1907180964946747,
      "num_tokens": 37245026.0,
      "step": 21465
    },
    {
      "entropy": 5.680355262756348,
      "epoch": 1.670453219218051,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004719930305709285,
      "loss": 5.0714,
      "mean_token_accuracy": 0.19570827186107637,
      "num_tokens": 37253295.0,
      "step": 21470
    },
    {
      "entropy": 5.651104497909546,
      "epoch": 1.6708422485897685,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047197963763671033,
      "loss": 5.0467,
      "mean_token_accuracy": 0.2063677117228508,
      "num_tokens": 37262785.0,
      "step": 21475
    },
    {
      "entropy": 5.642996978759766,
      "epoch": 1.671231277961486,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004719662417136397,
      "loss": 5.1256,
      "mean_token_accuracy": 0.2006133183836937,
      "num_tokens": 37271937.0,
      "step": 21480
    },
    {
      "entropy": 5.6505739212036135,
      "epoch": 1.6716203073332037,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047195284280191987,
      "loss": 5.0212,
      "mean_token_accuracy": 0.20324522256851196,
      "num_tokens": 37280861.0,
      "step": 21485
    },
    {
      "entropy": 5.671285104751587,
      "epoch": 1.6720093367049214,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004719394409017541,
      "loss": 5.0643,
      "mean_token_accuracy": 0.19922929108142853,
      "num_tokens": 37289235.0,
      "step": 21490
    },
    {
      "entropy": 5.64128360748291,
      "epoch": 1.6723983660766388,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047192603601334585,
      "loss": 5.0703,
      "mean_token_accuracy": 0.19294965714216233,
      "num_tokens": 37297778.0,
      "step": 21495
    },
    {
      "entropy": 5.645361471176147,
      "epoch": 1.6727873954483563,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047191262813689835,
      "loss": 5.0826,
      "mean_token_accuracy": 0.19001952558755875,
      "num_tokens": 37306960.0,
      "step": 21500
    },
    {
      "entropy": 5.659552669525146,
      "epoch": 1.673176424820074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004718992172726152,
      "loss": 5.1064,
      "mean_token_accuracy": 0.19469638764858246,
      "num_tokens": 37316050.0,
      "step": 21505
    },
    {
      "entropy": 5.586802673339844,
      "epoch": 1.6735654541917915,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047188580342069983,
      "loss": 5.0591,
      "mean_token_accuracy": 0.19923404604196548,
      "num_tokens": 37325058.0,
      "step": 21510
    },
    {
      "entropy": 5.712506818771362,
      "epoch": 1.673954483563509,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004718723865813557,
      "loss": 5.1592,
      "mean_token_accuracy": 0.18860733211040498,
      "num_tokens": 37333512.0,
      "step": 21515
    },
    {
      "entropy": 5.677211666107178,
      "epoch": 1.6743435129352267,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004718589667547865,
      "loss": 5.0552,
      "mean_token_accuracy": 0.20085858553647995,
      "num_tokens": 37341695.0,
      "step": 21520
    },
    {
      "entropy": 5.693460464477539,
      "epoch": 1.6747325423069441,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004718455439411958,
      "loss": 5.2198,
      "mean_token_accuracy": 0.18530402928590775,
      "num_tokens": 37350917.0,
      "step": 21525
    },
    {
      "entropy": 5.735363435745239,
      "epoch": 1.6751215716786616,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004718321181407873,
      "loss": 5.1697,
      "mean_token_accuracy": 0.1849642351269722,
      "num_tokens": 37360206.0,
      "step": 21530
    },
    {
      "entropy": 5.744312524795532,
      "epoch": 1.6755106010503793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004718186893537647,
      "loss": 5.1339,
      "mean_token_accuracy": 0.19340480268001556,
      "num_tokens": 37369378.0,
      "step": 21535
    },
    {
      "entropy": 5.660828876495361,
      "epoch": 1.675899630422097,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004718052575803318,
      "loss": 5.0524,
      "mean_token_accuracy": 0.20309800207614898,
      "num_tokens": 37377744.0,
      "step": 21540
    },
    {
      "entropy": 5.667059135437012,
      "epoch": 1.6762886597938145,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004717918228206923,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19197254925966262,
      "num_tokens": 37387097.0,
      "step": 21545
    },
    {
      "entropy": 5.713493490219117,
      "epoch": 1.676677689165532,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004717783850750501,
      "loss": 5.1459,
      "mean_token_accuracy": 0.1895543411374092,
      "num_tokens": 37395551.0,
      "step": 21550
    },
    {
      "entropy": 5.630736780166626,
      "epoch": 1.6770667185372496,
      "grad_norm": 1.125,
      "learning_rate": 0.0004717649443436091,
      "loss": 4.9281,
      "mean_token_accuracy": 0.20800315886735915,
      "num_tokens": 37403876.0,
      "step": 21555
    },
    {
      "entropy": 5.658391809463501,
      "epoch": 1.6774557479089671,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047175150062657336,
      "loss": 5.0505,
      "mean_token_accuracy": 0.1969800129532814,
      "num_tokens": 37412658.0,
      "step": 21560
    },
    {
      "entropy": 5.656177282333374,
      "epoch": 1.6778447772806846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047173805392414664,
      "loss": 5.0293,
      "mean_token_accuracy": 0.19987383335828782,
      "num_tokens": 37420790.0,
      "step": 21565
    },
    {
      "entropy": 5.639275074005127,
      "epoch": 1.6782338066524023,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004717246042365332,
      "loss": 5.098,
      "mean_token_accuracy": 0.19355754554271698,
      "num_tokens": 37428866.0,
      "step": 21570
    },
    {
      "entropy": 5.6372967720031735,
      "epoch": 1.6786228360241198,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047171115156393695,
      "loss": 5.032,
      "mean_token_accuracy": 0.2024747237563133,
      "num_tokens": 37437636.0,
      "step": 21575
    },
    {
      "entropy": 5.6486818313598635,
      "epoch": 1.6790118653958372,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047169769590656207,
      "loss": 5.0879,
      "mean_token_accuracy": 0.19749243259429933,
      "num_tokens": 37446455.0,
      "step": 21580
    },
    {
      "entropy": 5.641884803771973,
      "epoch": 1.679400894767555,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047168423726461273,
      "loss": 5.1577,
      "mean_token_accuracy": 0.1887023076415062,
      "num_tokens": 37454997.0,
      "step": 21585
    },
    {
      "entropy": 5.637234306335449,
      "epoch": 1.6797899241392726,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004716707756382931,
      "loss": 4.9943,
      "mean_token_accuracy": 0.20316977947950363,
      "num_tokens": 37463701.0,
      "step": 21590
    },
    {
      "entropy": 5.703650093078613,
      "epoch": 1.6801789535109901,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047165731102780757,
      "loss": 5.1235,
      "mean_token_accuracy": 0.18962299823760986,
      "num_tokens": 37472460.0,
      "step": 21595
    },
    {
      "entropy": 5.677802467346192,
      "epoch": 1.6805679828827076,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047164384343336037,
      "loss": 5.0337,
      "mean_token_accuracy": 0.19802031815052032,
      "num_tokens": 37481114.0,
      "step": 21600
    },
    {
      "entropy": 5.6594819068908695,
      "epoch": 1.6809570122544253,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047163037285515583,
      "loss": 5.1129,
      "mean_token_accuracy": 0.19172435253858566,
      "num_tokens": 37489593.0,
      "step": 21605
    },
    {
      "entropy": 5.726067781448364,
      "epoch": 1.6813460416261428,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004716168992933982,
      "loss": 5.1599,
      "mean_token_accuracy": 0.19118998050689698,
      "num_tokens": 37498044.0,
      "step": 21610
    },
    {
      "entropy": 5.741387605667114,
      "epoch": 1.6817350709978602,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047160342274829234,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19408333748579026,
      "num_tokens": 37507339.0,
      "step": 21615
    },
    {
      "entropy": 5.608314943313599,
      "epoch": 1.682124100369578,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047158994322004223,
      "loss": 4.9657,
      "mean_token_accuracy": 0.20372917652130126,
      "num_tokens": 37515919.0,
      "step": 21620
    },
    {
      "entropy": 5.628144121170044,
      "epoch": 1.6825131297412956,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004715764607088527,
      "loss": 4.9686,
      "mean_token_accuracy": 0.20786874294281005,
      "num_tokens": 37524990.0,
      "step": 21625
    },
    {
      "entropy": 5.628019189834594,
      "epoch": 1.6829021591130129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004715629752149283,
      "loss": 5.1045,
      "mean_token_accuracy": 0.19200923144817353,
      "num_tokens": 37534125.0,
      "step": 21630
    },
    {
      "entropy": 5.648524332046509,
      "epoch": 1.6832911884847306,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047154948673847356,
      "loss": 5.0221,
      "mean_token_accuracy": 0.1976946175098419,
      "num_tokens": 37541715.0,
      "step": 21635
    },
    {
      "entropy": 5.7356569290161135,
      "epoch": 1.6836802178564483,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004715359952796933,
      "loss": 5.156,
      "mean_token_accuracy": 0.20350012481212615,
      "num_tokens": 37550097.0,
      "step": 21640
    },
    {
      "entropy": 5.585136222839355,
      "epoch": 1.6840692472281658,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047152250083879203,
      "loss": 5.0188,
      "mean_token_accuracy": 0.19679685980081557,
      "num_tokens": 37558950.0,
      "step": 21645
    },
    {
      "entropy": 5.675404357910156,
      "epoch": 1.6844582765998832,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047150900341597464,
      "loss": 5.034,
      "mean_token_accuracy": 0.20156317204236984,
      "num_tokens": 37567737.0,
      "step": 21650
    },
    {
      "entropy": 5.700136137008667,
      "epoch": 1.684847305971601,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004714955030114459,
      "loss": 5.0773,
      "mean_token_accuracy": 0.19584817737340926,
      "num_tokens": 37576649.0,
      "step": 21655
    },
    {
      "entropy": 5.655124568939209,
      "epoch": 1.6852363353433184,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047148199962541065,
      "loss": 5.0558,
      "mean_token_accuracy": 0.20388414412736894,
      "num_tokens": 37584455.0,
      "step": 21660
    },
    {
      "entropy": 5.659012794494629,
      "epoch": 1.6856253647150359,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004714684932580738,
      "loss": 5.1272,
      "mean_token_accuracy": 0.18838134557008743,
      "num_tokens": 37593653.0,
      "step": 21665
    },
    {
      "entropy": 5.693011713027954,
      "epoch": 1.6860143940867536,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004714549839096403,
      "loss": 5.1093,
      "mean_token_accuracy": 0.1929284319281578,
      "num_tokens": 37603212.0,
      "step": 21670
    },
    {
      "entropy": 5.686216735839844,
      "epoch": 1.6864034234584713,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047144147158031507,
      "loss": 5.0331,
      "mean_token_accuracy": 0.19574832767248154,
      "num_tokens": 37611871.0,
      "step": 21675
    },
    {
      "entropy": 5.7182615280151365,
      "epoch": 1.6867924528301885,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004714279562703032,
      "loss": 5.2188,
      "mean_token_accuracy": 0.18723901510238647,
      "num_tokens": 37621410.0,
      "step": 21680
    },
    {
      "entropy": 5.664160490036011,
      "epoch": 1.6871814822019062,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004714144379798098,
      "loss": 5.0297,
      "mean_token_accuracy": 0.19567696303129195,
      "num_tokens": 37629525.0,
      "step": 21685
    },
    {
      "entropy": 5.723068618774414,
      "epoch": 1.687570511573624,
      "grad_norm": 1.125,
      "learning_rate": 0.00047140091670903986,
      "loss": 5.0854,
      "mean_token_accuracy": 0.19774255752563477,
      "num_tokens": 37638183.0,
      "step": 21690
    },
    {
      "entropy": 5.686976766586303,
      "epoch": 1.6879595409453414,
      "grad_norm": 1.125,
      "learning_rate": 0.00047138739245819857,
      "loss": 5.0693,
      "mean_token_accuracy": 0.19803808480501175,
      "num_tokens": 37646884.0,
      "step": 21695
    },
    {
      "entropy": 5.738629531860352,
      "epoch": 1.6883485703170589,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047137386522749124,
      "loss": 5.1196,
      "mean_token_accuracy": 0.18730611950159073,
      "num_tokens": 37655482.0,
      "step": 21700
    },
    {
      "entropy": 5.612303066253662,
      "epoch": 1.6887375996887766,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004713603350171231,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2021462708711624,
      "num_tokens": 37664056.0,
      "step": 21705
    },
    {
      "entropy": 5.669025564193726,
      "epoch": 1.689126629060494,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047134680182729926,
      "loss": 5.2382,
      "mean_token_accuracy": 0.19384651780128478,
      "num_tokens": 37673198.0,
      "step": 21710
    },
    {
      "entropy": 5.664942216873169,
      "epoch": 1.6895156584322115,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004713332656582254,
      "loss": 5.0697,
      "mean_token_accuracy": 0.19532050788402558,
      "num_tokens": 37682660.0,
      "step": 21715
    },
    {
      "entropy": 5.633370590209961,
      "epoch": 1.6899046878039292,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004713197265101066,
      "loss": 5.0389,
      "mean_token_accuracy": 0.202335362136364,
      "num_tokens": 37691232.0,
      "step": 21720
    },
    {
      "entropy": 5.61824893951416,
      "epoch": 1.690293717175647,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047130618438314856,
      "loss": 5.0005,
      "mean_token_accuracy": 0.20015470683574677,
      "num_tokens": 37699734.0,
      "step": 21725
    },
    {
      "entropy": 5.700785541534424,
      "epoch": 1.6906827465473642,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004712926392775565,
      "loss": 5.0839,
      "mean_token_accuracy": 0.1999967008829117,
      "num_tokens": 37708529.0,
      "step": 21730
    },
    {
      "entropy": 5.63576717376709,
      "epoch": 1.6910717759190819,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004712790911935362,
      "loss": 4.9562,
      "mean_token_accuracy": 0.2098645582795143,
      "num_tokens": 37717197.0,
      "step": 21735
    },
    {
      "entropy": 5.618352890014648,
      "epoch": 1.6914608052907996,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000471265540131293,
      "loss": 5.0982,
      "mean_token_accuracy": 0.19188545197248458,
      "num_tokens": 37726276.0,
      "step": 21740
    },
    {
      "entropy": 5.72824912071228,
      "epoch": 1.691849834662517,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047125198609103267,
      "loss": 5.137,
      "mean_token_accuracy": 0.19199802875518798,
      "num_tokens": 37735016.0,
      "step": 21745
    },
    {
      "entropy": 5.729806423187256,
      "epoch": 1.6922388640342345,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004712384290729607,
      "loss": 5.104,
      "mean_token_accuracy": 0.20036841332912445,
      "num_tokens": 37744332.0,
      "step": 21750
    },
    {
      "entropy": 5.7336780548095705,
      "epoch": 1.6926278934059522,
      "grad_norm": 1.171875,
      "learning_rate": 0.000471224869077283,
      "loss": 5.164,
      "mean_token_accuracy": 0.1810240551829338,
      "num_tokens": 37752805.0,
      "step": 21755
    },
    {
      "entropy": 5.708140134811401,
      "epoch": 1.6930169227776697,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047121130610420527,
      "loss": 5.0816,
      "mean_token_accuracy": 0.19440166354179383,
      "num_tokens": 37760979.0,
      "step": 21760
    },
    {
      "entropy": 5.588563919067383,
      "epoch": 1.6934059521493872,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047119774015393323,
      "loss": 5.0224,
      "mean_token_accuracy": 0.19733347296714782,
      "num_tokens": 37769096.0,
      "step": 21765
    },
    {
      "entropy": 5.656375885009766,
      "epoch": 1.6937949815211049,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047118417122667285,
      "loss": 5.0283,
      "mean_token_accuracy": 0.19535085558891296,
      "num_tokens": 37776998.0,
      "step": 21770
    },
    {
      "entropy": 5.7147716045379635,
      "epoch": 1.6941840108928226,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004711705993226298,
      "loss": 5.1449,
      "mean_token_accuracy": 0.19495135098695754,
      "num_tokens": 37785547.0,
      "step": 21775
    },
    {
      "entropy": 5.667989873886109,
      "epoch": 1.6945730402645398,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004711570244420102,
      "loss": 5.0958,
      "mean_token_accuracy": 0.1913059949874878,
      "num_tokens": 37793659.0,
      "step": 21780
    },
    {
      "entropy": 5.6489544868469235,
      "epoch": 1.6949620696362575,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004711434465850199,
      "loss": 5.0083,
      "mean_token_accuracy": 0.19947948008775712,
      "num_tokens": 37802282.0,
      "step": 21785
    },
    {
      "entropy": 5.69829363822937,
      "epoch": 1.6953510990079752,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004711298657518651,
      "loss": 5.0728,
      "mean_token_accuracy": 0.19906374365091323,
      "num_tokens": 37810692.0,
      "step": 21790
    },
    {
      "entropy": 5.606528997421265,
      "epoch": 1.6957401283796927,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004711162819427518,
      "loss": 5.0343,
      "mean_token_accuracy": 0.20089835524559022,
      "num_tokens": 37819264.0,
      "step": 21795
    },
    {
      "entropy": 5.662407445907593,
      "epoch": 1.6961291577514102,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000471102695157886,
      "loss": 5.0801,
      "mean_token_accuracy": 0.20163782984018325,
      "num_tokens": 37827650.0,
      "step": 21800
    },
    {
      "entropy": 5.565398263931274,
      "epoch": 1.6965181871231279,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004710891053974739,
      "loss": 5.0159,
      "mean_token_accuracy": 0.2013427272439003,
      "num_tokens": 37835683.0,
      "step": 21805
    },
    {
      "entropy": 5.652868175506592,
      "epoch": 1.6969072164948453,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004710755126617217,
      "loss": 5.0309,
      "mean_token_accuracy": 0.1954900309443474,
      "num_tokens": 37844334.0,
      "step": 21810
    },
    {
      "entropy": 5.617059993743896,
      "epoch": 1.6972962458665628,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004710619169508358,
      "loss": 5.0088,
      "mean_token_accuracy": 0.19692396521568298,
      "num_tokens": 37853609.0,
      "step": 21815
    },
    {
      "entropy": 5.665457820892334,
      "epoch": 1.6976852752382805,
      "grad_norm": 1.125,
      "learning_rate": 0.0004710483182650223,
      "loss": 5.0612,
      "mean_token_accuracy": 0.19669616967439651,
      "num_tokens": 37862817.0,
      "step": 21820
    },
    {
      "entropy": 5.589251852035522,
      "epoch": 1.6980743046099982,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047103471660448767,
      "loss": 5.0005,
      "mean_token_accuracy": 0.2045843094587326,
      "num_tokens": 37871630.0,
      "step": 21825
    },
    {
      "entropy": 5.552453231811524,
      "epoch": 1.6984633339817157,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047102111196943813,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20275750160217285,
      "num_tokens": 37881098.0,
      "step": 21830
    },
    {
      "entropy": 5.659877634048462,
      "epoch": 1.6988523633534331,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047100750436008035,
      "loss": 5.024,
      "mean_token_accuracy": 0.2032433718442917,
      "num_tokens": 37889760.0,
      "step": 21835
    },
    {
      "entropy": 5.740854406356812,
      "epoch": 1.6992413927251508,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047099389377662054,
      "loss": 5.132,
      "mean_token_accuracy": 0.19285643249750137,
      "num_tokens": 37898417.0,
      "step": 21840
    },
    {
      "entropy": 5.639759302139282,
      "epoch": 1.6996304220968683,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004709802802192654,
      "loss": 5.0801,
      "mean_token_accuracy": 0.2006111890077591,
      "num_tokens": 37907117.0,
      "step": 21845
    },
    {
      "entropy": 5.614550828933716,
      "epoch": 1.7000194514685858,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047096666368822153,
      "loss": 5.0148,
      "mean_token_accuracy": 0.1985253646969795,
      "num_tokens": 37915781.0,
      "step": 21850
    },
    {
      "entropy": 5.650159883499145,
      "epoch": 1.7004084808403035,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047095304418369536,
      "loss": 5.1034,
      "mean_token_accuracy": 0.19175428450107573,
      "num_tokens": 37924770.0,
      "step": 21855
    },
    {
      "entropy": 5.657611083984375,
      "epoch": 1.700797510212021,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004709394217058936,
      "loss": 5.1124,
      "mean_token_accuracy": 0.19348921179771422,
      "num_tokens": 37933512.0,
      "step": 21860
    },
    {
      "entropy": 5.612676429748535,
      "epoch": 1.7011865395837384,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004709257962550231,
      "loss": 4.935,
      "mean_token_accuracy": 0.20104921907186507,
      "num_tokens": 37942439.0,
      "step": 21865
    },
    {
      "entropy": 5.672794818878174,
      "epoch": 1.7015755689554561,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047091216783129035,
      "loss": 5.1241,
      "mean_token_accuracy": 0.19222945719957352,
      "num_tokens": 37951388.0,
      "step": 21870
    },
    {
      "entropy": 5.689729785919189,
      "epoch": 1.7019645983271738,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004708985364349024,
      "loss": 5.116,
      "mean_token_accuracy": 0.19365573674440384,
      "num_tokens": 37960372.0,
      "step": 21875
    },
    {
      "entropy": 5.71558632850647,
      "epoch": 1.7023536276988913,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047088490206606586,
      "loss": 5.2004,
      "mean_token_accuracy": 0.17831635028123854,
      "num_tokens": 37969472.0,
      "step": 21880
    },
    {
      "entropy": 5.616215753555298,
      "epoch": 1.7027426570706088,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047087126472498785,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2066580057144165,
      "num_tokens": 37977422.0,
      "step": 21885
    },
    {
      "entropy": 5.596018838882446,
      "epoch": 1.7031316864423265,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000470857624411875,
      "loss": 4.9122,
      "mean_token_accuracy": 0.20602083951234818,
      "num_tokens": 37985861.0,
      "step": 21890
    },
    {
      "entropy": 5.690075063705445,
      "epoch": 1.703520715814044,
      "grad_norm": 1.25,
      "learning_rate": 0.00047084398112693456,
      "loss": 5.0423,
      "mean_token_accuracy": 0.20235230773687363,
      "num_tokens": 37994535.0,
      "step": 21895
    },
    {
      "entropy": 5.6552125930786135,
      "epoch": 1.7039097451857614,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004708303348703734,
      "loss": 5.0822,
      "mean_token_accuracy": 0.19286412000656128,
      "num_tokens": 38003351.0,
      "step": 21900
    },
    {
      "entropy": 5.678764867782593,
      "epoch": 1.7042987745574791,
      "grad_norm": 1.125,
      "learning_rate": 0.0004708166856423986,
      "loss": 5.0784,
      "mean_token_accuracy": 0.19291504323482514,
      "num_tokens": 38012210.0,
      "step": 21905
    },
    {
      "entropy": 5.696633291244507,
      "epoch": 1.7046878039291966,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047080303344321727,
      "loss": 5.0751,
      "mean_token_accuracy": 0.19998956620693206,
      "num_tokens": 38020548.0,
      "step": 21910
    },
    {
      "entropy": 5.687703037261963,
      "epoch": 1.705076833300914,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047078937827303653,
      "loss": 5.0388,
      "mean_token_accuracy": 0.19409146904945374,
      "num_tokens": 38029451.0,
      "step": 21915
    },
    {
      "entropy": 5.656308364868164,
      "epoch": 1.7054658626726318,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004707757201320636,
      "loss": 5.0799,
      "mean_token_accuracy": 0.1945350632071495,
      "num_tokens": 38037821.0,
      "step": 21920
    },
    {
      "entropy": 5.615873718261719,
      "epoch": 1.7058548920443495,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047076205902050577,
      "loss": 5.1307,
      "mean_token_accuracy": 0.19154444336891174,
      "num_tokens": 38046188.0,
      "step": 21925
    },
    {
      "entropy": 5.679753208160401,
      "epoch": 1.706243921416067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047074839493857024,
      "loss": 5.1302,
      "mean_token_accuracy": 0.1952909231185913,
      "num_tokens": 38054821.0,
      "step": 21930
    },
    {
      "entropy": 5.672739744186401,
      "epoch": 1.7066329507877844,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004707347278864644,
      "loss": 5.0532,
      "mean_token_accuracy": 0.19533757120370865,
      "num_tokens": 38063057.0,
      "step": 21935
    },
    {
      "entropy": 5.706013154983521,
      "epoch": 1.7070219801595021,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047072105786439563,
      "loss": 5.2027,
      "mean_token_accuracy": 0.1914782240986824,
      "num_tokens": 38071694.0,
      "step": 21940
    },
    {
      "entropy": 5.691050481796265,
      "epoch": 1.7074110095312196,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047070738487257137,
      "loss": 5.0219,
      "mean_token_accuracy": 0.1982439175248146,
      "num_tokens": 38080965.0,
      "step": 21945
    },
    {
      "entropy": 5.667789697647095,
      "epoch": 1.707800038902937,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047069370891119895,
      "loss": 5.0521,
      "mean_token_accuracy": 0.19038837552070617,
      "num_tokens": 38089786.0,
      "step": 21950
    },
    {
      "entropy": 5.6412077903747555,
      "epoch": 1.7081890682746548,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000470680029980486,
      "loss": 5.0622,
      "mean_token_accuracy": 0.19882871955633163,
      "num_tokens": 38097994.0,
      "step": 21955
    },
    {
      "entropy": 5.690302467346191,
      "epoch": 1.7085780976463723,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047066634808064006,
      "loss": 5.1178,
      "mean_token_accuracy": 0.19154055416584015,
      "num_tokens": 38106910.0,
      "step": 21960
    },
    {
      "entropy": 5.646644115447998,
      "epoch": 1.7089671270180897,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047065266321186873,
      "loss": 5.0332,
      "mean_token_accuracy": 0.2043434962630272,
      "num_tokens": 38115162.0,
      "step": 21965
    },
    {
      "entropy": 5.705600547790527,
      "epoch": 1.7093561563898074,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004706389753743797,
      "loss": 5.083,
      "mean_token_accuracy": 0.20416899919509887,
      "num_tokens": 38123776.0,
      "step": 21970
    },
    {
      "entropy": 5.705074787139893,
      "epoch": 1.7097451857615251,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004706252845683805,
      "loss": 5.1659,
      "mean_token_accuracy": 0.19011609256267548,
      "num_tokens": 38132591.0,
      "step": 21975
    },
    {
      "entropy": 5.6537495136260985,
      "epoch": 1.7101342151332426,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047061159079407903,
      "loss": 5.027,
      "mean_token_accuracy": 0.2028370201587677,
      "num_tokens": 38141092.0,
      "step": 21980
    },
    {
      "entropy": 5.633610391616822,
      "epoch": 1.71052324450496,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000470597894051683,
      "loss": 5.0078,
      "mean_token_accuracy": 0.1986740291118622,
      "num_tokens": 38149315.0,
      "step": 21985
    },
    {
      "entropy": 5.604947948455811,
      "epoch": 1.7109122738766778,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004705841943414003,
      "loss": 4.9733,
      "mean_token_accuracy": 0.19915590435266495,
      "num_tokens": 38158208.0,
      "step": 21990
    },
    {
      "entropy": 5.74879937171936,
      "epoch": 1.7113013032483952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047057049166343876,
      "loss": 5.126,
      "mean_token_accuracy": 0.19257869869470595,
      "num_tokens": 38166339.0,
      "step": 21995
    },
    {
      "entropy": 5.717361402511597,
      "epoch": 1.7116903326201127,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047055678601800623,
      "loss": 5.084,
      "mean_token_accuracy": 0.19625718891620636,
      "num_tokens": 38174813.0,
      "step": 22000
    },
    {
      "entropy": 5.656217098236084,
      "epoch": 1.7120793619918304,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047054307740531076,
      "loss": 4.9596,
      "mean_token_accuracy": 0.20155562162399293,
      "num_tokens": 38183440.0,
      "step": 22005
    },
    {
      "entropy": 5.596694850921631,
      "epoch": 1.712468391363548,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047052936582556035,
      "loss": 5.0202,
      "mean_token_accuracy": 0.19960058182477952,
      "num_tokens": 38192510.0,
      "step": 22010
    },
    {
      "entropy": 5.650761985778809,
      "epoch": 1.7128574207352654,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047051565127896307,
      "loss": 5.0144,
      "mean_token_accuracy": 0.20671189874410628,
      "num_tokens": 38200854.0,
      "step": 22015
    },
    {
      "entropy": 5.6916844844818115,
      "epoch": 1.713246450106983,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047050193376572686,
      "loss": 5.0717,
      "mean_token_accuracy": 0.1918088510632515,
      "num_tokens": 38209632.0,
      "step": 22020
    },
    {
      "entropy": 5.615683746337891,
      "epoch": 1.7136354794787008,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047048821328606,
      "loss": 4.9976,
      "mean_token_accuracy": 0.20116380751132965,
      "num_tokens": 38218657.0,
      "step": 22025
    },
    {
      "entropy": 5.7197511196136475,
      "epoch": 1.7140245088504182,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004704744898401708,
      "loss": 5.0643,
      "mean_token_accuracy": 0.19772231131792067,
      "num_tokens": 38227507.0,
      "step": 22030
    },
    {
      "entropy": 5.6676534652709964,
      "epoch": 1.7144135382221357,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004704607634282672,
      "loss": 5.0377,
      "mean_token_accuracy": 0.1997448906302452,
      "num_tokens": 38236822.0,
      "step": 22035
    },
    {
      "entropy": 5.563470411300659,
      "epoch": 1.7148025675938534,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047044703405055765,
      "loss": 4.9497,
      "mean_token_accuracy": 0.2040942892432213,
      "num_tokens": 38245144.0,
      "step": 22040
    },
    {
      "entropy": 5.621599626541138,
      "epoch": 1.7151915969655709,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047043330170725046,
      "loss": 5.0516,
      "mean_token_accuracy": 0.1953385666012764,
      "num_tokens": 38254363.0,
      "step": 22045
    },
    {
      "entropy": 5.634972953796387,
      "epoch": 1.7155806263372884,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047041956639855406,
      "loss": 5.051,
      "mean_token_accuracy": 0.19931643605232238,
      "num_tokens": 38263117.0,
      "step": 22050
    },
    {
      "entropy": 5.632564926147461,
      "epoch": 1.715969655709006,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004704058281246766,
      "loss": 5.0341,
      "mean_token_accuracy": 0.1971964493393898,
      "num_tokens": 38271217.0,
      "step": 22055
    },
    {
      "entropy": 5.716578531265259,
      "epoch": 1.7163586850807238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004703920868858268,
      "loss": 5.1623,
      "mean_token_accuracy": 0.19212710112333298,
      "num_tokens": 38279789.0,
      "step": 22060
    },
    {
      "entropy": 5.626923418045044,
      "epoch": 1.716747714452441,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047037834268221315,
      "loss": 4.9823,
      "mean_token_accuracy": 0.20163310319185257,
      "num_tokens": 38288845.0,
      "step": 22065
    },
    {
      "entropy": 5.656175184249878,
      "epoch": 1.7171367438241587,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004703645955140441,
      "loss": 5.0493,
      "mean_token_accuracy": 0.1967256933450699,
      "num_tokens": 38297237.0,
      "step": 22070
    },
    {
      "entropy": 5.745203351974487,
      "epoch": 1.7175257731958764,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047035084538152815,
      "loss": 5.1585,
      "mean_token_accuracy": 0.18956682980060577,
      "num_tokens": 38306582.0,
      "step": 22075
    },
    {
      "entropy": 5.658037710189819,
      "epoch": 1.7179148025675939,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004703370922848742,
      "loss": 4.9744,
      "mean_token_accuracy": 0.21153550297021867,
      "num_tokens": 38315406.0,
      "step": 22080
    },
    {
      "entropy": 5.62490701675415,
      "epoch": 1.7183038319393114,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047032333622429074,
      "loss": 4.9744,
      "mean_token_accuracy": 0.20097845196723937,
      "num_tokens": 38324271.0,
      "step": 22085
    },
    {
      "entropy": 5.664039182662964,
      "epoch": 1.718692861311029,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047030957719998656,
      "loss": 5.0756,
      "mean_token_accuracy": 0.19401168525218965,
      "num_tokens": 38332831.0,
      "step": 22090
    },
    {
      "entropy": 5.632407712936401,
      "epoch": 1.7190818906827465,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004702958152121704,
      "loss": 4.9909,
      "mean_token_accuracy": 0.20227778404951097,
      "num_tokens": 38341263.0,
      "step": 22095
    },
    {
      "entropy": 5.648492622375488,
      "epoch": 1.719470920054464,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004702820502610511,
      "loss": 5.1847,
      "mean_token_accuracy": 0.18876439929008484,
      "num_tokens": 38349890.0,
      "step": 22100
    },
    {
      "entropy": 5.610504007339477,
      "epoch": 1.7198599494261817,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004702682823468375,
      "loss": 4.9655,
      "mean_token_accuracy": 0.20201875567436217,
      "num_tokens": 38358071.0,
      "step": 22105
    },
    {
      "entropy": 5.711025524139404,
      "epoch": 1.7202489787978994,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047025451146973844,
      "loss": 5.1283,
      "mean_token_accuracy": 0.19127406924962997,
      "num_tokens": 38366859.0,
      "step": 22110
    },
    {
      "entropy": 5.770093822479248,
      "epoch": 1.7206380081696167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047024073762996305,
      "loss": 5.2786,
      "mean_token_accuracy": 0.180878247320652,
      "num_tokens": 38375897.0,
      "step": 22115
    },
    {
      "entropy": 5.6647388458251955,
      "epoch": 1.7210270375413343,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004702269608277202,
      "loss": 5.0572,
      "mean_token_accuracy": 0.20210765302181244,
      "num_tokens": 38384700.0,
      "step": 22120
    },
    {
      "entropy": 5.698626279830933,
      "epoch": 1.721416066913052,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004702131810632189,
      "loss": 5.1123,
      "mean_token_accuracy": 0.19869651645421982,
      "num_tokens": 38393212.0,
      "step": 22125
    },
    {
      "entropy": 5.620493793487549,
      "epoch": 1.7218050962847695,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047019939833666837,
      "loss": 5.0233,
      "mean_token_accuracy": 0.20437860637903213,
      "num_tokens": 38401572.0,
      "step": 22130
    },
    {
      "entropy": 5.618623733520508,
      "epoch": 1.722194125656487,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004701856126482776,
      "loss": 5.0555,
      "mean_token_accuracy": 0.19549354761838914,
      "num_tokens": 38410669.0,
      "step": 22135
    },
    {
      "entropy": 5.661244440078735,
      "epoch": 1.7225831550282047,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004701718239982559,
      "loss": 5.0658,
      "mean_token_accuracy": 0.1931559458374977,
      "num_tokens": 38419038.0,
      "step": 22140
    },
    {
      "entropy": 5.6170848369598385,
      "epoch": 1.7229721843999222,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047015803238681234,
      "loss": 5.0027,
      "mean_token_accuracy": 0.2055936947464943,
      "num_tokens": 38427314.0,
      "step": 22145
    },
    {
      "entropy": 5.608303117752075,
      "epoch": 1.7233612137716396,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004701442378141563,
      "loss": 4.9995,
      "mean_token_accuracy": 0.19678326845169067,
      "num_tokens": 38435373.0,
      "step": 22150
    },
    {
      "entropy": 5.613267707824707,
      "epoch": 1.7237502431433573,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000470130440280497,
      "loss": 5.0228,
      "mean_token_accuracy": 0.20060820132493973,
      "num_tokens": 38444448.0,
      "step": 22155
    },
    {
      "entropy": 5.6791112422943115,
      "epoch": 1.724139272515075,
      "grad_norm": 1.125,
      "learning_rate": 0.0004701166397860439,
      "loss": 4.9721,
      "mean_token_accuracy": 0.20241123139858247,
      "num_tokens": 38453576.0,
      "step": 22160
    },
    {
      "entropy": 5.686106252670288,
      "epoch": 1.7245283018867923,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004701028363310064,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20553840398788453,
      "num_tokens": 38462848.0,
      "step": 22165
    },
    {
      "entropy": 5.6319732666015625,
      "epoch": 1.72491733125851,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047008902991559385,
      "loss": 5.0021,
      "mean_token_accuracy": 0.20384716987609863,
      "num_tokens": 38470953.0,
      "step": 22170
    },
    {
      "entropy": 5.687946414947509,
      "epoch": 1.7253063606302277,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004700752205400158,
      "loss": 5.1323,
      "mean_token_accuracy": 0.19119604974985122,
      "num_tokens": 38479835.0,
      "step": 22175
    },
    {
      "entropy": 5.679265928268433,
      "epoch": 1.7256953900019452,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047006140820448174,
      "loss": 5.0515,
      "mean_token_accuracy": 0.19489602446556092,
      "num_tokens": 38488342.0,
      "step": 22180
    },
    {
      "entropy": 5.664249897003174,
      "epoch": 1.7260844193736626,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047004759290920136,
      "loss": 5.0676,
      "mean_token_accuracy": 0.1976734146475792,
      "num_tokens": 38496716.0,
      "step": 22185
    },
    {
      "entropy": 5.658633279800415,
      "epoch": 1.7264734487453803,
      "grad_norm": 1.125,
      "learning_rate": 0.0004700337746543841,
      "loss": 5.0953,
      "mean_token_accuracy": 0.19553611129522325,
      "num_tokens": 38506615.0,
      "step": 22190
    },
    {
      "entropy": 5.609146404266357,
      "epoch": 1.7268624781170978,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004700199534402398,
      "loss": 4.9031,
      "mean_token_accuracy": 0.20573283582925797,
      "num_tokens": 38515936.0,
      "step": 22195
    },
    {
      "entropy": 5.622857475280762,
      "epoch": 1.7272515074888153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004700061292669781,
      "loss": 5.0328,
      "mean_token_accuracy": 0.19693392515182495,
      "num_tokens": 38524829.0,
      "step": 22200
    },
    {
      "entropy": 5.681061124801635,
      "epoch": 1.727640536860533,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004699923021348088,
      "loss": 5.1863,
      "mean_token_accuracy": 0.18945951163768768,
      "num_tokens": 38533686.0,
      "step": 22205
    },
    {
      "entropy": 5.704560089111328,
      "epoch": 1.7280295662322507,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046997847204394166,
      "loss": 5.0564,
      "mean_token_accuracy": 0.19742194563150406,
      "num_tokens": 38542449.0,
      "step": 22210
    },
    {
      "entropy": 5.679417324066162,
      "epoch": 1.728418595603968,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046996463899458653,
      "loss": 5.0275,
      "mean_token_accuracy": 0.20108308494091034,
      "num_tokens": 38550888.0,
      "step": 22215
    },
    {
      "entropy": 5.5916379451751705,
      "epoch": 1.7288076249756856,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004699508029869533,
      "loss": 5.0696,
      "mean_token_accuracy": 0.19759927093982696,
      "num_tokens": 38559366.0,
      "step": 22220
    },
    {
      "entropy": 5.692618656158447,
      "epoch": 1.7291966543474033,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046993696402125205,
      "loss": 5.0914,
      "mean_token_accuracy": 0.20067006051540376,
      "num_tokens": 38567080.0,
      "step": 22225
    },
    {
      "entropy": 5.657074213027954,
      "epoch": 1.7295856837191208,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004699231220976925,
      "loss": 5.0689,
      "mean_token_accuracy": 0.2045196369290352,
      "num_tokens": 38575520.0,
      "step": 22230
    },
    {
      "entropy": 5.7178449630737305,
      "epoch": 1.7299747130908383,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004699092772164849,
      "loss": 5.1606,
      "mean_token_accuracy": 0.1892016500234604,
      "num_tokens": 38583974.0,
      "step": 22235
    },
    {
      "entropy": 5.6052460193634035,
      "epoch": 1.730363742462556,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004698954293778392,
      "loss": 5.0696,
      "mean_token_accuracy": 0.19698223024606704,
      "num_tokens": 38592677.0,
      "step": 22240
    },
    {
      "entropy": 5.621514415740966,
      "epoch": 1.7307527718342735,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046988157858196555,
      "loss": 4.9104,
      "mean_token_accuracy": 0.201981620490551,
      "num_tokens": 38601342.0,
      "step": 22245
    },
    {
      "entropy": 5.712474775314331,
      "epoch": 1.731141801205991,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046986772482907426,
      "loss": 5.1634,
      "mean_token_accuracy": 0.18830789625644684,
      "num_tokens": 38610895.0,
      "step": 22250
    },
    {
      "entropy": 5.668631267547608,
      "epoch": 1.7315308305777086,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004698538681193754,
      "loss": 5.1195,
      "mean_token_accuracy": 0.1954927295446396,
      "num_tokens": 38619059.0,
      "step": 22255
    },
    {
      "entropy": 5.679808139801025,
      "epoch": 1.7319198599494263,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046984000845307907,
      "loss": 5.0718,
      "mean_token_accuracy": 0.1958228886127472,
      "num_tokens": 38628084.0,
      "step": 22260
    },
    {
      "entropy": 5.6559711456298825,
      "epoch": 1.7323088893211438,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046982614583039584,
      "loss": 5.0758,
      "mean_token_accuracy": 0.19177769422531127,
      "num_tokens": 38636205.0,
      "step": 22265
    },
    {
      "entropy": 5.6353994846344,
      "epoch": 1.7326979186928613,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004698122802515359,
      "loss": 5.0316,
      "mean_token_accuracy": 0.19626931250095367,
      "num_tokens": 38644583.0,
      "step": 22270
    },
    {
      "entropy": 5.642191600799561,
      "epoch": 1.733086948064579,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004697984117167097,
      "loss": 5.0163,
      "mean_token_accuracy": 0.20118715465068818,
      "num_tokens": 38653517.0,
      "step": 22275
    },
    {
      "entropy": 5.6460182666778564,
      "epoch": 1.7334759774362964,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046978454022612767,
      "loss": 5.0765,
      "mean_token_accuracy": 0.1928579553961754,
      "num_tokens": 38661926.0,
      "step": 22280
    },
    {
      "entropy": 5.753850650787354,
      "epoch": 1.733865006808014,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046977066578000025,
      "loss": 5.1241,
      "mean_token_accuracy": 0.19831566363573075,
      "num_tokens": 38671567.0,
      "step": 22285
    },
    {
      "entropy": 5.679229211807251,
      "epoch": 1.7342540361797316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046975678837853806,
      "loss": 5.0494,
      "mean_token_accuracy": 0.20398346036672593,
      "num_tokens": 38680439.0,
      "step": 22290
    },
    {
      "entropy": 5.642826557159424,
      "epoch": 1.734643065551449,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046974290802195156,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19669349044561385,
      "num_tokens": 38689168.0,
      "step": 22295
    },
    {
      "entropy": 5.689180564880371,
      "epoch": 1.7350320949231666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004697290247104513,
      "loss": 5.0323,
      "mean_token_accuracy": 0.19757416248321533,
      "num_tokens": 38697592.0,
      "step": 22300
    },
    {
      "entropy": 5.649882888793945,
      "epoch": 1.7354211242948843,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046971513844424814,
      "loss": 5.0798,
      "mean_token_accuracy": 0.19310355484485625,
      "num_tokens": 38705561.0,
      "step": 22305
    },
    {
      "entropy": 5.614341115951538,
      "epoch": 1.735810153666602,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046970124922355265,
      "loss": 5.0392,
      "mean_token_accuracy": 0.2010719284415245,
      "num_tokens": 38714694.0,
      "step": 22310
    },
    {
      "entropy": 5.679735565185547,
      "epoch": 1.7361991830383194,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004696873570485756,
      "loss": 5.0254,
      "mean_token_accuracy": 0.20323633998632432,
      "num_tokens": 38722797.0,
      "step": 22315
    },
    {
      "entropy": 5.682542705535889,
      "epoch": 1.736588212410037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004696734619195278,
      "loss": 5.0992,
      "mean_token_accuracy": 0.1997784748673439,
      "num_tokens": 38732076.0,
      "step": 22320
    },
    {
      "entropy": 5.699263191223144,
      "epoch": 1.7369772417817546,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046965956383662,
      "loss": 5.1075,
      "mean_token_accuracy": 0.20008494555950165,
      "num_tokens": 38740992.0,
      "step": 22325
    },
    {
      "entropy": 5.6915748596191404,
      "epoch": 1.737366271153472,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004696456628000632,
      "loss": 5.0761,
      "mean_token_accuracy": 0.19764744490385056,
      "num_tokens": 38749266.0,
      "step": 22330
    },
    {
      "entropy": 5.677693510055542,
      "epoch": 1.7377553005251896,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046963175881006825,
      "loss": 5.1521,
      "mean_token_accuracy": 0.19179669469594957,
      "num_tokens": 38757983.0,
      "step": 22335
    },
    {
      "entropy": 5.6361168384552,
      "epoch": 1.7381443298969073,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004696178518668462,
      "loss": 4.9896,
      "mean_token_accuracy": 0.20907741338014602,
      "num_tokens": 38766258.0,
      "step": 22340
    },
    {
      "entropy": 5.68007230758667,
      "epoch": 1.7385333592686247,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046960394197060804,
      "loss": 5.1095,
      "mean_token_accuracy": 0.19478160440921782,
      "num_tokens": 38775105.0,
      "step": 22345
    },
    {
      "entropy": 5.695080280303955,
      "epoch": 1.7389223886403422,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046959002912156473,
      "loss": 5.1197,
      "mean_token_accuracy": 0.1957464560866356,
      "num_tokens": 38783415.0,
      "step": 22350
    },
    {
      "entropy": 5.6615049839019775,
      "epoch": 1.73931141801206,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004695761133199275,
      "loss": 5.1281,
      "mean_token_accuracy": 0.1939013972878456,
      "num_tokens": 38791703.0,
      "step": 22355
    },
    {
      "entropy": 5.666936206817627,
      "epoch": 1.7397004473837776,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004695621945659074,
      "loss": 5.1025,
      "mean_token_accuracy": 0.19301100075244904,
      "num_tokens": 38800823.0,
      "step": 22360
    },
    {
      "entropy": 5.674952459335327,
      "epoch": 1.740089476755495,
      "grad_norm": 1.0,
      "learning_rate": 0.0004695482728597157,
      "loss": 5.0987,
      "mean_token_accuracy": 0.18985410928726196,
      "num_tokens": 38810607.0,
      "step": 22365
    },
    {
      "entropy": 5.652855157852173,
      "epoch": 1.7404785061272126,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046953434820156363,
      "loss": 5.0429,
      "mean_token_accuracy": 0.19221502989530564,
      "num_tokens": 38819288.0,
      "step": 22370
    },
    {
      "entropy": 5.639925289154053,
      "epoch": 1.7408675354989303,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004695204205916624,
      "loss": 4.9931,
      "mean_token_accuracy": 0.2042757272720337,
      "num_tokens": 38827791.0,
      "step": 22375
    },
    {
      "entropy": 5.689849328994751,
      "epoch": 1.7412565648706477,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004695064900302235,
      "loss": 5.0529,
      "mean_token_accuracy": 0.19893163293600083,
      "num_tokens": 38836338.0,
      "step": 22380
    },
    {
      "entropy": 5.7244312286376955,
      "epoch": 1.7416455942423652,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004694925565174581,
      "loss": 5.0148,
      "mean_token_accuracy": 0.1985758975148201,
      "num_tokens": 38844768.0,
      "step": 22385
    },
    {
      "entropy": 5.75396466255188,
      "epoch": 1.742034623614083,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046947862005357777,
      "loss": 5.1498,
      "mean_token_accuracy": 0.19053892642259598,
      "num_tokens": 38854022.0,
      "step": 22390
    },
    {
      "entropy": 5.655837488174439,
      "epoch": 1.7424236529858004,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004694646806387939,
      "loss": 5.0215,
      "mean_token_accuracy": 0.2034000650048256,
      "num_tokens": 38862035.0,
      "step": 22395
    },
    {
      "entropy": 5.659757041931153,
      "epoch": 1.7428126823575179,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004694507382733181,
      "loss": 5.0617,
      "mean_token_accuracy": 0.2006596252322197,
      "num_tokens": 38870762.0,
      "step": 22400
    },
    {
      "entropy": 5.655742597579956,
      "epoch": 1.7432017117292355,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004694367929573618,
      "loss": 5.0009,
      "mean_token_accuracy": 0.2085031747817993,
      "num_tokens": 38879791.0,
      "step": 22405
    },
    {
      "entropy": 5.789531087875366,
      "epoch": 1.7435907411009532,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004694228446911367,
      "loss": 5.1339,
      "mean_token_accuracy": 0.19194365888834,
      "num_tokens": 38889926.0,
      "step": 22410
    },
    {
      "entropy": 5.698483085632324,
      "epoch": 1.7439797704726707,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004694088934748542,
      "loss": 5.0695,
      "mean_token_accuracy": 0.19272553473711013,
      "num_tokens": 38898428.0,
      "step": 22415
    },
    {
      "entropy": 5.623120498657227,
      "epoch": 1.7443687998443882,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004693949393087263,
      "loss": 5.1,
      "mean_token_accuracy": 0.19020944982767105,
      "num_tokens": 38906800.0,
      "step": 22420
    },
    {
      "entropy": 5.66190276145935,
      "epoch": 1.744757829216106,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004693809821929647,
      "loss": 5.0619,
      "mean_token_accuracy": 0.18886170089244841,
      "num_tokens": 38915947.0,
      "step": 22425
    },
    {
      "entropy": 5.74968729019165,
      "epoch": 1.7451468585878234,
      "grad_norm": 1.078125,
      "learning_rate": 0.000469367022127781,
      "loss": 5.1052,
      "mean_token_accuracy": 0.1926722154021263,
      "num_tokens": 38924895.0,
      "step": 22430
    },
    {
      "entropy": 5.715599250793457,
      "epoch": 1.7455358879595408,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046935305911338703,
      "loss": 5.086,
      "mean_token_accuracy": 0.18825379759073257,
      "num_tokens": 38933393.0,
      "step": 22435
    },
    {
      "entropy": 5.702757740020752,
      "epoch": 1.7459249173312585,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004693390931499948,
      "loss": 5.0827,
      "mean_token_accuracy": 0.2023254230618477,
      "num_tokens": 38942219.0,
      "step": 22440
    },
    {
      "entropy": 5.676919794082641,
      "epoch": 1.7463139467029762,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004693251242378162,
      "loss": 5.0381,
      "mean_token_accuracy": 0.1928280547261238,
      "num_tokens": 38950938.0,
      "step": 22445
    },
    {
      "entropy": 5.643939352035522,
      "epoch": 1.7467029760746935,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046931115237706304,
      "loss": 5.0497,
      "mean_token_accuracy": 0.197927089035511,
      "num_tokens": 38959805.0,
      "step": 22450
    },
    {
      "entropy": 5.6327300548553465,
      "epoch": 1.7470920054464112,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004692971775679475,
      "loss": 4.9843,
      "mean_token_accuracy": 0.2004704847931862,
      "num_tokens": 38968308.0,
      "step": 22455
    },
    {
      "entropy": 5.784936761856079,
      "epoch": 1.7474810348181289,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046928319981068154,
      "loss": 5.1239,
      "mean_token_accuracy": 0.19116262793540956,
      "num_tokens": 38976409.0,
      "step": 22460
    },
    {
      "entropy": 5.636967039108276,
      "epoch": 1.7478700641898464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004692692191054773,
      "loss": 5.0552,
      "mean_token_accuracy": 0.19898292571306228,
      "num_tokens": 38985193.0,
      "step": 22465
    },
    {
      "entropy": 5.633849000930786,
      "epoch": 1.7482590935615638,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004692552354525468,
      "loss": 5.0477,
      "mean_token_accuracy": 0.20176565051078796,
      "num_tokens": 38994141.0,
      "step": 22470
    },
    {
      "entropy": 5.696846437454224,
      "epoch": 1.7486481229332815,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004692412488521023,
      "loss": 5.1808,
      "mean_token_accuracy": 0.19120857119560242,
      "num_tokens": 39002953.0,
      "step": 22475
    },
    {
      "entropy": 5.722127914428711,
      "epoch": 1.749037152304999,
      "grad_norm": 1.296875,
      "learning_rate": 0.000469227259304356,
      "loss": 5.0954,
      "mean_token_accuracy": 0.20289715826511384,
      "num_tokens": 39010672.0,
      "step": 22480
    },
    {
      "entropy": 5.66537618637085,
      "epoch": 1.7494261816767165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046921326680952023,
      "loss": 5.1277,
      "mean_token_accuracy": 0.1905087947845459,
      "num_tokens": 39019677.0,
      "step": 22485
    },
    {
      "entropy": 5.680766582489014,
      "epoch": 1.7498152110484342,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004691992713678072,
      "loss": 5.0727,
      "mean_token_accuracy": 0.1929251730442047,
      "num_tokens": 39028165.0,
      "step": 22490
    },
    {
      "entropy": 5.621655464172363,
      "epoch": 1.7502042404201519,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004691852729794293,
      "loss": 5.0054,
      "mean_token_accuracy": 0.19845127910375596,
      "num_tokens": 39037537.0,
      "step": 22495
    },
    {
      "entropy": 5.6673445224761965,
      "epoch": 1.7505932697918691,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004691712716445991,
      "loss": 5.035,
      "mean_token_accuracy": 0.19844138622283936,
      "num_tokens": 39045861.0,
      "step": 22500
    },
    {
      "entropy": 5.709296083450317,
      "epoch": 1.7509822991635868,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004691572673635288,
      "loss": 5.0493,
      "mean_token_accuracy": 0.19929910451173782,
      "num_tokens": 39054526.0,
      "step": 22505
    },
    {
      "entropy": 5.66934986114502,
      "epoch": 1.7513713285353045,
      "grad_norm": 1.09375,
      "learning_rate": 0.000469143260136431,
      "loss": 5.063,
      "mean_token_accuracy": 0.19108586758375168,
      "num_tokens": 39063296.0,
      "step": 22510
    },
    {
      "entropy": 5.674559259414673,
      "epoch": 1.751760357907022,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004691292499635183,
      "loss": 4.9905,
      "mean_token_accuracy": 0.19899203181266784,
      "num_tokens": 39071570.0,
      "step": 22515
    },
    {
      "entropy": 5.643053960800171,
      "epoch": 1.7521493872787395,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004691152368450032,
      "loss": 5.0791,
      "mean_token_accuracy": 0.19975041300058366,
      "num_tokens": 39080798.0,
      "step": 22520
    },
    {
      "entropy": 5.658553981781006,
      "epoch": 1.7525384166504572,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046910122078109835,
      "loss": 5.0581,
      "mean_token_accuracy": 0.19790079891681672,
      "num_tokens": 39089572.0,
      "step": 22525
    },
    {
      "entropy": 5.7552674293518065,
      "epoch": 1.7529274460221747,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004690872017720163,
      "loss": 5.1612,
      "mean_token_accuracy": 0.19588136970996856,
      "num_tokens": 39098391.0,
      "step": 22530
    },
    {
      "entropy": 5.6929107189178465,
      "epoch": 1.7533164753938921,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046907317981797006,
      "loss": 5.1215,
      "mean_token_accuracy": 0.1927901938557625,
      "num_tokens": 39106892.0,
      "step": 22535
    },
    {
      "entropy": 5.680017471313477,
      "epoch": 1.7537055047656098,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046905915491917213,
      "loss": 5.0546,
      "mean_token_accuracy": 0.19367746710777284,
      "num_tokens": 39115635.0,
      "step": 22540
    },
    {
      "entropy": 5.704803323745727,
      "epoch": 1.7540945341373275,
      "grad_norm": 1.125,
      "learning_rate": 0.0004690451270758354,
      "loss": 5.0492,
      "mean_token_accuracy": 0.19545793235301973,
      "num_tokens": 39123665.0,
      "step": 22545
    },
    {
      "entropy": 5.704788017272949,
      "epoch": 1.7544835635090448,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046903109628817276,
      "loss": 5.1127,
      "mean_token_accuracy": 0.19106509834527968,
      "num_tokens": 39132727.0,
      "step": 22550
    },
    {
      "entropy": 5.669383955001831,
      "epoch": 1.7548725928807625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046901706255639703,
      "loss": 5.0842,
      "mean_token_accuracy": 0.1927453249692917,
      "num_tokens": 39140814.0,
      "step": 22555
    },
    {
      "entropy": 5.762030696868896,
      "epoch": 1.7552616222524802,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004690030258807212,
      "loss": 5.1575,
      "mean_token_accuracy": 0.18692436665296555,
      "num_tokens": 39149877.0,
      "step": 22560
    },
    {
      "entropy": 5.675040626525879,
      "epoch": 1.7556506516241976,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046898898626135827,
      "loss": 5.0909,
      "mean_token_accuracy": 0.19413626044988633,
      "num_tokens": 39158698.0,
      "step": 22565
    },
    {
      "entropy": 5.603366994857788,
      "epoch": 1.7560396809959151,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004689749436985211,
      "loss": 5.003,
      "mean_token_accuracy": 0.20513325482606887,
      "num_tokens": 39167546.0,
      "step": 22570
    },
    {
      "entropy": 5.669141006469727,
      "epoch": 1.7564287103676328,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046896089819242304,
      "loss": 5.0588,
      "mean_token_accuracy": 0.1938392236828804,
      "num_tokens": 39176547.0,
      "step": 22575
    },
    {
      "entropy": 5.745855236053467,
      "epoch": 1.7568177397393503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004689468497432771,
      "loss": 5.0725,
      "mean_token_accuracy": 0.198039710521698,
      "num_tokens": 39185273.0,
      "step": 22580
    },
    {
      "entropy": 5.703879451751709,
      "epoch": 1.7572067691110678,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004689327983512963,
      "loss": 5.0321,
      "mean_token_accuracy": 0.20252016484737395,
      "num_tokens": 39194204.0,
      "step": 22585
    },
    {
      "entropy": 5.667798328399658,
      "epoch": 1.7575957984827855,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046891874401669404,
      "loss": 4.9768,
      "mean_token_accuracy": 0.20301975756883622,
      "num_tokens": 39202951.0,
      "step": 22590
    },
    {
      "entropy": 5.588111639022827,
      "epoch": 1.7579848278545032,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004689046867396834,
      "loss": 5.0512,
      "mean_token_accuracy": 0.19673458784818648,
      "num_tokens": 39211508.0,
      "step": 22595
    },
    {
      "entropy": 5.7373357772827145,
      "epoch": 1.7583738572262204,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004688906265204779,
      "loss": 5.113,
      "mean_token_accuracy": 0.19004614800214767,
      "num_tokens": 39219804.0,
      "step": 22600
    },
    {
      "entropy": 5.753034400939941,
      "epoch": 1.7587628865979381,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004688765633592907,
      "loss": 5.142,
      "mean_token_accuracy": 0.1904366448521614,
      "num_tokens": 39228204.0,
      "step": 22605
    },
    {
      "entropy": 5.6577208042144775,
      "epoch": 1.7591519159696558,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004688624972563352,
      "loss": 5.0685,
      "mean_token_accuracy": 0.19090791046619415,
      "num_tokens": 39237003.0,
      "step": 22610
    },
    {
      "entropy": 5.621747016906738,
      "epoch": 1.7595409453413733,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004688484282118249,
      "loss": 5.0332,
      "mean_token_accuracy": 0.19215899407863618,
      "num_tokens": 39245636.0,
      "step": 22615
    },
    {
      "entropy": 5.5834471702575685,
      "epoch": 1.7599299747130908,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004688343562259732,
      "loss": 4.9189,
      "mean_token_accuracy": 0.2004231721162796,
      "num_tokens": 39254288.0,
      "step": 22620
    },
    {
      "entropy": 5.679382562637329,
      "epoch": 1.7603190040848085,
      "grad_norm": 1.125,
      "learning_rate": 0.0004688202812989937,
      "loss": 5.1094,
      "mean_token_accuracy": 0.1992802456021309,
      "num_tokens": 39263217.0,
      "step": 22625
    },
    {
      "entropy": 5.604801511764526,
      "epoch": 1.760708033456526,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004688062034311,
      "loss": 5.0664,
      "mean_token_accuracy": 0.19958146810531616,
      "num_tokens": 39271240.0,
      "step": 22630
    },
    {
      "entropy": 5.609039688110352,
      "epoch": 1.7610970628282434,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046879212262250546,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20655782222747804,
      "num_tokens": 39279665.0,
      "step": 22635
    },
    {
      "entropy": 5.699652242660522,
      "epoch": 1.761486092199961,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046877803887342406,
      "loss": 5.175,
      "mean_token_accuracy": 0.192455293238163,
      "num_tokens": 39287975.0,
      "step": 22640
    },
    {
      "entropy": 5.716793346405029,
      "epoch": 1.7618751215716788,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004687639521840693,
      "loss": 5.1186,
      "mean_token_accuracy": 0.19389519840478897,
      "num_tokens": 39296541.0,
      "step": 22645
    },
    {
      "entropy": 5.695902109146118,
      "epoch": 1.7622641509433963,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046874986255465495,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20217987298965454,
      "num_tokens": 39304710.0,
      "step": 22650
    },
    {
      "entropy": 5.657218360900879,
      "epoch": 1.7626531803151138,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046873576998539485,
      "loss": 4.9624,
      "mean_token_accuracy": 0.20321538746356965,
      "num_tokens": 39313236.0,
      "step": 22655
    },
    {
      "entropy": 5.607696008682251,
      "epoch": 1.7630422096868315,
      "grad_norm": 1.125,
      "learning_rate": 0.0004687216744765028,
      "loss": 4.9099,
      "mean_token_accuracy": 0.20135416239500045,
      "num_tokens": 39322656.0,
      "step": 22660
    },
    {
      "entropy": 5.656253337860107,
      "epoch": 1.763431239058549,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004687075760281927,
      "loss": 4.9794,
      "mean_token_accuracy": 0.2120069906115532,
      "num_tokens": 39331074.0,
      "step": 22665
    },
    {
      "entropy": 5.643962144851685,
      "epoch": 1.7638202684302664,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004686934746406784,
      "loss": 5.0419,
      "mean_token_accuracy": 0.1975744664669037,
      "num_tokens": 39339450.0,
      "step": 22670
    },
    {
      "entropy": 5.678130435943603,
      "epoch": 1.764209297801984,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046867937031417397,
      "loss": 5.1565,
      "mean_token_accuracy": 0.19002707600593566,
      "num_tokens": 39347270.0,
      "step": 22675
    },
    {
      "entropy": 5.656103849411011,
      "epoch": 1.7645983271737016,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004686652630488933,
      "loss": 5.0679,
      "mean_token_accuracy": 0.1878320664167404,
      "num_tokens": 39355853.0,
      "step": 22680
    },
    {
      "entropy": 5.580417728424072,
      "epoch": 1.764987356545419,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046865115284505063,
      "loss": 4.9699,
      "mean_token_accuracy": 0.20845221728086472,
      "num_tokens": 39364160.0,
      "step": 22685
    },
    {
      "entropy": 5.622933101654053,
      "epoch": 1.7653763859171367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046863703970285986,
      "loss": 5.0646,
      "mean_token_accuracy": 0.19981067329645158,
      "num_tokens": 39372504.0,
      "step": 22690
    },
    {
      "entropy": 5.663434171676636,
      "epoch": 1.7657654152888544,
      "grad_norm": 1.125,
      "learning_rate": 0.0004686229236225352,
      "loss": 5.056,
      "mean_token_accuracy": 0.2035764843225479,
      "num_tokens": 39380883.0,
      "step": 22695
    },
    {
      "entropy": 5.7430390357971195,
      "epoch": 1.766154444660572,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004686088046042909,
      "loss": 5.1099,
      "mean_token_accuracy": 0.1917836457490921,
      "num_tokens": 39389943.0,
      "step": 22700
    },
    {
      "entropy": 5.596102619171143,
      "epoch": 1.7665434740322894,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046859468264834114,
      "loss": 5.0603,
      "mean_token_accuracy": 0.20420622676610947,
      "num_tokens": 39398919.0,
      "step": 22705
    },
    {
      "entropy": 5.67539701461792,
      "epoch": 1.766932503404007,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046858055775490017,
      "loss": 5.1429,
      "mean_token_accuracy": 0.1941673770546913,
      "num_tokens": 39407547.0,
      "step": 22710
    },
    {
      "entropy": 5.648145771026611,
      "epoch": 1.7673215327757246,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004685664299241825,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20077628642320633,
      "num_tokens": 39416264.0,
      "step": 22715
    },
    {
      "entropy": 5.62908411026001,
      "epoch": 1.767710562147442,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004685522991564022,
      "loss": 5.0399,
      "mean_token_accuracy": 0.19342577159404756,
      "num_tokens": 39424104.0,
      "step": 22720
    },
    {
      "entropy": 5.693596601486206,
      "epoch": 1.7680995915191597,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004685381654517738,
      "loss": 5.0853,
      "mean_token_accuracy": 0.19596892595291138,
      "num_tokens": 39432443.0,
      "step": 22725
    },
    {
      "entropy": 5.620149850845337,
      "epoch": 1.7684886208908772,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004685240288105119,
      "loss": 5.0557,
      "mean_token_accuracy": 0.19603675305843354,
      "num_tokens": 39441620.0,
      "step": 22730
    },
    {
      "entropy": 5.628264617919922,
      "epoch": 1.7688776502625947,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046850988923283085,
      "loss": 5.0249,
      "mean_token_accuracy": 0.2046712026000023,
      "num_tokens": 39449854.0,
      "step": 22735
    },
    {
      "entropy": 5.644801044464112,
      "epoch": 1.7692666796343124,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046849574671894523,
      "loss": 5.0969,
      "mean_token_accuracy": 0.19389274567365647,
      "num_tokens": 39458496.0,
      "step": 22740
    },
    {
      "entropy": 5.695472192764282,
      "epoch": 1.76965570900603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046848160126906956,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19473907947540284,
      "num_tokens": 39466545.0,
      "step": 22745
    },
    {
      "entropy": 5.68575439453125,
      "epoch": 1.7700447383777476,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004684674528834186,
      "loss": 5.0967,
      "mean_token_accuracy": 0.19515854269266128,
      "num_tokens": 39475771.0,
      "step": 22750
    },
    {
      "entropy": 5.6835036277771,
      "epoch": 1.770433767749465,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046845330156220703,
      "loss": 5.1075,
      "mean_token_accuracy": 0.19743928909301758,
      "num_tokens": 39484244.0,
      "step": 22755
    },
    {
      "entropy": 5.718806219100952,
      "epoch": 1.7708227971211827,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004684391473056495,
      "loss": 5.1383,
      "mean_token_accuracy": 0.19019536375999452,
      "num_tokens": 39493138.0,
      "step": 22760
    },
    {
      "entropy": 5.737482309341431,
      "epoch": 1.7712118264929002,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046842499011396076,
      "loss": 5.0948,
      "mean_token_accuracy": 0.1941129133105278,
      "num_tokens": 39502065.0,
      "step": 22765
    },
    {
      "entropy": 5.697163152694702,
      "epoch": 1.7716008558646177,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046841082998735575,
      "loss": 5.0726,
      "mean_token_accuracy": 0.20459358245134354,
      "num_tokens": 39510095.0,
      "step": 22770
    },
    {
      "entropy": 5.635035753250122,
      "epoch": 1.7719898852363354,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046839666692604916,
      "loss": 5.1274,
      "mean_token_accuracy": 0.19442155808210373,
      "num_tokens": 39518104.0,
      "step": 22775
    },
    {
      "entropy": 5.716279029846191,
      "epoch": 1.7723789146080529,
      "grad_norm": 1.5,
      "learning_rate": 0.000468382500930256,
      "loss": 5.086,
      "mean_token_accuracy": 0.1993977263569832,
      "num_tokens": 39526768.0,
      "step": 22780
    },
    {
      "entropy": 5.673862409591675,
      "epoch": 1.7727679439797703,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046836833200019116,
      "loss": 5.0487,
      "mean_token_accuracy": 0.19186717867851258,
      "num_tokens": 39535847.0,
      "step": 22785
    },
    {
      "entropy": 5.734010219573975,
      "epoch": 1.773156973351488,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004683541601360697,
      "loss": 5.1132,
      "mean_token_accuracy": 0.20216183960437775,
      "num_tokens": 39544364.0,
      "step": 22790
    },
    {
      "entropy": 5.702363157272339,
      "epoch": 1.7735460027232057,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046833998533810653,
      "loss": 5.1274,
      "mean_token_accuracy": 0.18689947426319123,
      "num_tokens": 39552733.0,
      "step": 22795
    },
    {
      "entropy": 5.695834112167359,
      "epoch": 1.7739350320949232,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004683258076065169,
      "loss": 5.0619,
      "mean_token_accuracy": 0.19932250082492828,
      "num_tokens": 39561580.0,
      "step": 22800
    },
    {
      "entropy": 5.625617504119873,
      "epoch": 1.7743240614666407,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046831162694151586,
      "loss": 5.0517,
      "mean_token_accuracy": 0.19228088706731797,
      "num_tokens": 39570283.0,
      "step": 22805
    },
    {
      "entropy": 5.668644189834595,
      "epoch": 1.7747130908383584,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046829744334331855,
      "loss": 5.0251,
      "mean_token_accuracy": 0.19798846840858458,
      "num_tokens": 39578835.0,
      "step": 22810
    },
    {
      "entropy": 5.684339761734009,
      "epoch": 1.7751021202100759,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046828325681214013,
      "loss": 5.0779,
      "mean_token_accuracy": 0.20166678726673126,
      "num_tokens": 39587053.0,
      "step": 22815
    },
    {
      "entropy": 5.676909780502319,
      "epoch": 1.7754911495817933,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000468269067348196,
      "loss": 5.1579,
      "mean_token_accuracy": 0.18229772597551347,
      "num_tokens": 39595799.0,
      "step": 22820
    },
    {
      "entropy": 5.707350969314575,
      "epoch": 1.775880178953511,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004682548749517014,
      "loss": 5.0331,
      "mean_token_accuracy": 0.1991925686597824,
      "num_tokens": 39604248.0,
      "step": 22825
    },
    {
      "entropy": 5.624576568603516,
      "epoch": 1.7762692083252287,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046824067962287163,
      "loss": 5.0319,
      "mean_token_accuracy": 0.19990004897117614,
      "num_tokens": 39612690.0,
      "step": 22830
    },
    {
      "entropy": 5.666665887832641,
      "epoch": 1.776658237696946,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004682264813619221,
      "loss": 5.0725,
      "mean_token_accuracy": 0.19386529922485352,
      "num_tokens": 39621784.0,
      "step": 22835
    },
    {
      "entropy": 5.658817148208618,
      "epoch": 1.7770472670686637,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046821228016906836,
      "loss": 5.0013,
      "mean_token_accuracy": 0.2040756016969681,
      "num_tokens": 39630455.0,
      "step": 22840
    },
    {
      "entropy": 5.635995864868164,
      "epoch": 1.7774362964403814,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004681980760445257,
      "loss": 5.0147,
      "mean_token_accuracy": 0.19693631827831268,
      "num_tokens": 39639017.0,
      "step": 22845
    },
    {
      "entropy": 5.642162275314331,
      "epoch": 1.7778253258120988,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004681838689885098,
      "loss": 4.9665,
      "mean_token_accuracy": 0.2065414234995842,
      "num_tokens": 39648083.0,
      "step": 22850
    },
    {
      "entropy": 5.661451816558838,
      "epoch": 1.7782143551838163,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004681696590012362,
      "loss": 5.0582,
      "mean_token_accuracy": 0.20174534022808074,
      "num_tokens": 39656887.0,
      "step": 22855
    },
    {
      "entropy": 5.6726624965667725,
      "epoch": 1.778603384555534,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046815544608292043,
      "loss": 5.089,
      "mean_token_accuracy": 0.1949265718460083,
      "num_tokens": 39665658.0,
      "step": 22860
    },
    {
      "entropy": 5.6686193466186525,
      "epoch": 1.7789924139272515,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004681412302337782,
      "loss": 5.0508,
      "mean_token_accuracy": 0.19710324704647064,
      "num_tokens": 39673423.0,
      "step": 22865
    },
    {
      "entropy": 5.691291570663452,
      "epoch": 1.779381443298969,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004681270114540252,
      "loss": 5.0719,
      "mean_token_accuracy": 0.20220151543617249,
      "num_tokens": 39682382.0,
      "step": 22870
    },
    {
      "entropy": 5.700251150131225,
      "epoch": 1.7797704726706867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004681127897438772,
      "loss": 5.1107,
      "mean_token_accuracy": 0.19690333157777787,
      "num_tokens": 39691089.0,
      "step": 22875
    },
    {
      "entropy": 5.729765319824219,
      "epoch": 1.7801595020424044,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046809856510355007,
      "loss": 5.1577,
      "mean_token_accuracy": 0.19188273549079896,
      "num_tokens": 39700888.0,
      "step": 22880
    },
    {
      "entropy": 5.723038196563721,
      "epoch": 1.7805485314141216,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004680843375332595,
      "loss": 5.1142,
      "mean_token_accuracy": 0.20196729749441147,
      "num_tokens": 39708515.0,
      "step": 22885
    },
    {
      "entropy": 5.619906806945801,
      "epoch": 1.7809375607858393,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046807010703322143,
      "loss": 4.9656,
      "mean_token_accuracy": 0.2009779006242752,
      "num_tokens": 39716790.0,
      "step": 22890
    },
    {
      "entropy": 5.72461314201355,
      "epoch": 1.781326590157557,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004680558736036518,
      "loss": 5.0329,
      "mean_token_accuracy": 0.19692874401807786,
      "num_tokens": 39726292.0,
      "step": 22895
    },
    {
      "entropy": 5.641946840286255,
      "epoch": 1.7817156195292745,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004680416372447666,
      "loss": 5.0403,
      "mean_token_accuracy": 0.20342497080564498,
      "num_tokens": 39735222.0,
      "step": 22900
    },
    {
      "entropy": 5.627364683151245,
      "epoch": 1.782104648900992,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046802739795678177,
      "loss": 5.0891,
      "mean_token_accuracy": 0.1946094438433647,
      "num_tokens": 39744619.0,
      "step": 22905
    },
    {
      "entropy": 5.673235034942627,
      "epoch": 1.7824936782727097,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046801315573991346,
      "loss": 5.0901,
      "mean_token_accuracy": 0.19993328750133516,
      "num_tokens": 39752394.0,
      "step": 22910
    },
    {
      "entropy": 5.633034992218017,
      "epoch": 1.7828827076444271,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046799891059437764,
      "loss": 4.991,
      "mean_token_accuracy": 0.20043096244335173,
      "num_tokens": 39760529.0,
      "step": 22915
    },
    {
      "entropy": 5.720476579666138,
      "epoch": 1.7832717370161446,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046798466252039056,
      "loss": 5.1383,
      "mean_token_accuracy": 0.19344329684972764,
      "num_tokens": 39768231.0,
      "step": 22920
    },
    {
      "entropy": 5.75451340675354,
      "epoch": 1.7836607663878623,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046797041151816845,
      "loss": 5.1984,
      "mean_token_accuracy": 0.19174774885177612,
      "num_tokens": 39776773.0,
      "step": 22925
    },
    {
      "entropy": 5.7062615871429445,
      "epoch": 1.78404979575958,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046795615758792747,
      "loss": 5.0567,
      "mean_token_accuracy": 0.1905716598033905,
      "num_tokens": 39785248.0,
      "step": 22930
    },
    {
      "entropy": 5.632964611053467,
      "epoch": 1.7844388251312973,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004679419007298839,
      "loss": 5.0575,
      "mean_token_accuracy": 0.19742142856121064,
      "num_tokens": 39794794.0,
      "step": 22935
    },
    {
      "entropy": 5.6968883037567135,
      "epoch": 1.784827854503015,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004679276409442541,
      "loss": 5.0396,
      "mean_token_accuracy": 0.1929635375738144,
      "num_tokens": 39803662.0,
      "step": 22940
    },
    {
      "entropy": 5.7244353771209715,
      "epoch": 1.7852168838747327,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004679133782312544,
      "loss": 5.0432,
      "mean_token_accuracy": 0.19755503684282302,
      "num_tokens": 39812189.0,
      "step": 22945
    },
    {
      "entropy": 5.679898691177368,
      "epoch": 1.7856059132464501,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004678991125911013,
      "loss": 5.0813,
      "mean_token_accuracy": 0.19720999300479888,
      "num_tokens": 39820831.0,
      "step": 22950
    },
    {
      "entropy": 5.571303224563598,
      "epoch": 1.7859949426181676,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004678848440240111,
      "loss": 4.9429,
      "mean_token_accuracy": 0.2035401001572609,
      "num_tokens": 39829380.0,
      "step": 22955
    },
    {
      "entropy": 5.631348943710327,
      "epoch": 1.7863839719898853,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004678705725302005,
      "loss": 5.0633,
      "mean_token_accuracy": 0.20108917355537415,
      "num_tokens": 39838243.0,
      "step": 22960
    },
    {
      "entropy": 5.689360809326172,
      "epoch": 1.7867730013616028,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004678562981098859,
      "loss": 5.0397,
      "mean_token_accuracy": 0.20027730166912078,
      "num_tokens": 39846885.0,
      "step": 22965
    },
    {
      "entropy": 5.665248441696167,
      "epoch": 1.7871620307333203,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046784202076328394,
      "loss": 5.0054,
      "mean_token_accuracy": 0.20157646387815475,
      "num_tokens": 39855337.0,
      "step": 22970
    },
    {
      "entropy": 5.590274238586426,
      "epoch": 1.787551060105038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046782774049061124,
      "loss": 5.0752,
      "mean_token_accuracy": 0.20414217114448546,
      "num_tokens": 39864606.0,
      "step": 22975
    },
    {
      "entropy": 5.74508376121521,
      "epoch": 1.7879400894767556,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004678134572920845,
      "loss": 5.0775,
      "mean_token_accuracy": 0.19494454711675643,
      "num_tokens": 39872621.0,
      "step": 22980
    },
    {
      "entropy": 5.710721445083618,
      "epoch": 1.788329118848473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004677991711679204,
      "loss": 5.0525,
      "mean_token_accuracy": 0.20031729489564895,
      "num_tokens": 39880525.0,
      "step": 22985
    },
    {
      "entropy": 5.645658349990844,
      "epoch": 1.7887181482201906,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046778488211833585,
      "loss": 4.9711,
      "mean_token_accuracy": 0.20704260915517808,
      "num_tokens": 39889541.0,
      "step": 22990
    },
    {
      "entropy": 5.672270727157593,
      "epoch": 1.7891071775919083,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004677705901435475,
      "loss": 5.0438,
      "mean_token_accuracy": 0.1979015737771988,
      "num_tokens": 39897682.0,
      "step": 22995
    },
    {
      "entropy": 5.703026056289673,
      "epoch": 1.7894962069636258,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004677562952437723,
      "loss": 5.0597,
      "mean_token_accuracy": 0.19432690292596816,
      "num_tokens": 39906656.0,
      "step": 23000
    },
    {
      "entropy": 5.723121500015258,
      "epoch": 1.7898852363353432,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046774199741922705,
      "loss": 5.1327,
      "mean_token_accuracy": 0.19010985791683196,
      "num_tokens": 39915202.0,
      "step": 23005
    },
    {
      "entropy": 5.621491575241089,
      "epoch": 1.790274265707061,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046772769667012884,
      "loss": 4.978,
      "mean_token_accuracy": 0.20201105028390884,
      "num_tokens": 39923510.0,
      "step": 23010
    },
    {
      "entropy": 5.5676689624786375,
      "epoch": 1.7906632950787784,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004677133929966946,
      "loss": 5.0231,
      "mean_token_accuracy": 0.2080300882458687,
      "num_tokens": 39932468.0,
      "step": 23015
    },
    {
      "entropy": 5.71773247718811,
      "epoch": 1.791052324450496,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046769908639914134,
      "loss": 5.1256,
      "mean_token_accuracy": 0.19488838613033294,
      "num_tokens": 39942758.0,
      "step": 23020
    },
    {
      "entropy": 5.7325835704803465,
      "epoch": 1.7914413538222136,
      "grad_norm": 1.125,
      "learning_rate": 0.0004676847768776861,
      "loss": 5.1069,
      "mean_token_accuracy": 0.19283661395311355,
      "num_tokens": 39950829.0,
      "step": 23025
    },
    {
      "entropy": 5.620270252227783,
      "epoch": 1.7918303831939313,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004676704644325462,
      "loss": 5.0978,
      "mean_token_accuracy": 0.19478842318058015,
      "num_tokens": 39960204.0,
      "step": 23030
    },
    {
      "entropy": 5.662912130355835,
      "epoch": 1.7922194125656488,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004676561490639385,
      "loss": 5.0521,
      "mean_token_accuracy": 0.19848431050777435,
      "num_tokens": 39968513.0,
      "step": 23035
    },
    {
      "entropy": 5.654798793792724,
      "epoch": 1.7926084419373662,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004676418307720805,
      "loss": 5.0886,
      "mean_token_accuracy": 0.19719878286123277,
      "num_tokens": 39977181.0,
      "step": 23040
    },
    {
      "entropy": 5.558326864242554,
      "epoch": 1.792997471309084,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004676275095571892,
      "loss": 4.9012,
      "mean_token_accuracy": 0.2106527015566826,
      "num_tokens": 39986760.0,
      "step": 23045
    },
    {
      "entropy": 5.6533793926239015,
      "epoch": 1.7933865006808014,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046761318541948215,
      "loss": 5.1154,
      "mean_token_accuracy": 0.18526081889867782,
      "num_tokens": 39995203.0,
      "step": 23050
    },
    {
      "entropy": 5.643955230712891,
      "epoch": 1.7937755300525189,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004675988583591766,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20396423041820527,
      "num_tokens": 40003526.0,
      "step": 23055
    },
    {
      "entropy": 5.681190490722656,
      "epoch": 1.7941645594242366,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046758452837648997,
      "loss": 5.0547,
      "mean_token_accuracy": 0.19564736485481263,
      "num_tokens": 40012036.0,
      "step": 23060
    },
    {
      "entropy": 5.635527181625366,
      "epoch": 1.794553588795954,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004675701954716395,
      "loss": 5.0653,
      "mean_token_accuracy": 0.19711072444915773,
      "num_tokens": 40020237.0,
      "step": 23065
    },
    {
      "entropy": 5.611155700683594,
      "epoch": 1.7949426181676715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046755585964484286,
      "loss": 4.9757,
      "mean_token_accuracy": 0.20652034133672714,
      "num_tokens": 40028749.0,
      "step": 23070
    },
    {
      "entropy": 5.6394758224487305,
      "epoch": 1.7953316475393892,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004675415208963177,
      "loss": 4.9477,
      "mean_token_accuracy": 0.20087110251188278,
      "num_tokens": 40036799.0,
      "step": 23075
    },
    {
      "entropy": 5.629435968399048,
      "epoch": 1.795720676911107,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046752717922628125,
      "loss": 5.0887,
      "mean_token_accuracy": 0.19493671059608458,
      "num_tokens": 40045416.0,
      "step": 23080
    },
    {
      "entropy": 5.7006933212280275,
      "epoch": 1.7961097062828244,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004675128346349514,
      "loss": 5.1208,
      "mean_token_accuracy": 0.1985117867588997,
      "num_tokens": 40054128.0,
      "step": 23085
    },
    {
      "entropy": 5.775534915924072,
      "epoch": 1.7964987356545419,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046749848712254554,
      "loss": 5.1354,
      "mean_token_accuracy": 0.19225320518016814,
      "num_tokens": 40062765.0,
      "step": 23090
    },
    {
      "entropy": 5.642290782928467,
      "epoch": 1.7968877650262596,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046748413668928164,
      "loss": 5.0555,
      "mean_token_accuracy": 0.20141281187534332,
      "num_tokens": 40071190.0,
      "step": 23095
    },
    {
      "entropy": 5.654160833358764,
      "epoch": 1.797276794397977,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004674697833353774,
      "loss": 5.0654,
      "mean_token_accuracy": 0.1998469427227974,
      "num_tokens": 40080552.0,
      "step": 23100
    },
    {
      "entropy": 5.67447862625122,
      "epoch": 1.7976658237696945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046745542706105045,
      "loss": 5.0817,
      "mean_token_accuracy": 0.20118483752012253,
      "num_tokens": 40089407.0,
      "step": 23105
    },
    {
      "entropy": 5.794737243652344,
      "epoch": 1.7980548531414122,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046744106786651875,
      "loss": 5.2701,
      "mean_token_accuracy": 0.18606453239917756,
      "num_tokens": 40097946.0,
      "step": 23110
    },
    {
      "entropy": 5.655776882171631,
      "epoch": 1.7984438825131297,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004674267057520001,
      "loss": 5.0128,
      "mean_token_accuracy": 0.20463781356811522,
      "num_tokens": 40105901.0,
      "step": 23115
    },
    {
      "entropy": 5.61436710357666,
      "epoch": 1.7988329118848472,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004674123407177125,
      "loss": 4.9325,
      "mean_token_accuracy": 0.2023484319448471,
      "num_tokens": 40114481.0,
      "step": 23120
    },
    {
      "entropy": 5.646229028701782,
      "epoch": 1.7992219412565649,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046739797276387394,
      "loss": 5.0051,
      "mean_token_accuracy": 0.20337729305028915,
      "num_tokens": 40123086.0,
      "step": 23125
    },
    {
      "entropy": 5.654345750808716,
      "epoch": 1.7996109706282826,
      "grad_norm": 1.125,
      "learning_rate": 0.00046738360189070235,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20031450390815736,
      "num_tokens": 40131497.0,
      "step": 23130
    },
    {
      "entropy": 5.619352340698242,
      "epoch": 1.8,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004673692280984157,
      "loss": 5.0442,
      "mean_token_accuracy": 0.19908270388841628,
      "num_tokens": 40139632.0,
      "step": 23135
    },
    {
      "entropy": 5.62560658454895,
      "epoch": 1.8003890293717175,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004673548513872324,
      "loss": 5.0156,
      "mean_token_accuracy": 0.2011853873729706,
      "num_tokens": 40148312.0,
      "step": 23140
    },
    {
      "entropy": 5.677383899688721,
      "epoch": 1.8007780587434352,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046734047175737026,
      "loss": 5.089,
      "mean_token_accuracy": 0.1920654758810997,
      "num_tokens": 40156750.0,
      "step": 23145
    },
    {
      "entropy": 5.630202770233154,
      "epoch": 1.8011670881151527,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046732608920904754,
      "loss": 5.0582,
      "mean_token_accuracy": 0.19776277244091034,
      "num_tokens": 40165334.0,
      "step": 23150
    },
    {
      "entropy": 5.642322587966919,
      "epoch": 1.8015561174868702,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004673117037424827,
      "loss": 5.0427,
      "mean_token_accuracy": 0.20004768669605255,
      "num_tokens": 40174217.0,
      "step": 23155
    },
    {
      "entropy": 5.602140998840332,
      "epoch": 1.8019451468585879,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046729731535789375,
      "loss": 5.012,
      "mean_token_accuracy": 0.20319269448518754,
      "num_tokens": 40182889.0,
      "step": 23160
    },
    {
      "entropy": 5.66202449798584,
      "epoch": 1.8023341762303053,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046728292405549913,
      "loss": 5.069,
      "mean_token_accuracy": 0.19812913239002228,
      "num_tokens": 40191428.0,
      "step": 23165
    },
    {
      "entropy": 5.659556674957275,
      "epoch": 1.8027232056020228,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004672685298355172,
      "loss": 5.0341,
      "mean_token_accuracy": 0.20688219219446183,
      "num_tokens": 40200204.0,
      "step": 23170
    },
    {
      "entropy": 5.641662454605102,
      "epoch": 1.8031122349737405,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004672541326981664,
      "loss": 4.9308,
      "mean_token_accuracy": 0.21110827773809432,
      "num_tokens": 40208326.0,
      "step": 23175
    },
    {
      "entropy": 5.569920396804809,
      "epoch": 1.8035012643454582,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000467239732643665,
      "loss": 5.0495,
      "mean_token_accuracy": 0.20019952654838563,
      "num_tokens": 40217300.0,
      "step": 23180
    },
    {
      "entropy": 5.642634057998658,
      "epoch": 1.8038902937171757,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046722532967223183,
      "loss": 5.1421,
      "mean_token_accuracy": 0.19416471868753432,
      "num_tokens": 40226695.0,
      "step": 23185
    },
    {
      "entropy": 5.692462491989136,
      "epoch": 1.8042793230888932,
      "grad_norm": 1.125,
      "learning_rate": 0.0004672109237840851,
      "loss": 5.0292,
      "mean_token_accuracy": 0.19761148393154143,
      "num_tokens": 40235286.0,
      "step": 23190
    },
    {
      "entropy": 5.609381532669067,
      "epoch": 1.8046683524606109,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046719651497944355,
      "loss": 4.9935,
      "mean_token_accuracy": 0.20100178122520446,
      "num_tokens": 40244065.0,
      "step": 23195
    },
    {
      "entropy": 5.70858244895935,
      "epoch": 1.8050573818323283,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004671821032585259,
      "loss": 5.1978,
      "mean_token_accuracy": 0.18573150485754014,
      "num_tokens": 40253058.0,
      "step": 23200
    },
    {
      "entropy": 5.712513828277588,
      "epoch": 1.8054464112040458,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004671676886215506,
      "loss": 5.0845,
      "mean_token_accuracy": 0.19421554058790208,
      "num_tokens": 40261797.0,
      "step": 23205
    },
    {
      "entropy": 5.646021223068237,
      "epoch": 1.8058354405757635,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004671532710687365,
      "loss": 5.0395,
      "mean_token_accuracy": 0.20096675753593446,
      "num_tokens": 40270003.0,
      "step": 23210
    },
    {
      "entropy": 5.638779211044311,
      "epoch": 1.806224469947481,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004671388506003024,
      "loss": 5.0679,
      "mean_token_accuracy": 0.19866203665733337,
      "num_tokens": 40278241.0,
      "step": 23215
    },
    {
      "entropy": 5.677653694152832,
      "epoch": 1.8066134993191985,
      "grad_norm": 1.25,
      "learning_rate": 0.0004671244272164671,
      "loss": 5.1198,
      "mean_token_accuracy": 0.19523296654224395,
      "num_tokens": 40286316.0,
      "step": 23220
    },
    {
      "entropy": 5.667910099029541,
      "epoch": 1.8070025286909162,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046711000091744935,
      "loss": 5.0748,
      "mean_token_accuracy": 0.19464613497257233,
      "num_tokens": 40295378.0,
      "step": 23225
    },
    {
      "entropy": 5.717720699310303,
      "epoch": 1.8073915580626339,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004670955717034681,
      "loss": 5.0814,
      "mean_token_accuracy": 0.19408326596021652,
      "num_tokens": 40305379.0,
      "step": 23230
    },
    {
      "entropy": 5.710579967498779,
      "epoch": 1.8077805874343513,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046708113957474233,
      "loss": 5.1186,
      "mean_token_accuracy": 0.1927325263619423,
      "num_tokens": 40314174.0,
      "step": 23235
    },
    {
      "entropy": 5.71313853263855,
      "epoch": 1.8081696168060688,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000467066704531491,
      "loss": 5.1042,
      "mean_token_accuracy": 0.20000478625297546,
      "num_tokens": 40323708.0,
      "step": 23240
    },
    {
      "entropy": 5.647215175628662,
      "epoch": 1.8085586461777865,
      "grad_norm": 1.125,
      "learning_rate": 0.0004670522665739332,
      "loss": 5.0004,
      "mean_token_accuracy": 0.20029812306165695,
      "num_tokens": 40333664.0,
      "step": 23245
    },
    {
      "entropy": 5.663864660263061,
      "epoch": 1.808947675549504,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004670378257022878,
      "loss": 5.0542,
      "mean_token_accuracy": 0.20027079284191132,
      "num_tokens": 40342037.0,
      "step": 23250
    },
    {
      "entropy": 5.657285118103028,
      "epoch": 1.8093367049212215,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046702338191677414,
      "loss": 4.9995,
      "mean_token_accuracy": 0.20434896498918534,
      "num_tokens": 40350472.0,
      "step": 23255
    },
    {
      "entropy": 5.667795658111572,
      "epoch": 1.8097257342929391,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004670089352176113,
      "loss": 5.0727,
      "mean_token_accuracy": 0.19985506683588028,
      "num_tokens": 40358829.0,
      "step": 23260
    },
    {
      "entropy": 5.629911088943482,
      "epoch": 1.8101147636646568,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046699448560501847,
      "loss": 5.0737,
      "mean_token_accuracy": 0.19637651294469832,
      "num_tokens": 40367414.0,
      "step": 23265
    },
    {
      "entropy": 5.667527914047241,
      "epoch": 1.810503793036374,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004669800330792149,
      "loss": 5.1191,
      "mean_token_accuracy": 0.1909766376018524,
      "num_tokens": 40375969.0,
      "step": 23270
    },
    {
      "entropy": 5.722210788726807,
      "epoch": 1.8108928224080918,
      "grad_norm": 1.125,
      "learning_rate": 0.00046696557764041994,
      "loss": 5.0594,
      "mean_token_accuracy": 0.19282193928956987,
      "num_tokens": 40384200.0,
      "step": 23275
    },
    {
      "entropy": 5.741775274276733,
      "epoch": 1.8112818517798095,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004669511192888528,
      "loss": 5.1682,
      "mean_token_accuracy": 0.19382085800170898,
      "num_tokens": 40392891.0,
      "step": 23280
    },
    {
      "entropy": 5.717968320846557,
      "epoch": 1.811670881151527,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046693665802473294,
      "loss": 5.1208,
      "mean_token_accuracy": 0.19663182497024537,
      "num_tokens": 40402009.0,
      "step": 23285
    },
    {
      "entropy": 5.72480263710022,
      "epoch": 1.8120599105232444,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046692219384827986,
      "loss": 5.1051,
      "mean_token_accuracy": 0.1851302906870842,
      "num_tokens": 40410756.0,
      "step": 23290
    },
    {
      "entropy": 5.706399059295654,
      "epoch": 1.8124489398949621,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004669077267597129,
      "loss": 5.1325,
      "mean_token_accuracy": 0.19384849965572357,
      "num_tokens": 40420087.0,
      "step": 23295
    },
    {
      "entropy": 5.619845056533814,
      "epoch": 1.8128379692666796,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046689325675925174,
      "loss": 4.9897,
      "mean_token_accuracy": 0.20281611680984496,
      "num_tokens": 40428741.0,
      "step": 23300
    },
    {
      "entropy": 5.759622097015381,
      "epoch": 1.813226998638397,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046687878384711574,
      "loss": 5.198,
      "mean_token_accuracy": 0.18550204932689668,
      "num_tokens": 40438128.0,
      "step": 23305
    },
    {
      "entropy": 5.712845325469971,
      "epoch": 1.8136160280101148,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046686430802352476,
      "loss": 5.1376,
      "mean_token_accuracy": 0.19512750059366227,
      "num_tokens": 40446990.0,
      "step": 23310
    },
    {
      "entropy": 5.647995328903198,
      "epoch": 1.8140050573818325,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004668498292886982,
      "loss": 5.0422,
      "mean_token_accuracy": 0.19947749823331834,
      "num_tokens": 40454627.0,
      "step": 23315
    },
    {
      "entropy": 5.634478950500489,
      "epoch": 1.8143940867535497,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046683534764285577,
      "loss": 5.0716,
      "mean_token_accuracy": 0.2054267942905426,
      "num_tokens": 40463321.0,
      "step": 23320
    },
    {
      "entropy": 5.609573316574097,
      "epoch": 1.8147831161252674,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046682086308621725,
      "loss": 4.9214,
      "mean_token_accuracy": 0.2108432486653328,
      "num_tokens": 40472156.0,
      "step": 23325
    },
    {
      "entropy": 5.643345928192138,
      "epoch": 1.8151721454969851,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004668063756190026,
      "loss": 5.0675,
      "mean_token_accuracy": 0.1969673976302147,
      "num_tokens": 40480938.0,
      "step": 23330
    },
    {
      "entropy": 5.632159280776977,
      "epoch": 1.8155611748687026,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046679188524143136,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20527398139238356,
      "num_tokens": 40489740.0,
      "step": 23335
    },
    {
      "entropy": 5.705085706710816,
      "epoch": 1.81595020424042,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004667773919537235,
      "loss": 5.0025,
      "mean_token_accuracy": 0.2028025820851326,
      "num_tokens": 40498420.0,
      "step": 23340
    },
    {
      "entropy": 5.634736204147339,
      "epoch": 1.8163392336121378,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046676289575609914,
      "loss": 4.9234,
      "mean_token_accuracy": 0.20825934708118438,
      "num_tokens": 40506634.0,
      "step": 23345
    },
    {
      "entropy": 5.678823232650757,
      "epoch": 1.8167282629838553,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046674839664877796,
      "loss": 5.1488,
      "mean_token_accuracy": 0.19508560448884965,
      "num_tokens": 40515310.0,
      "step": 23350
    },
    {
      "entropy": 5.676866865158081,
      "epoch": 1.8171172923555727,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004667338946319801,
      "loss": 5.0091,
      "mean_token_accuracy": 0.20593808293342591,
      "num_tokens": 40523739.0,
      "step": 23355
    },
    {
      "entropy": 5.721298503875732,
      "epoch": 1.8175063217272904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004667193897059255,
      "loss": 5.1639,
      "mean_token_accuracy": 0.19017911404371263,
      "num_tokens": 40533070.0,
      "step": 23360
    },
    {
      "entropy": 5.731092691421509,
      "epoch": 1.8178953510990081,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046670488187083436,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20166377425193788,
      "num_tokens": 40541975.0,
      "step": 23365
    },
    {
      "entropy": 5.725260829925537,
      "epoch": 1.8182843804707254,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004666903711269267,
      "loss": 5.141,
      "mean_token_accuracy": 0.19650277644395828,
      "num_tokens": 40550896.0,
      "step": 23370
    },
    {
      "entropy": 5.635598659515381,
      "epoch": 1.818673409842443,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004666758574744228,
      "loss": 5.1201,
      "mean_token_accuracy": 0.19900337010622024,
      "num_tokens": 40559719.0,
      "step": 23375
    },
    {
      "entropy": 5.709807014465332,
      "epoch": 1.8190624392141608,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004666613409135429,
      "loss": 5.1162,
      "mean_token_accuracy": 0.19977762550115585,
      "num_tokens": 40568448.0,
      "step": 23380
    },
    {
      "entropy": 5.785470867156983,
      "epoch": 1.8194514685858783,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004666468214445072,
      "loss": 5.1308,
      "mean_token_accuracy": 0.19481135606765748,
      "num_tokens": 40576443.0,
      "step": 23385
    },
    {
      "entropy": 5.645090579986572,
      "epoch": 1.8198404979575957,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046663229906753595,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21064617484807968,
      "num_tokens": 40584440.0,
      "step": 23390
    },
    {
      "entropy": 5.655154848098755,
      "epoch": 1.8202295273293134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046661777378284965,
      "loss": 5.172,
      "mean_token_accuracy": 0.19538187235593796,
      "num_tokens": 40593696.0,
      "step": 23395
    },
    {
      "entropy": 5.65268874168396,
      "epoch": 1.820618556701031,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004666032455906685,
      "loss": 5.0023,
      "mean_token_accuracy": 0.2076433554291725,
      "num_tokens": 40601368.0,
      "step": 23400
    },
    {
      "entropy": 5.7503540992736815,
      "epoch": 1.8210075860727484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046658871449121325,
      "loss": 5.086,
      "mean_token_accuracy": 0.2028270110487938,
      "num_tokens": 40609143.0,
      "step": 23405
    },
    {
      "entropy": 5.696317195892334,
      "epoch": 1.821396615444466,
      "grad_norm": 1.125,
      "learning_rate": 0.0004665741804847041,
      "loss": 5.0632,
      "mean_token_accuracy": 0.2003426656126976,
      "num_tokens": 40617602.0,
      "step": 23410
    },
    {
      "entropy": 5.612908935546875,
      "epoch": 1.8217856448161838,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046655964357136164,
      "loss": 4.9612,
      "mean_token_accuracy": 0.20739205926656723,
      "num_tokens": 40626412.0,
      "step": 23415
    },
    {
      "entropy": 5.652625846862793,
      "epoch": 1.822174674187901,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046654510375140657,
      "loss": 5.0202,
      "mean_token_accuracy": 0.206109456717968,
      "num_tokens": 40635095.0,
      "step": 23420
    },
    {
      "entropy": 5.678670167922974,
      "epoch": 1.8225637035596187,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004665305610250594,
      "loss": 5.0384,
      "mean_token_accuracy": 0.20043517649173737,
      "num_tokens": 40643421.0,
      "step": 23425
    },
    {
      "entropy": 5.647865676879883,
      "epoch": 1.8229527329313364,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004665160153925408,
      "loss": 5.0729,
      "mean_token_accuracy": 0.20355336666107177,
      "num_tokens": 40652180.0,
      "step": 23430
    },
    {
      "entropy": 5.735098266601563,
      "epoch": 1.823341762303054,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046650146685407154,
      "loss": 5.1481,
      "mean_token_accuracy": 0.18693066388368607,
      "num_tokens": 40660398.0,
      "step": 23435
    },
    {
      "entropy": 5.714107656478882,
      "epoch": 1.8237307916747714,
      "grad_norm": 1.125,
      "learning_rate": 0.00046648691540987226,
      "loss": 5.1676,
      "mean_token_accuracy": 0.1879961296916008,
      "num_tokens": 40668827.0,
      "step": 23440
    },
    {
      "entropy": 5.688480520248413,
      "epoch": 1.824119821046489,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046647236106016387,
      "loss": 5.0699,
      "mean_token_accuracy": 0.1986662283539772,
      "num_tokens": 40676920.0,
      "step": 23445
    },
    {
      "entropy": 5.670010995864868,
      "epoch": 1.8245088504182065,
      "grad_norm": 1.25,
      "learning_rate": 0.0004664578038051672,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20432327389717103,
      "num_tokens": 40685115.0,
      "step": 23450
    },
    {
      "entropy": 5.676260232925415,
      "epoch": 1.824897879789924,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000466443243645103,
      "loss": 5.039,
      "mean_token_accuracy": 0.1987512916326523,
      "num_tokens": 40693870.0,
      "step": 23455
    },
    {
      "entropy": 5.718591928482056,
      "epoch": 1.8252869091616417,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004664286805801923,
      "loss": 5.022,
      "mean_token_accuracy": 0.194333516061306,
      "num_tokens": 40703031.0,
      "step": 23460
    },
    {
      "entropy": 5.622937917709351,
      "epoch": 1.8256759385333594,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004664141146106562,
      "loss": 4.985,
      "mean_token_accuracy": 0.20394688844680786,
      "num_tokens": 40711690.0,
      "step": 23465
    },
    {
      "entropy": 5.668556261062622,
      "epoch": 1.8260649679050769,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046639954573671547,
      "loss": 5.0804,
      "mean_token_accuracy": 0.1969800263643265,
      "num_tokens": 40720086.0,
      "step": 23470
    },
    {
      "entropy": 5.764101028442383,
      "epoch": 1.8264539972767944,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046638497395859127,
      "loss": 5.1254,
      "mean_token_accuracy": 0.1944834589958191,
      "num_tokens": 40728117.0,
      "step": 23475
    },
    {
      "entropy": 5.727489233016968,
      "epoch": 1.826843026648512,
      "grad_norm": 1.125,
      "learning_rate": 0.0004663703992765047,
      "loss": 5.1156,
      "mean_token_accuracy": 0.1994257539510727,
      "num_tokens": 40736761.0,
      "step": 23480
    },
    {
      "entropy": 5.659313726425171,
      "epoch": 1.8272320560202295,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046635582169067693,
      "loss": 5.0645,
      "mean_token_accuracy": 0.19522079676389695,
      "num_tokens": 40745291.0,
      "step": 23485
    },
    {
      "entropy": 5.743591356277466,
      "epoch": 1.827621085391947,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046634124120132915,
      "loss": 5.0786,
      "mean_token_accuracy": 0.20272588282823562,
      "num_tokens": 40754308.0,
      "step": 23490
    },
    {
      "entropy": 5.633688020706177,
      "epoch": 1.8280101147636647,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004663266578086826,
      "loss": 4.9807,
      "mean_token_accuracy": 0.19849610179662705,
      "num_tokens": 40763046.0,
      "step": 23495
    },
    {
      "entropy": 5.5965883255004885,
      "epoch": 1.8283991441353822,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004663120715129585,
      "loss": 5.0297,
      "mean_token_accuracy": 0.19359540343284606,
      "num_tokens": 40770912.0,
      "step": 23500
    },
    {
      "entropy": 5.704968357086182,
      "epoch": 1.8287881735070997,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004662974823143783,
      "loss": 5.1235,
      "mean_token_accuracy": 0.1852446749806404,
      "num_tokens": 40779748.0,
      "step": 23505
    },
    {
      "entropy": 5.742366933822632,
      "epoch": 1.8291772028788174,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004662828902131632,
      "loss": 5.1075,
      "mean_token_accuracy": 0.195328725874424,
      "num_tokens": 40787732.0,
      "step": 23510
    },
    {
      "entropy": 5.6843358993530275,
      "epoch": 1.829566232250535,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004662682952095348,
      "loss": 5.0101,
      "mean_token_accuracy": 0.20013401806354522,
      "num_tokens": 40796521.0,
      "step": 23515
    },
    {
      "entropy": 5.713075876235962,
      "epoch": 1.8299552616222525,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046625369730371436,
      "loss": 5.1469,
      "mean_token_accuracy": 0.19386830925941467,
      "num_tokens": 40805479.0,
      "step": 23520
    },
    {
      "entropy": 5.654767990112305,
      "epoch": 1.83034429099397,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004662390964959235,
      "loss": 4.9644,
      "mean_token_accuracy": 0.20261916816234588,
      "num_tokens": 40813647.0,
      "step": 23525
    },
    {
      "entropy": 5.60133318901062,
      "epoch": 1.8307333203656877,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046622449278638375,
      "loss": 4.9544,
      "mean_token_accuracy": 0.20324818044900894,
      "num_tokens": 40821753.0,
      "step": 23530
    },
    {
      "entropy": 5.660799503326416,
      "epoch": 1.8311223497374052,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004662098861753167,
      "loss": 5.1023,
      "mean_token_accuracy": 0.20108500570058824,
      "num_tokens": 40830391.0,
      "step": 23535
    },
    {
      "entropy": 5.656362628936767,
      "epoch": 1.8315113791091227,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046619527666294394,
      "loss": 5.0444,
      "mean_token_accuracy": 0.20144785940647125,
      "num_tokens": 40839372.0,
      "step": 23540
    },
    {
      "entropy": 5.70982518196106,
      "epoch": 1.8319004084808403,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004661806642494872,
      "loss": 5.1384,
      "mean_token_accuracy": 0.1974942773580551,
      "num_tokens": 40847945.0,
      "step": 23545
    },
    {
      "entropy": 5.786671686172485,
      "epoch": 1.8322894378525578,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004661660489351681,
      "loss": 5.1726,
      "mean_token_accuracy": 0.19213981181383133,
      "num_tokens": 40857600.0,
      "step": 23550
    },
    {
      "entropy": 5.670353746414184,
      "epoch": 1.8326784672242753,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004661514307202086,
      "loss": 5.0074,
      "mean_token_accuracy": 0.2023478180170059,
      "num_tokens": 40866672.0,
      "step": 23555
    },
    {
      "entropy": 5.6903105735778805,
      "epoch": 1.833067496595993,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046613680960483036,
      "loss": 5.0503,
      "mean_token_accuracy": 0.1918072909116745,
      "num_tokens": 40875240.0,
      "step": 23560
    },
    {
      "entropy": 5.62783727645874,
      "epoch": 1.8334565259677107,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004661221855892552,
      "loss": 5.0031,
      "mean_token_accuracy": 0.19587404429912567,
      "num_tokens": 40883731.0,
      "step": 23565
    },
    {
      "entropy": 5.625275087356568,
      "epoch": 1.8338455553394282,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046610755867370506,
      "loss": 4.9792,
      "mean_token_accuracy": 0.202235047519207,
      "num_tokens": 40892572.0,
      "step": 23570
    },
    {
      "entropy": 5.6437853336334225,
      "epoch": 1.8342345847111456,
      "grad_norm": 1.15625,
      "learning_rate": 0.000466092928858402,
      "loss": 5.0611,
      "mean_token_accuracy": 0.20040431171655654,
      "num_tokens": 40901603.0,
      "step": 23575
    },
    {
      "entropy": 5.6633320331573485,
      "epoch": 1.8346236140828633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046607829614356787,
      "loss": 5.0458,
      "mean_token_accuracy": 0.20046085715293885,
      "num_tokens": 40910502.0,
      "step": 23580
    },
    {
      "entropy": 5.680140495300293,
      "epoch": 1.8350126434545808,
      "grad_norm": 1.0,
      "learning_rate": 0.0004660636605294247,
      "loss": 5.0009,
      "mean_token_accuracy": 0.2056643158197403,
      "num_tokens": 40919858.0,
      "step": 23585
    },
    {
      "entropy": 5.726168966293335,
      "epoch": 1.8354016728262983,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004660490220161946,
      "loss": 5.0369,
      "mean_token_accuracy": 0.201455157995224,
      "num_tokens": 40928795.0,
      "step": 23590
    },
    {
      "entropy": 5.6757513046264645,
      "epoch": 1.835790702198016,
      "grad_norm": 1.125,
      "learning_rate": 0.00046603438060409966,
      "loss": 5.0282,
      "mean_token_accuracy": 0.20020510405302047,
      "num_tokens": 40937240.0,
      "step": 23595
    },
    {
      "entropy": 5.668379831314087,
      "epoch": 1.8361797315697335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000466019736293362,
      "loss": 5.0757,
      "mean_token_accuracy": 0.1974781259894371,
      "num_tokens": 40945974.0,
      "step": 23600
    },
    {
      "entropy": 5.644535923004151,
      "epoch": 1.836568760941451,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046600508908420396,
      "loss": 4.9809,
      "mean_token_accuracy": 0.21096906661987305,
      "num_tokens": 40953553.0,
      "step": 23605
    },
    {
      "entropy": 5.6556085586547855,
      "epoch": 1.8369577903131686,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046599043897684766,
      "loss": 5.0216,
      "mean_token_accuracy": 0.20918246656656264,
      "num_tokens": 40961287.0,
      "step": 23610
    },
    {
      "entropy": 5.617972469329834,
      "epoch": 1.8373468196848863,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004659757859715155,
      "loss": 4.9763,
      "mean_token_accuracy": 0.20169522911310195,
      "num_tokens": 40970266.0,
      "step": 23615
    },
    {
      "entropy": 5.691685390472412,
      "epoch": 1.8377358490566038,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046596113006842975,
      "loss": 5.0141,
      "mean_token_accuracy": 0.20533781200647355,
      "num_tokens": 40978303.0,
      "step": 23620
    },
    {
      "entropy": 5.759164428710937,
      "epoch": 1.8381248784283213,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004659464712678128,
      "loss": 5.0657,
      "mean_token_accuracy": 0.1988115653395653,
      "num_tokens": 40986360.0,
      "step": 23625
    },
    {
      "entropy": 5.641524171829223,
      "epoch": 1.838513907800039,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004659318095698871,
      "loss": 5.1188,
      "mean_token_accuracy": 0.19440290480852127,
      "num_tokens": 40995323.0,
      "step": 23630
    },
    {
      "entropy": 5.668115949630737,
      "epoch": 1.8389029371717565,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000465917144974875,
      "loss": 5.0173,
      "mean_token_accuracy": 0.20526110827922822,
      "num_tokens": 41004032.0,
      "step": 23635
    },
    {
      "entropy": 5.694147968292237,
      "epoch": 1.839291966543474,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004659024774829992,
      "loss": 5.049,
      "mean_token_accuracy": 0.19918417781591416,
      "num_tokens": 41011992.0,
      "step": 23640
    },
    {
      "entropy": 5.699076128005982,
      "epoch": 1.8396809959151916,
      "grad_norm": 1.203125,
      "learning_rate": 0.000465887807094482,
      "loss": 5.0863,
      "mean_token_accuracy": 0.1953320950269699,
      "num_tokens": 41020252.0,
      "step": 23645
    },
    {
      "entropy": 5.683829641342163,
      "epoch": 1.8400700252869093,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046587313380954635,
      "loss": 4.9953,
      "mean_token_accuracy": 0.19899878799915313,
      "num_tokens": 41028858.0,
      "step": 23650
    },
    {
      "entropy": 5.691145896911621,
      "epoch": 1.8404590546586266,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046585845762841453,
      "loss": 4.973,
      "mean_token_accuracy": 0.20388407856225968,
      "num_tokens": 41037539.0,
      "step": 23655
    },
    {
      "entropy": 5.663805770874023,
      "epoch": 1.8408480840303443,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004658437785513095,
      "loss": 5.081,
      "mean_token_accuracy": 0.19854643791913987,
      "num_tokens": 41045809.0,
      "step": 23660
    },
    {
      "entropy": 5.607454776763916,
      "epoch": 1.841237113402062,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004658290965784539,
      "loss": 4.9958,
      "mean_token_accuracy": 0.2029963865876198,
      "num_tokens": 41054675.0,
      "step": 23665
    },
    {
      "entropy": 5.711282205581665,
      "epoch": 1.8416261427737795,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004658144117100704,
      "loss": 5.1612,
      "mean_token_accuracy": 0.19147171080112457,
      "num_tokens": 41063493.0,
      "step": 23670
    },
    {
      "entropy": 5.699791812896729,
      "epoch": 1.842015172145497,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046579972394638204,
      "loss": 5.0894,
      "mean_token_accuracy": 0.1946207493543625,
      "num_tokens": 41072985.0,
      "step": 23675
    },
    {
      "entropy": 5.691910934448242,
      "epoch": 1.8424042015172146,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046578503328761146,
      "loss": 5.0647,
      "mean_token_accuracy": 0.20021508634090424,
      "num_tokens": 41080926.0,
      "step": 23680
    },
    {
      "entropy": 5.601751804351807,
      "epoch": 1.842793230888932,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046577033973398173,
      "loss": 5.0031,
      "mean_token_accuracy": 0.20392018258571626,
      "num_tokens": 41090251.0,
      "step": 23685
    },
    {
      "entropy": 5.62412109375,
      "epoch": 1.8431822602606496,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004657556432857157,
      "loss": 4.9653,
      "mean_token_accuracy": 0.20614240914583207,
      "num_tokens": 41098681.0,
      "step": 23690
    },
    {
      "entropy": 5.670771551132202,
      "epoch": 1.8435712896323673,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004657409439430364,
      "loss": 5.0977,
      "mean_token_accuracy": 0.2012905165553093,
      "num_tokens": 41107399.0,
      "step": 23695
    },
    {
      "entropy": 5.712767457962036,
      "epoch": 1.843960319004085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004657262417061669,
      "loss": 5.048,
      "mean_token_accuracy": 0.19650247246026992,
      "num_tokens": 41116476.0,
      "step": 23700
    },
    {
      "entropy": 5.627057886123657,
      "epoch": 1.8443493483758022,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004657115365753302,
      "loss": 4.9791,
      "mean_token_accuracy": 0.19985273480415344,
      "num_tokens": 41125191.0,
      "step": 23705
    },
    {
      "entropy": 5.624863481521606,
      "epoch": 1.84473837774752,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046569682855074953,
      "loss": 4.9925,
      "mean_token_accuracy": 0.2109066978096962,
      "num_tokens": 41133811.0,
      "step": 23710
    },
    {
      "entropy": 5.6927660465240475,
      "epoch": 1.8451274071192376,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046568211763264796,
      "loss": 5.1323,
      "mean_token_accuracy": 0.19276395440101624,
      "num_tokens": 41143093.0,
      "step": 23715
    },
    {
      "entropy": 5.762273597717285,
      "epoch": 1.845516436490955,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004656674038212488,
      "loss": 5.1412,
      "mean_token_accuracy": 0.19303966015577317,
      "num_tokens": 41152106.0,
      "step": 23720
    },
    {
      "entropy": 5.684381103515625,
      "epoch": 1.8459054658626726,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046565268711677525,
      "loss": 5.0205,
      "mean_token_accuracy": 0.19906811267137528,
      "num_tokens": 41161100.0,
      "step": 23725
    },
    {
      "entropy": 5.6614970684051515,
      "epoch": 1.8462944952343903,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046563796751945065,
      "loss": 5.057,
      "mean_token_accuracy": 0.19930259585380555,
      "num_tokens": 41170339.0,
      "step": 23730
    },
    {
      "entropy": 5.714292240142822,
      "epoch": 1.8466835246061077,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046562324502949834,
      "loss": 5.0549,
      "mean_token_accuracy": 0.19824913889169693,
      "num_tokens": 41178821.0,
      "step": 23735
    },
    {
      "entropy": 5.719708728790283,
      "epoch": 1.8470725539778252,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004656085196471416,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19827462434768678,
      "num_tokens": 41187578.0,
      "step": 23740
    },
    {
      "entropy": 5.545157957077026,
      "epoch": 1.847461583349543,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046559379137260404,
      "loss": 4.9357,
      "mean_token_accuracy": 0.20713798701763153,
      "num_tokens": 41196799.0,
      "step": 23745
    },
    {
      "entropy": 5.650405645370483,
      "epoch": 1.8478506127212606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000465579060206109,
      "loss": 5.1154,
      "mean_token_accuracy": 0.19511368423700332,
      "num_tokens": 41206300.0,
      "step": 23750
    },
    {
      "entropy": 5.766040182113647,
      "epoch": 1.8482396420929779,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046556432614788015,
      "loss": 5.1069,
      "mean_token_accuracy": 0.19853024184703827,
      "num_tokens": 41214377.0,
      "step": 23755
    },
    {
      "entropy": 5.676653671264648,
      "epoch": 1.8486286714646956,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004655495891981409,
      "loss": 5.0063,
      "mean_token_accuracy": 0.19895326644182204,
      "num_tokens": 41223303.0,
      "step": 23760
    },
    {
      "entropy": 5.687079858779907,
      "epoch": 1.8490177008364133,
      "grad_norm": 1.09375,
      "learning_rate": 0.000465534849357115,
      "loss": 5.1443,
      "mean_token_accuracy": 0.19204787611961366,
      "num_tokens": 41232323.0,
      "step": 23765
    },
    {
      "entropy": 5.726064872741699,
      "epoch": 1.8494067302081307,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046552010662502595,
      "loss": 5.1038,
      "mean_token_accuracy": 0.1950084537267685,
      "num_tokens": 41241605.0,
      "step": 23770
    },
    {
      "entropy": 5.756454133987427,
      "epoch": 1.8497957595798482,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004655053610020976,
      "loss": 5.174,
      "mean_token_accuracy": 0.18778659105300904,
      "num_tokens": 41250604.0,
      "step": 23775
    },
    {
      "entropy": 5.65646448135376,
      "epoch": 1.850184788951566,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004654906124885535,
      "loss": 4.9695,
      "mean_token_accuracy": 0.20573434829711915,
      "num_tokens": 41258640.0,
      "step": 23780
    },
    {
      "entropy": 5.657597780227661,
      "epoch": 1.8505738183232834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004654758610846176,
      "loss": 5.1129,
      "mean_token_accuracy": 0.20204756408929825,
      "num_tokens": 41268104.0,
      "step": 23785
    },
    {
      "entropy": 5.690503025054932,
      "epoch": 1.8509628476950009,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004654611067905137,
      "loss": 5.1337,
      "mean_token_accuracy": 0.1977001339197159,
      "num_tokens": 41276276.0,
      "step": 23790
    },
    {
      "entropy": 5.688591146469117,
      "epoch": 1.8513518770667186,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046544634960646563,
      "loss": 4.9828,
      "mean_token_accuracy": 0.2058488130569458,
      "num_tokens": 41284464.0,
      "step": 23795
    },
    {
      "entropy": 5.679538917541504,
      "epoch": 1.8517409064384363,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004654315895326974,
      "loss": 4.9948,
      "mean_token_accuracy": 0.1992371916770935,
      "num_tokens": 41293108.0,
      "step": 23800
    },
    {
      "entropy": 5.658460426330566,
      "epoch": 1.8521299358101535,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004654168265694328,
      "loss": 5.0377,
      "mean_token_accuracy": 0.19509689807891845,
      "num_tokens": 41301830.0,
      "step": 23805
    },
    {
      "entropy": 5.7098856449127195,
      "epoch": 1.8525189651818712,
      "grad_norm": 1.203125,
      "learning_rate": 0.000465402060716896,
      "loss": 5.0706,
      "mean_token_accuracy": 0.19837660789489747,
      "num_tokens": 41310487.0,
      "step": 23810
    },
    {
      "entropy": 5.735160303115845,
      "epoch": 1.852907994553589,
      "grad_norm": 1.140625,
      "learning_rate": 0.000465387291975311,
      "loss": 5.1457,
      "mean_token_accuracy": 0.19527142941951753,
      "num_tokens": 41319730.0,
      "step": 23815
    },
    {
      "entropy": 5.6724484920501705,
      "epoch": 1.8532970239253064,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046537252034490185,
      "loss": 5.0451,
      "mean_token_accuracy": 0.20284042954444886,
      "num_tokens": 41328914.0,
      "step": 23820
    },
    {
      "entropy": 5.647346448898316,
      "epoch": 1.8536860532970239,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046535774582589275,
      "loss": 5.0819,
      "mean_token_accuracy": 0.19659153819084169,
      "num_tokens": 41337060.0,
      "step": 23825
    },
    {
      "entropy": 5.684302282333374,
      "epoch": 1.8540750826687415,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046534296841850776,
      "loss": 5.0051,
      "mean_token_accuracy": 0.20757822394371034,
      "num_tokens": 41344966.0,
      "step": 23830
    },
    {
      "entropy": 5.608697366714478,
      "epoch": 1.854464112040459,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046532818812297124,
      "loss": 4.926,
      "mean_token_accuracy": 0.20916413962841035,
      "num_tokens": 41353678.0,
      "step": 23835
    },
    {
      "entropy": 5.745792627334595,
      "epoch": 1.8548531414121765,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004653134049395074,
      "loss": 5.1785,
      "mean_token_accuracy": 0.20162003636360168,
      "num_tokens": 41362782.0,
      "step": 23840
    },
    {
      "entropy": 5.646778774261475,
      "epoch": 1.8552421707838942,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004652986188683406,
      "loss": 5.0171,
      "mean_token_accuracy": 0.20401924699544907,
      "num_tokens": 41370668.0,
      "step": 23845
    },
    {
      "entropy": 5.611280870437622,
      "epoch": 1.855631200155612,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046528382990969504,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2035282790660858,
      "num_tokens": 41378945.0,
      "step": 23850
    },
    {
      "entropy": 5.7154212474823,
      "epoch": 1.8560202295273294,
      "grad_norm": 1.125,
      "learning_rate": 0.0004652690380637953,
      "loss": 5.0693,
      "mean_token_accuracy": 0.1947337046265602,
      "num_tokens": 41387579.0,
      "step": 23855
    },
    {
      "entropy": 5.696177768707275,
      "epoch": 1.8564092588990468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004652542433308657,
      "loss": 5.0718,
      "mean_token_accuracy": 0.2017604887485504,
      "num_tokens": 41396472.0,
      "step": 23860
    },
    {
      "entropy": 5.7154511451721195,
      "epoch": 1.8567982882707645,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004652394457111309,
      "loss": 5.1161,
      "mean_token_accuracy": 0.19253923296928405,
      "num_tokens": 41404599.0,
      "step": 23865
    },
    {
      "entropy": 5.739647674560547,
      "epoch": 1.857187317642482,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046522464520481517,
      "loss": 5.1399,
      "mean_token_accuracy": 0.18762982040643691,
      "num_tokens": 41412799.0,
      "step": 23870
    },
    {
      "entropy": 5.64436707496643,
      "epoch": 1.8575763470141995,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046520984181214333,
      "loss": 4.9524,
      "mean_token_accuracy": 0.20920240730047227,
      "num_tokens": 41421193.0,
      "step": 23875
    },
    {
      "entropy": 5.665311288833618,
      "epoch": 1.8579653763859172,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004651950355333398,
      "loss": 5.0175,
      "mean_token_accuracy": 0.19525519758462906,
      "num_tokens": 41430201.0,
      "step": 23880
    },
    {
      "entropy": 5.605517768859864,
      "epoch": 1.8583544057576347,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004651802263686294,
      "loss": 4.9876,
      "mean_token_accuracy": 0.19989306330680848,
      "num_tokens": 41439122.0,
      "step": 23885
    },
    {
      "entropy": 5.671520757675171,
      "epoch": 1.8587434351293521,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004651654143182367,
      "loss": 5.05,
      "mean_token_accuracy": 0.19741638153791427,
      "num_tokens": 41448553.0,
      "step": 23890
    },
    {
      "entropy": 5.651991558074951,
      "epoch": 1.8591324645010698,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004651505993823866,
      "loss": 5.0183,
      "mean_token_accuracy": 0.2022734746336937,
      "num_tokens": 41457098.0,
      "step": 23895
    },
    {
      "entropy": 5.669555187225342,
      "epoch": 1.8595214938727875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046513578156130383,
      "loss": 5.075,
      "mean_token_accuracy": 0.19775795042514802,
      "num_tokens": 41465714.0,
      "step": 23900
    },
    {
      "entropy": 5.747840929031372,
      "epoch": 1.859910523244505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004651209608552131,
      "loss": 5.0953,
      "mean_token_accuracy": 0.1980939269065857,
      "num_tokens": 41474418.0,
      "step": 23905
    },
    {
      "entropy": 5.645845603942871,
      "epoch": 1.8602995526162225,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046510613726433945,
      "loss": 4.9773,
      "mean_token_accuracy": 0.2044762283563614,
      "num_tokens": 41482430.0,
      "step": 23910
    },
    {
      "entropy": 5.711126470565796,
      "epoch": 1.8606885819879402,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004650913107889078,
      "loss": 5.1362,
      "mean_token_accuracy": 0.19826145470142365,
      "num_tokens": 41490759.0,
      "step": 23915
    },
    {
      "entropy": 5.594354343414307,
      "epoch": 1.8610776113596577,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000465076481429143,
      "loss": 4.9887,
      "mean_token_accuracy": 0.20265743881464005,
      "num_tokens": 41499660.0,
      "step": 23920
    },
    {
      "entropy": 5.68090500831604,
      "epoch": 1.8614666407313751,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004650616491852701,
      "loss": 5.0168,
      "mean_token_accuracy": 0.20643742382526398,
      "num_tokens": 41507515.0,
      "step": 23925
    },
    {
      "entropy": 5.722912311553955,
      "epoch": 1.8618556701030928,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046504681405751426,
      "loss": 5.08,
      "mean_token_accuracy": 0.19374623000621796,
      "num_tokens": 41515859.0,
      "step": 23930
    },
    {
      "entropy": 5.685977268218994,
      "epoch": 1.8622446994748103,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046503197604610047,
      "loss": 5.1096,
      "mean_token_accuracy": 0.20370736271142958,
      "num_tokens": 41524776.0,
      "step": 23935
    },
    {
      "entropy": 5.701795148849487,
      "epoch": 1.8626337288465278,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046501713515125393,
      "loss": 5.0243,
      "mean_token_accuracy": 0.20092787891626357,
      "num_tokens": 41532854.0,
      "step": 23940
    },
    {
      "entropy": 5.72903528213501,
      "epoch": 1.8630227582182455,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004650022913731998,
      "loss": 5.0099,
      "mean_token_accuracy": 0.21019555926322936,
      "num_tokens": 41542038.0,
      "step": 23945
    },
    {
      "entropy": 5.686867141723633,
      "epoch": 1.8634117875899632,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046498744471216335,
      "loss": 5.022,
      "mean_token_accuracy": 0.20487914979457855,
      "num_tokens": 41550428.0,
      "step": 23950
    },
    {
      "entropy": 5.635628890991211,
      "epoch": 1.8638008169616807,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046497259516836974,
      "loss": 4.9958,
      "mean_token_accuracy": 0.20121567100286483,
      "num_tokens": 41559289.0,
      "step": 23955
    },
    {
      "entropy": 5.691625165939331,
      "epoch": 1.8641898463333981,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046495774274204446,
      "loss": 5.0109,
      "mean_token_accuracy": 0.20548161417245864,
      "num_tokens": 41567901.0,
      "step": 23960
    },
    {
      "entropy": 5.760720205307007,
      "epoch": 1.8645788757051158,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004649428874334127,
      "loss": 5.1296,
      "mean_token_accuracy": 0.1914092168211937,
      "num_tokens": 41577589.0,
      "step": 23965
    },
    {
      "entropy": 5.736770391464233,
      "epoch": 1.8649679050768333,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004649280292427,
      "loss": 5.1248,
      "mean_token_accuracy": 0.19287013858556748,
      "num_tokens": 41586319.0,
      "step": 23970
    },
    {
      "entropy": 5.736889457702636,
      "epoch": 1.8653569344485508,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004649131681701318,
      "loss": 5.1106,
      "mean_token_accuracy": 0.19541349709033967,
      "num_tokens": 41594316.0,
      "step": 23975
    },
    {
      "entropy": 5.660229873657227,
      "epoch": 1.8657459638202685,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046489830421593347,
      "loss": 5.0369,
      "mean_token_accuracy": 0.19791144132614136,
      "num_tokens": 41603221.0,
      "step": 23980
    },
    {
      "entropy": 5.64997034072876,
      "epoch": 1.866134993191986,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004648834373803307,
      "loss": 5.0503,
      "mean_token_accuracy": 0.19683537185192107,
      "num_tokens": 41612082.0,
      "step": 23985
    },
    {
      "entropy": 5.710425853729248,
      "epoch": 1.8665240225637034,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046486856766354893,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2030527263879776,
      "num_tokens": 41619762.0,
      "step": 23990
    },
    {
      "entropy": 5.636745882034302,
      "epoch": 1.8669130519354211,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046485369506581387,
      "loss": 5.0095,
      "mean_token_accuracy": 0.19832469820976256,
      "num_tokens": 41627968.0,
      "step": 23995
    },
    {
      "entropy": 5.69230694770813,
      "epoch": 1.8673020813071388,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046483881958735124,
      "loss": 5.0917,
      "mean_token_accuracy": 0.19699103087186814,
      "num_tokens": 41636841.0,
      "step": 24000
    },
    {
      "epoch": 1.8673020813071388,
      "eval_entropy": 5.483582375758651,
      "eval_loss": 5.10653018951416,
      "eval_mean_token_accuracy": 0.20658851980273832,
      "eval_num_tokens": 41636841.0,
      "eval_runtime": 21.8321,
      "eval_samples_per_second": 1903.529,
      "eval_steps_per_second": 237.953,
      "step": 24000
    },
    {
      "entropy": 5.741102457046509,
      "epoch": 1.8676911106788563,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046482394122838663,
      "loss": 5.0506,
      "mean_token_accuracy": 0.2047205686569214,
      "num_tokens": 41645352.0,
      "step": 24005
    },
    {
      "entropy": 5.660536670684815,
      "epoch": 1.8680801400505738,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046480905998914587,
      "loss": 4.9577,
      "mean_token_accuracy": 0.20603823363780976,
      "num_tokens": 41654296.0,
      "step": 24010
    },
    {
      "entropy": 5.590140914916992,
      "epoch": 1.8684691694222915,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004647941758698547,
      "loss": 5.0156,
      "mean_token_accuracy": 0.1979769691824913,
      "num_tokens": 41662008.0,
      "step": 24015
    },
    {
      "entropy": 5.7127727508544925,
      "epoch": 1.868858198794009,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000464779288870739,
      "loss": 5.1653,
      "mean_token_accuracy": 0.1959889277815819,
      "num_tokens": 41671765.0,
      "step": 24020
    },
    {
      "entropy": 5.760813999176025,
      "epoch": 1.8692472281657264,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046476439899202464,
      "loss": 5.1483,
      "mean_token_accuracy": 0.1913057327270508,
      "num_tokens": 41681035.0,
      "step": 24025
    },
    {
      "entropy": 5.700266933441162,
      "epoch": 1.8696362575374441,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046474950623393756,
      "loss": 5.0134,
      "mean_token_accuracy": 0.20269862115383147,
      "num_tokens": 41689974.0,
      "step": 24030
    },
    {
      "entropy": 5.692906045913697,
      "epoch": 1.8700252869091616,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004647346105967038,
      "loss": 5.0764,
      "mean_token_accuracy": 0.19065997898578643,
      "num_tokens": 41698560.0,
      "step": 24035
    },
    {
      "entropy": 5.697722101211548,
      "epoch": 1.870414316280879,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004647197120805493,
      "loss": 5.0303,
      "mean_token_accuracy": 0.20352455228567123,
      "num_tokens": 41706741.0,
      "step": 24040
    },
    {
      "entropy": 5.6599808692932125,
      "epoch": 1.8708033456525968,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046470481068570013,
      "loss": 5.0118,
      "mean_token_accuracy": 0.19611548781394958,
      "num_tokens": 41715685.0,
      "step": 24045
    },
    {
      "entropy": 5.648841094970703,
      "epoch": 1.8711923750243145,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004646899064123824,
      "loss": 5.0843,
      "mean_token_accuracy": 0.2012360453605652,
      "num_tokens": 41724123.0,
      "step": 24050
    },
    {
      "entropy": 5.7200220108032225,
      "epoch": 1.871581404396032,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004646749992608223,
      "loss": 5.0574,
      "mean_token_accuracy": 0.20124853700399398,
      "num_tokens": 41732604.0,
      "step": 24055
    },
    {
      "entropy": 5.630925798416138,
      "epoch": 1.8719704337677494,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004646600892312459,
      "loss": 5.016,
      "mean_token_accuracy": 0.2035152032971382,
      "num_tokens": 41740444.0,
      "step": 24060
    },
    {
      "entropy": 5.654859256744385,
      "epoch": 1.872359463139467,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004646451763238796,
      "loss": 5.0105,
      "mean_token_accuracy": 0.19737775027751922,
      "num_tokens": 41748800.0,
      "step": 24065
    },
    {
      "entropy": 5.648532152175903,
      "epoch": 1.8727484925111846,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004646302605389496,
      "loss": 5.0589,
      "mean_token_accuracy": 0.19410253912210465,
      "num_tokens": 41757402.0,
      "step": 24070
    },
    {
      "entropy": 5.699870777130127,
      "epoch": 1.873137521882902,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004646153418766822,
      "loss": 5.0262,
      "mean_token_accuracy": 0.20737480372190475,
      "num_tokens": 41766604.0,
      "step": 24075
    },
    {
      "entropy": 5.756013679504394,
      "epoch": 1.8735265512546198,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046460042033730377,
      "loss": 5.2053,
      "mean_token_accuracy": 0.19025604277849198,
      "num_tokens": 41775450.0,
      "step": 24080
    },
    {
      "entropy": 5.7421306610107425,
      "epoch": 1.8739155806263375,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004645854959210408,
      "loss": 5.1726,
      "mean_token_accuracy": 0.19606088250875472,
      "num_tokens": 41784355.0,
      "step": 24085
    },
    {
      "entropy": 5.655627822875976,
      "epoch": 1.8743046099980547,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046457056862811956,
      "loss": 4.999,
      "mean_token_accuracy": 0.19285477101802825,
      "num_tokens": 41793441.0,
      "step": 24090
    },
    {
      "entropy": 5.672564315795898,
      "epoch": 1.8746936393697724,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004645556384587668,
      "loss": 4.9977,
      "mean_token_accuracy": 0.19842258393764495,
      "num_tokens": 41802407.0,
      "step": 24095
    },
    {
      "entropy": 5.700974321365356,
      "epoch": 1.87508266874149,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004645407054132089,
      "loss": 5.0544,
      "mean_token_accuracy": 0.1979052484035492,
      "num_tokens": 41810362.0,
      "step": 24100
    },
    {
      "entropy": 5.676897716522217,
      "epoch": 1.8754716981132076,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004645257694916725,
      "loss": 5.1575,
      "mean_token_accuracy": 0.19177428483963013,
      "num_tokens": 41819332.0,
      "step": 24105
    },
    {
      "entropy": 5.698676872253418,
      "epoch": 1.875860727484925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004645108306943842,
      "loss": 5.0183,
      "mean_token_accuracy": 0.20641182065010072,
      "num_tokens": 41828048.0,
      "step": 24110
    },
    {
      "entropy": 5.636488342285157,
      "epoch": 1.8762497568566427,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004644958890215707,
      "loss": 4.9914,
      "mean_token_accuracy": 0.20486603081226348,
      "num_tokens": 41836583.0,
      "step": 24115
    },
    {
      "entropy": 5.660585546493531,
      "epoch": 1.8766387862283602,
      "grad_norm": 1.25,
      "learning_rate": 0.0004644809444734586,
      "loss": 5.0557,
      "mean_token_accuracy": 0.19707500636577607,
      "num_tokens": 41845454.0,
      "step": 24120
    },
    {
      "entropy": 5.700658845901489,
      "epoch": 1.8770278156000777,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046446599705027487,
      "loss": 5.0334,
      "mean_token_accuracy": 0.19468516558408738,
      "num_tokens": 41854481.0,
      "step": 24125
    },
    {
      "entropy": 5.730928039550781,
      "epoch": 1.8774168449717954,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046445104675224607,
      "loss": 5.0904,
      "mean_token_accuracy": 0.2011136829853058,
      "num_tokens": 41863015.0,
      "step": 24130
    },
    {
      "entropy": 5.723241806030273,
      "epoch": 1.877805874343513,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004644360935795993,
      "loss": 5.0601,
      "mean_token_accuracy": 0.19522613734006883,
      "num_tokens": 41871703.0,
      "step": 24135
    },
    {
      "entropy": 5.7228978157043455,
      "epoch": 1.8781949037152303,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046442113753256126,
      "loss": 5.0013,
      "mean_token_accuracy": 0.19627539813518524,
      "num_tokens": 41880136.0,
      "step": 24140
    },
    {
      "entropy": 5.722740173339844,
      "epoch": 1.878583933086948,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046440617861135905,
      "loss": 5.1315,
      "mean_token_accuracy": 0.19552311301231384,
      "num_tokens": 41889065.0,
      "step": 24145
    },
    {
      "entropy": 5.618831777572632,
      "epoch": 1.8789729624586657,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004643912168162194,
      "loss": 4.9717,
      "mean_token_accuracy": 0.20740085244178771,
      "num_tokens": 41898160.0,
      "step": 24150
    },
    {
      "entropy": 5.725120782852173,
      "epoch": 1.8793619918303832,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004643762521473696,
      "loss": 5.0554,
      "mean_token_accuracy": 0.2016186848282814,
      "num_tokens": 41907037.0,
      "step": 24155
    },
    {
      "entropy": 5.724745416641236,
      "epoch": 1.8797510212021007,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004643612846050366,
      "loss": 5.1157,
      "mean_token_accuracy": 0.1896385058760643,
      "num_tokens": 41915176.0,
      "step": 24160
    },
    {
      "entropy": 5.728715610504151,
      "epoch": 1.8801400505738184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046434631418944744,
      "loss": 5.1057,
      "mean_token_accuracy": 0.19158462584018707,
      "num_tokens": 41923586.0,
      "step": 24165
    },
    {
      "entropy": 5.703259468078613,
      "epoch": 1.8805290799455359,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004643313409008294,
      "loss": 5.1898,
      "mean_token_accuracy": 0.18386437743902206,
      "num_tokens": 41932803.0,
      "step": 24170
    },
    {
      "entropy": 5.718877363204956,
      "epoch": 1.8809181093172533,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046431636473940956,
      "loss": 5.0844,
      "mean_token_accuracy": 0.19435103833675385,
      "num_tokens": 41941737.0,
      "step": 24175
    },
    {
      "entropy": 5.6895629405975345,
      "epoch": 1.881307138688971,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004643013857054152,
      "loss": 4.999,
      "mean_token_accuracy": 0.2022035926580429,
      "num_tokens": 41950099.0,
      "step": 24180
    },
    {
      "entropy": 5.658486413955688,
      "epoch": 1.8816961680606887,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046428640379907367,
      "loss": 5.0662,
      "mean_token_accuracy": 0.19750165045261384,
      "num_tokens": 41959091.0,
      "step": 24185
    },
    {
      "entropy": 5.728923988342285,
      "epoch": 1.882085197432406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046427141902061225,
      "loss": 4.9813,
      "mean_token_accuracy": 0.20358682423830032,
      "num_tokens": 41968462.0,
      "step": 24190
    },
    {
      "entropy": 5.680650281906128,
      "epoch": 1.8824742268041237,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004642564313702582,
      "loss": 5.0571,
      "mean_token_accuracy": 0.19784417897462844,
      "num_tokens": 41976514.0,
      "step": 24195
    },
    {
      "entropy": 5.626406717300415,
      "epoch": 1.8828632561758414,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046424144084823916,
      "loss": 4.9735,
      "mean_token_accuracy": 0.2012020990252495,
      "num_tokens": 41985583.0,
      "step": 24200
    },
    {
      "entropy": 5.643293714523315,
      "epoch": 1.8832522855475589,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004642264474547824,
      "loss": 4.9842,
      "mean_token_accuracy": 0.20644300431013107,
      "num_tokens": 41993684.0,
      "step": 24205
    },
    {
      "entropy": 5.641506958007812,
      "epoch": 1.8836413149192763,
      "grad_norm": 1.125,
      "learning_rate": 0.00046421145119011556,
      "loss": 4.9498,
      "mean_token_accuracy": 0.20412273854017257,
      "num_tokens": 42002239.0,
      "step": 24210
    },
    {
      "entropy": 5.5814532279968265,
      "epoch": 1.884030344290994,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000464196452054466,
      "loss": 4.9239,
      "mean_token_accuracy": 0.20882173329591752,
      "num_tokens": 42010886.0,
      "step": 24215
    },
    {
      "entropy": 5.678795433044433,
      "epoch": 1.8844193736627115,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004641814500480615,
      "loss": 5.1051,
      "mean_token_accuracy": 0.1926621153950691,
      "num_tokens": 42019703.0,
      "step": 24220
    },
    {
      "entropy": 5.7432952404022215,
      "epoch": 1.884808403034429,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004641664451711296,
      "loss": 5.158,
      "mean_token_accuracy": 0.19239576607942582,
      "num_tokens": 42028467.0,
      "step": 24225
    },
    {
      "entropy": 5.684470176696777,
      "epoch": 1.8851974324061467,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046415143742389793,
      "loss": 5.0396,
      "mean_token_accuracy": 0.1998426377773285,
      "num_tokens": 42037220.0,
      "step": 24230
    },
    {
      "entropy": 5.706933927536011,
      "epoch": 1.8855864617778644,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004641364268065943,
      "loss": 5.0975,
      "mean_token_accuracy": 0.1953598141670227,
      "num_tokens": 42046034.0,
      "step": 24235
    },
    {
      "entropy": 5.654673051834107,
      "epoch": 1.8859754911495816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046412141331944644,
      "loss": 5.0442,
      "mean_token_accuracy": 0.1991073340177536,
      "num_tokens": 42054659.0,
      "step": 24240
    },
    {
      "entropy": 5.709597110748291,
      "epoch": 1.8863645205212993,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046410639696268206,
      "loss": 5.0999,
      "mean_token_accuracy": 0.1921329081058502,
      "num_tokens": 42063130.0,
      "step": 24245
    },
    {
      "entropy": 5.702262830734253,
      "epoch": 1.886753549893017,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004640913777365292,
      "loss": 5.1088,
      "mean_token_accuracy": 0.19227679818868637,
      "num_tokens": 42071162.0,
      "step": 24250
    },
    {
      "entropy": 5.678764772415161,
      "epoch": 1.8871425792647345,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046407635564121565,
      "loss": 5.1064,
      "mean_token_accuracy": 0.19842315167188646,
      "num_tokens": 42079948.0,
      "step": 24255
    },
    {
      "entropy": 5.7053149223327635,
      "epoch": 1.887531608636452,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046406133067696936,
      "loss": 5.0567,
      "mean_token_accuracy": 0.19605812579393386,
      "num_tokens": 42089010.0,
      "step": 24260
    },
    {
      "entropy": 5.759976196289062,
      "epoch": 1.8879206380081697,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004640463028440182,
      "loss": 5.0605,
      "mean_token_accuracy": 0.19377840757369996,
      "num_tokens": 42097096.0,
      "step": 24265
    },
    {
      "entropy": 5.616702318191528,
      "epoch": 1.8883096673798871,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004640312721425904,
      "loss": 4.9776,
      "mean_token_accuracy": 0.19980015903711318,
      "num_tokens": 42105870.0,
      "step": 24270
    },
    {
      "entropy": 5.638563919067383,
      "epoch": 1.8886986967516046,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004640162385729139,
      "loss": 4.9755,
      "mean_token_accuracy": 0.21114583760499955,
      "num_tokens": 42114246.0,
      "step": 24275
    },
    {
      "entropy": 5.624612712860108,
      "epoch": 1.8890877261233223,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004640012021352168,
      "loss": 4.9844,
      "mean_token_accuracy": 0.19922612458467484,
      "num_tokens": 42122163.0,
      "step": 24280
    },
    {
      "entropy": 5.659454774856568,
      "epoch": 1.88947675549504,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004639861628297273,
      "loss": 5.0182,
      "mean_token_accuracy": 0.2026349723339081,
      "num_tokens": 42130255.0,
      "step": 24285
    },
    {
      "entropy": 5.734159326553344,
      "epoch": 1.8898657848667575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046397112065667354,
      "loss": 5.0919,
      "mean_token_accuracy": 0.19065866768360137,
      "num_tokens": 42138615.0,
      "step": 24290
    },
    {
      "entropy": 5.68282380104065,
      "epoch": 1.890254814238475,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046395607561628387,
      "loss": 4.9794,
      "mean_token_accuracy": 0.20117951631546022,
      "num_tokens": 42146502.0,
      "step": 24295
    },
    {
      "entropy": 5.668199634552002,
      "epoch": 1.8906438436101927,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046394102770878643,
      "loss": 5.0517,
      "mean_token_accuracy": 0.2024393394589424,
      "num_tokens": 42155548.0,
      "step": 24300
    },
    {
      "entropy": 5.560957574844361,
      "epoch": 1.8910328729819101,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046392597693440974,
      "loss": 4.8722,
      "mean_token_accuracy": 0.20830923765897752,
      "num_tokens": 42163813.0,
      "step": 24305
    },
    {
      "entropy": 5.743336915969849,
      "epoch": 1.8914219023536276,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004639109232933819,
      "loss": 5.0829,
      "mean_token_accuracy": 0.1985733166337013,
      "num_tokens": 42172702.0,
      "step": 24310
    },
    {
      "entropy": 5.735151672363282,
      "epoch": 1.8918109317253453,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004638958667859316,
      "loss": 5.1321,
      "mean_token_accuracy": 0.20116187632083893,
      "num_tokens": 42180776.0,
      "step": 24315
    },
    {
      "entropy": 5.71647367477417,
      "epoch": 1.8921999610970628,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004638808074122872,
      "loss": 5.0649,
      "mean_token_accuracy": 0.19686400294303893,
      "num_tokens": 42189695.0,
      "step": 24320
    },
    {
      "entropy": 5.66493821144104,
      "epoch": 1.8925889904687803,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004638657451726771,
      "loss": 5.0428,
      "mean_token_accuracy": 0.19719133228063584,
      "num_tokens": 42197884.0,
      "step": 24325
    },
    {
      "entropy": 5.744034194946289,
      "epoch": 1.892978019840498,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046385068006732995,
      "loss": 5.1461,
      "mean_token_accuracy": 0.1902386024594307,
      "num_tokens": 42206800.0,
      "step": 24330
    },
    {
      "entropy": 5.702206563949585,
      "epoch": 1.8933670492122157,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046383561209647437,
      "loss": 5.0202,
      "mean_token_accuracy": 0.21366167962551116,
      "num_tokens": 42216091.0,
      "step": 24335
    },
    {
      "entropy": 5.629003572463989,
      "epoch": 1.8937560785839331,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046382054126033884,
      "loss": 5.0525,
      "mean_token_accuracy": 0.19962407499551774,
      "num_tokens": 42224675.0,
      "step": 24340
    },
    {
      "entropy": 5.709082508087159,
      "epoch": 1.8941451079556506,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004638054675591523,
      "loss": 5.079,
      "mean_token_accuracy": 0.19774367809295654,
      "num_tokens": 42232964.0,
      "step": 24345
    },
    {
      "entropy": 5.733403730392456,
      "epoch": 1.8945341373273683,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004637903909931432,
      "loss": 5.0788,
      "mean_token_accuracy": 0.19271970093250274,
      "num_tokens": 42241932.0,
      "step": 24350
    },
    {
      "entropy": 5.645138120651245,
      "epoch": 1.8949231666990858,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004637753115625404,
      "loss": 4.8912,
      "mean_token_accuracy": 0.20349908322095872,
      "num_tokens": 42250146.0,
      "step": 24355
    },
    {
      "entropy": 5.656929779052734,
      "epoch": 1.8953121960708033,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004637602292675726,
      "loss": 5.1267,
      "mean_token_accuracy": 0.19452781826257706,
      "num_tokens": 42259183.0,
      "step": 24360
    },
    {
      "entropy": 5.732398843765258,
      "epoch": 1.895701225442521,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004637451441084689,
      "loss": 5.1551,
      "mean_token_accuracy": 0.189733025431633,
      "num_tokens": 42268436.0,
      "step": 24365
    },
    {
      "entropy": 5.777950572967529,
      "epoch": 1.8960902548142384,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004637300560854581,
      "loss": 5.1021,
      "mean_token_accuracy": 0.19860146641731263,
      "num_tokens": 42277146.0,
      "step": 24370
    },
    {
      "entropy": 5.742022466659546,
      "epoch": 1.896479284185956,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000463714965198769,
      "loss": 5.0604,
      "mean_token_accuracy": 0.1950198993086815,
      "num_tokens": 42286571.0,
      "step": 24375
    },
    {
      "entropy": 5.6356847286224365,
      "epoch": 1.8968683135576736,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004636998714486307,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20521741807460786,
      "num_tokens": 42295415.0,
      "step": 24380
    },
    {
      "entropy": 5.676912832260132,
      "epoch": 1.8972573429293913,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046368477483527224,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19570236802101135,
      "num_tokens": 42305105.0,
      "step": 24385
    },
    {
      "entropy": 5.688040494918823,
      "epoch": 1.8976463723011088,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004636696753589226,
      "loss": 4.981,
      "mean_token_accuracy": 0.20344528555870056,
      "num_tokens": 42313034.0,
      "step": 24390
    },
    {
      "entropy": 5.638887643814087,
      "epoch": 1.8980354016728263,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004636545730198109,
      "loss": 5.0353,
      "mean_token_accuracy": 0.19749893993139267,
      "num_tokens": 42321249.0,
      "step": 24395
    },
    {
      "entropy": 5.641788816452026,
      "epoch": 1.898424431044544,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046363946781816643,
      "loss": 5.0568,
      "mean_token_accuracy": 0.19804082810878754,
      "num_tokens": 42329519.0,
      "step": 24400
    },
    {
      "entropy": 5.742695283889771,
      "epoch": 1.8988134604162614,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046362435975421816,
      "loss": 5.1187,
      "mean_token_accuracy": 0.19449923038482667,
      "num_tokens": 42338318.0,
      "step": 24405
    },
    {
      "entropy": 5.689006710052491,
      "epoch": 1.899202489787979,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046360924882819554,
      "loss": 5.0372,
      "mean_token_accuracy": 0.19813034534454346,
      "num_tokens": 42346038.0,
      "step": 24410
    },
    {
      "entropy": 5.672586250305176,
      "epoch": 1.8995915191596966,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004635941350403277,
      "loss": 5.0187,
      "mean_token_accuracy": 0.20174821764230727,
      "num_tokens": 42356280.0,
      "step": 24415
    },
    {
      "entropy": 5.644523429870605,
      "epoch": 1.899980548531414,
      "grad_norm": 1.25,
      "learning_rate": 0.0004635790183908442,
      "loss": 4.9686,
      "mean_token_accuracy": 0.21053633987903594,
      "num_tokens": 42364716.0,
      "step": 24420
    },
    {
      "entropy": 5.6797035217285154,
      "epoch": 1.9003695779031315,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046356389887997415,
      "loss": 5.1051,
      "mean_token_accuracy": 0.19301884472370148,
      "num_tokens": 42373137.0,
      "step": 24425
    },
    {
      "entropy": 5.683261728286743,
      "epoch": 1.9007586072748492,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046354877650794707,
      "loss": 5.0089,
      "mean_token_accuracy": 0.19646963328123093,
      "num_tokens": 42382054.0,
      "step": 24430
    },
    {
      "entropy": 5.6304632186889645,
      "epoch": 1.901147636646567,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046353365127499235,
      "loss": 5.0157,
      "mean_token_accuracy": 0.1978391870856285,
      "num_tokens": 42391228.0,
      "step": 24435
    },
    {
      "entropy": 5.706356716156006,
      "epoch": 1.9015366660182844,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004635185231813396,
      "loss": 4.9921,
      "mean_token_accuracy": 0.20484830290079117,
      "num_tokens": 42400178.0,
      "step": 24440
    },
    {
      "entropy": 5.693290185928345,
      "epoch": 1.901925695390002,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004635033922272183,
      "loss": 5.0457,
      "mean_token_accuracy": 0.20605269819498062,
      "num_tokens": 42408515.0,
      "step": 24445
    },
    {
      "entropy": 5.738110303878784,
      "epoch": 1.9023147247617196,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046348825841285813,
      "loss": 5.0815,
      "mean_token_accuracy": 0.19458684772253038,
      "num_tokens": 42417291.0,
      "step": 24450
    },
    {
      "entropy": 5.811046600341797,
      "epoch": 1.902703754133437,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004634731217384886,
      "loss": 5.2364,
      "mean_token_accuracy": 0.1796120062470436,
      "num_tokens": 42426371.0,
      "step": 24455
    },
    {
      "entropy": 5.768460845947265,
      "epoch": 1.9030927835051545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004634579822043394,
      "loss": 5.1909,
      "mean_token_accuracy": 0.19514919221401214,
      "num_tokens": 42434767.0,
      "step": 24460
    },
    {
      "entropy": 5.685267496109009,
      "epoch": 1.9034818128768722,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046344283981064035,
      "loss": 5.0596,
      "mean_token_accuracy": 0.2002108722925186,
      "num_tokens": 42443618.0,
      "step": 24465
    },
    {
      "entropy": 5.6775439262390135,
      "epoch": 1.90387084224859,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046342769455762114,
      "loss": 5.0287,
      "mean_token_accuracy": 0.1962940812110901,
      "num_tokens": 42453134.0,
      "step": 24470
    },
    {
      "entropy": 5.713677597045899,
      "epoch": 1.9042598716203072,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004634125464455116,
      "loss": 5.0482,
      "mean_token_accuracy": 0.19347914606332778,
      "num_tokens": 42462037.0,
      "step": 24475
    },
    {
      "entropy": 5.6461576461792,
      "epoch": 1.9046489009920249,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046339739547454146,
      "loss": 4.9839,
      "mean_token_accuracy": 0.20335804373025895,
      "num_tokens": 42471318.0,
      "step": 24480
    },
    {
      "entropy": 5.675210475921631,
      "epoch": 1.9050379303637426,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046338224164494074,
      "loss": 5.0822,
      "mean_token_accuracy": 0.19868521839380265,
      "num_tokens": 42478851.0,
      "step": 24485
    },
    {
      "entropy": 5.618244314193726,
      "epoch": 1.90542695973546,
      "grad_norm": 1.125,
      "learning_rate": 0.00046336708495693933,
      "loss": 4.998,
      "mean_token_accuracy": 0.20081014335155487,
      "num_tokens": 42488406.0,
      "step": 24490
    },
    {
      "entropy": 5.691645097732544,
      "epoch": 1.9058159891071775,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046335192541076725,
      "loss": 4.9317,
      "mean_token_accuracy": 0.20910554379224777,
      "num_tokens": 42496001.0,
      "step": 24495
    },
    {
      "entropy": 5.675712966918946,
      "epoch": 1.9062050184788952,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004633367630066545,
      "loss": 5.093,
      "mean_token_accuracy": 0.192819344997406,
      "num_tokens": 42504808.0,
      "step": 24500
    },
    {
      "entropy": 5.6577520847320555,
      "epoch": 1.9065940478506127,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046332159774483116,
      "loss": 5.0329,
      "mean_token_accuracy": 0.201383475959301,
      "num_tokens": 42512885.0,
      "step": 24505
    },
    {
      "entropy": 5.660343360900879,
      "epoch": 1.9069830772223302,
      "grad_norm": 1.125,
      "learning_rate": 0.0004633064296255273,
      "loss": 5.0083,
      "mean_token_accuracy": 0.19845706075429917,
      "num_tokens": 42521481.0,
      "step": 24510
    },
    {
      "entropy": 5.714602041244507,
      "epoch": 1.9073721065940479,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046329125864897307,
      "loss": 5.0662,
      "mean_token_accuracy": 0.19484372437000275,
      "num_tokens": 42529784.0,
      "step": 24515
    },
    {
      "entropy": 5.656867122650146,
      "epoch": 1.9077611359657656,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004632760848153987,
      "loss": 4.9426,
      "mean_token_accuracy": 0.2061595305800438,
      "num_tokens": 42538802.0,
      "step": 24520
    },
    {
      "entropy": 5.672155904769897,
      "epoch": 1.9081501653374828,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004632609081250345,
      "loss": 4.985,
      "mean_token_accuracy": 0.19744465947151185,
      "num_tokens": 42547699.0,
      "step": 24525
    },
    {
      "entropy": 5.688351774215699,
      "epoch": 1.9085391947092005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046324572857811054,
      "loss": 5.0452,
      "mean_token_accuracy": 0.19838589280843735,
      "num_tokens": 42556583.0,
      "step": 24530
    },
    {
      "entropy": 5.611498403549194,
      "epoch": 1.9089282240809182,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004632305461748573,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21325928270816802,
      "num_tokens": 42565025.0,
      "step": 24535
    },
    {
      "entropy": 5.6442724704742435,
      "epoch": 1.9093172534526357,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046321536091550517,
      "loss": 5.0483,
      "mean_token_accuracy": 0.1989640474319458,
      "num_tokens": 42573798.0,
      "step": 24540
    },
    {
      "entropy": 5.718202304840088,
      "epoch": 1.9097062828243532,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004632001728002845,
      "loss": 5.0257,
      "mean_token_accuracy": 0.20260567516088485,
      "num_tokens": 42582002.0,
      "step": 24545
    },
    {
      "entropy": 5.777220106124878,
      "epoch": 1.9100953121960709,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004631849818294257,
      "loss": 5.0461,
      "mean_token_accuracy": 0.1970430076122284,
      "num_tokens": 42591387.0,
      "step": 24550
    },
    {
      "entropy": 5.741730785369873,
      "epoch": 1.9104843415677883,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046316978800315934,
      "loss": 5.1143,
      "mean_token_accuracy": 0.20090012401342391,
      "num_tokens": 42599769.0,
      "step": 24555
    },
    {
      "entropy": 5.60371618270874,
      "epoch": 1.9108733709395058,
      "grad_norm": 1.171875,
      "learning_rate": 0.000463154591321716,
      "loss": 4.9644,
      "mean_token_accuracy": 0.2052150085568428,
      "num_tokens": 42608279.0,
      "step": 24560
    },
    {
      "entropy": 5.641292524337769,
      "epoch": 1.9112624003112235,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046313939178532624,
      "loss": 5.0369,
      "mean_token_accuracy": 0.19688843190670013,
      "num_tokens": 42616652.0,
      "step": 24565
    },
    {
      "entropy": 5.638890123367309,
      "epoch": 1.9116514296829412,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004631241893942206,
      "loss": 5.0016,
      "mean_token_accuracy": 0.19857836067676543,
      "num_tokens": 42625629.0,
      "step": 24570
    },
    {
      "entropy": 5.677545595169067,
      "epoch": 1.9120404590546585,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046310898414862983,
      "loss": 5.0823,
      "mean_token_accuracy": 0.20221803337335587,
      "num_tokens": 42634159.0,
      "step": 24575
    },
    {
      "entropy": 5.743991088867188,
      "epoch": 1.9124294884263762,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004630937760487847,
      "loss": 5.1482,
      "mean_token_accuracy": 0.1890771880745888,
      "num_tokens": 42642880.0,
      "step": 24580
    },
    {
      "entropy": 5.697179937362671,
      "epoch": 1.9128185177980939,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004630785650949158,
      "loss": 5.0869,
      "mean_token_accuracy": 0.1961984694004059,
      "num_tokens": 42651481.0,
      "step": 24585
    },
    {
      "entropy": 5.7100200176239015,
      "epoch": 1.9132075471698113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004630633512872541,
      "loss": 5.0566,
      "mean_token_accuracy": 0.20454505532979966,
      "num_tokens": 42660131.0,
      "step": 24590
    },
    {
      "entropy": 5.664259052276611,
      "epoch": 1.9135965765415288,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046304813462603035,
      "loss": 5.0797,
      "mean_token_accuracy": 0.20387784540653228,
      "num_tokens": 42669094.0,
      "step": 24595
    },
    {
      "entropy": 5.676291418075562,
      "epoch": 1.9139856059132465,
      "grad_norm": 1.125,
      "learning_rate": 0.0004630329151114754,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20979381799697877,
      "num_tokens": 42677792.0,
      "step": 24600
    },
    {
      "entropy": 5.725220823287964,
      "epoch": 1.914374635284964,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046301769274382034,
      "loss": 5.0953,
      "mean_token_accuracy": 0.19348185509443283,
      "num_tokens": 42686397.0,
      "step": 24605
    },
    {
      "entropy": 5.697055244445801,
      "epoch": 1.9147636646566815,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004630024675232961,
      "loss": 5.0259,
      "mean_token_accuracy": 0.20506162643432618,
      "num_tokens": 42695832.0,
      "step": 24610
    },
    {
      "entropy": 5.674222087860107,
      "epoch": 1.9151526940283992,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046298723945013354,
      "loss": 5.0458,
      "mean_token_accuracy": 0.19670222103595733,
      "num_tokens": 42704378.0,
      "step": 24615
    },
    {
      "entropy": 5.6558003425598145,
      "epoch": 1.9155417234001169,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004629720085245639,
      "loss": 5.0032,
      "mean_token_accuracy": 0.19691879600286483,
      "num_tokens": 42713926.0,
      "step": 24620
    },
    {
      "entropy": 5.726751518249512,
      "epoch": 1.9159307527718341,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004629567747468182,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20231074392795562,
      "num_tokens": 42722779.0,
      "step": 24625
    },
    {
      "entropy": 5.689429044723511,
      "epoch": 1.9163197821435518,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004629415381171276,
      "loss": 4.9647,
      "mean_token_accuracy": 0.2019610345363617,
      "num_tokens": 42731460.0,
      "step": 24630
    },
    {
      "entropy": 5.6924498081207275,
      "epoch": 1.9167088115152695,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046292629863572325,
      "loss": 5.0535,
      "mean_token_accuracy": 0.19991564005613327,
      "num_tokens": 42739733.0,
      "step": 24635
    },
    {
      "entropy": 5.753908491134643,
      "epoch": 1.917097840886987,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004629110563028365,
      "loss": 5.0719,
      "mean_token_accuracy": 0.19914218932390212,
      "num_tokens": 42748330.0,
      "step": 24640
    },
    {
      "entropy": 5.695810317993164,
      "epoch": 1.9174868702587045,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004628958111186985,
      "loss": 5.0751,
      "mean_token_accuracy": 0.19527909755706788,
      "num_tokens": 42757423.0,
      "step": 24645
    },
    {
      "entropy": 5.668301963806153,
      "epoch": 1.9178758996304222,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004628805630835407,
      "loss": 5.0847,
      "mean_token_accuracy": 0.192294117808342,
      "num_tokens": 42766213.0,
      "step": 24650
    },
    {
      "entropy": 5.748909664154053,
      "epoch": 1.9182649290021396,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004628653121975944,
      "loss": 5.0907,
      "mean_token_accuracy": 0.20167043358087539,
      "num_tokens": 42774037.0,
      "step": 24655
    },
    {
      "entropy": 5.688054895401001,
      "epoch": 1.918653958373857,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000462850058461091,
      "loss": 5.1383,
      "mean_token_accuracy": 0.1927604928612709,
      "num_tokens": 42782806.0,
      "step": 24660
    },
    {
      "entropy": 5.599278163909912,
      "epoch": 1.9190429877455748,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004628348018742619,
      "loss": 4.9692,
      "mean_token_accuracy": 0.19859009683132173,
      "num_tokens": 42791790.0,
      "step": 24665
    },
    {
      "entropy": 5.725513744354248,
      "epoch": 1.9194320171172925,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004628195424373387,
      "loss": 5.1401,
      "mean_token_accuracy": 0.18950953036546708,
      "num_tokens": 42799966.0,
      "step": 24670
    },
    {
      "entropy": 5.696822452545166,
      "epoch": 1.91982104648901,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046280428015055286,
      "loss": 5.0379,
      "mean_token_accuracy": 0.20477413833141328,
      "num_tokens": 42808022.0,
      "step": 24675
    },
    {
      "entropy": 5.681642007827759,
      "epoch": 1.9202100758607274,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046278901501413606,
      "loss": 4.957,
      "mean_token_accuracy": 0.21134666502475738,
      "num_tokens": 42816191.0,
      "step": 24680
    },
    {
      "entropy": 5.693042469024658,
      "epoch": 1.9205991052324451,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046277374702831983,
      "loss": 5.0534,
      "mean_token_accuracy": 0.19525159299373626,
      "num_tokens": 42824840.0,
      "step": 24685
    },
    {
      "entropy": 5.672334671020508,
      "epoch": 1.9209881346041626,
      "grad_norm": 1.125,
      "learning_rate": 0.00046275847619333576,
      "loss": 5.0396,
      "mean_token_accuracy": 0.1973715215921402,
      "num_tokens": 42833788.0,
      "step": 24690
    },
    {
      "entropy": 5.734400939941406,
      "epoch": 1.92137716397588,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046274320250941576,
      "loss": 5.0836,
      "mean_token_accuracy": 0.19542493373155595,
      "num_tokens": 42842666.0,
      "step": 24695
    },
    {
      "entropy": 5.775894498825073,
      "epoch": 1.9217661933475978,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046272792597679146,
      "loss": 5.1169,
      "mean_token_accuracy": 0.1904724881052971,
      "num_tokens": 42850725.0,
      "step": 24700
    },
    {
      "entropy": 5.680692195892334,
      "epoch": 1.9221552227193153,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046271264659569474,
      "loss": 5.0681,
      "mean_token_accuracy": 0.19184989780187606,
      "num_tokens": 42859732.0,
      "step": 24705
    },
    {
      "entropy": 5.715995359420776,
      "epoch": 1.9225442520910327,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004626973643663573,
      "loss": 5.0564,
      "mean_token_accuracy": 0.19708100259304046,
      "num_tokens": 42868858.0,
      "step": 24710
    },
    {
      "entropy": 5.722647571563721,
      "epoch": 1.9229332814627504,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004626820792890112,
      "loss": 5.0865,
      "mean_token_accuracy": 0.19535237103700637,
      "num_tokens": 42877483.0,
      "step": 24715
    },
    {
      "entropy": 5.680809354782104,
      "epoch": 1.9233223108344681,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004626667913638882,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21316238939762117,
      "num_tokens": 42885156.0,
      "step": 24720
    },
    {
      "entropy": 5.730360412597657,
      "epoch": 1.9237113402061856,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004626515005912203,
      "loss": 5.1225,
      "mean_token_accuracy": 0.19758251160383225,
      "num_tokens": 42893599.0,
      "step": 24725
    },
    {
      "entropy": 5.679024171829224,
      "epoch": 1.924100369577903,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004626362069712397,
      "loss": 4.9986,
      "mean_token_accuracy": 0.2093589186668396,
      "num_tokens": 42902861.0,
      "step": 24730
    },
    {
      "entropy": 5.689429044723511,
      "epoch": 1.9244893989496208,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004626209105041782,
      "loss": 5.0547,
      "mean_token_accuracy": 0.19889709502458572,
      "num_tokens": 42911246.0,
      "step": 24735
    },
    {
      "entropy": 5.670091724395752,
      "epoch": 1.9248784283213383,
      "grad_norm": 1.21875,
      "learning_rate": 0.000462605611190268,
      "loss": 5.1332,
      "mean_token_accuracy": 0.194108085334301,
      "num_tokens": 42919526.0,
      "step": 24740
    },
    {
      "entropy": 5.666739797592163,
      "epoch": 1.9252674576930557,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046259030902974133,
      "loss": 4.952,
      "mean_token_accuracy": 0.20560158789157867,
      "num_tokens": 42927473.0,
      "step": 24745
    },
    {
      "entropy": 5.644410848617554,
      "epoch": 1.9256564870647734,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004625750040228302,
      "loss": 5.0159,
      "mean_token_accuracy": 0.20053688436746597,
      "num_tokens": 42936108.0,
      "step": 24750
    },
    {
      "entropy": 5.629378366470337,
      "epoch": 1.926045516436491,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046255969616976704,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20834380388259888,
      "num_tokens": 42944147.0,
      "step": 24755
    },
    {
      "entropy": 5.665427732467651,
      "epoch": 1.9264345458082084,
      "grad_norm": 1.0625,
      "learning_rate": 0.000462544385470784,
      "loss": 5.1342,
      "mean_token_accuracy": 0.1902235969901085,
      "num_tokens": 42952674.0,
      "step": 24760
    },
    {
      "entropy": 5.665197038650513,
      "epoch": 1.926823575179926,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004625290719261134,
      "loss": 5.0126,
      "mean_token_accuracy": 0.20286975353956221,
      "num_tokens": 42960907.0,
      "step": 24765
    },
    {
      "entropy": 5.685473823547364,
      "epoch": 1.9272126045516438,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004625137555359876,
      "loss": 5.0076,
      "mean_token_accuracy": 0.194828163087368,
      "num_tokens": 42968875.0,
      "step": 24770
    },
    {
      "entropy": 5.692585468292236,
      "epoch": 1.9276016339233613,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046249843630063905,
      "loss": 5.0266,
      "mean_token_accuracy": 0.20408463180065156,
      "num_tokens": 42977062.0,
      "step": 24775
    },
    {
      "entropy": 5.668549203872681,
      "epoch": 1.9279906632950787,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004624831142203001,
      "loss": 5.0032,
      "mean_token_accuracy": 0.20419415533542634,
      "num_tokens": 42985346.0,
      "step": 24780
    },
    {
      "entropy": 5.656054210662842,
      "epoch": 1.9283796926667964,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046246778929520346,
      "loss": 5.0654,
      "mean_token_accuracy": 0.18509457260370255,
      "num_tokens": 42993529.0,
      "step": 24785
    },
    {
      "entropy": 5.7392017364501955,
      "epoch": 1.928768722038514,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004624524615255814,
      "loss": 5.068,
      "mean_token_accuracy": 0.1970072254538536,
      "num_tokens": 43002254.0,
      "step": 24790
    },
    {
      "entropy": 5.679862833023071,
      "epoch": 1.9291577514102314,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046243713091166655,
      "loss": 5.0453,
      "mean_token_accuracy": 0.20193109959363936,
      "num_tokens": 43011130.0,
      "step": 24795
    },
    {
      "entropy": 5.642696666717529,
      "epoch": 1.929546780781949,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004624217974536916,
      "loss": 4.9932,
      "mean_token_accuracy": 0.20740202218294143,
      "num_tokens": 43019185.0,
      "step": 24800
    },
    {
      "entropy": 5.737049150466919,
      "epoch": 1.9299358101536666,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046240646115188925,
      "loss": 5.1968,
      "mean_token_accuracy": 0.18931230902671814,
      "num_tokens": 43028817.0,
      "step": 24805
    },
    {
      "entropy": 5.688579940795899,
      "epoch": 1.930324839525384,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004623911220064921,
      "loss": 5.0191,
      "mean_token_accuracy": 0.20310221016407012,
      "num_tokens": 43038205.0,
      "step": 24810
    },
    {
      "entropy": 5.654299974441528,
      "epoch": 1.9307138688971017,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046237578001773297,
      "loss": 5.0434,
      "mean_token_accuracy": 0.19733235687017442,
      "num_tokens": 43046844.0,
      "step": 24815
    },
    {
      "entropy": 5.680369710922241,
      "epoch": 1.9311028982688194,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046236043518584454,
      "loss": 5.1311,
      "mean_token_accuracy": 0.193811796605587,
      "num_tokens": 43055928.0,
      "step": 24820
    },
    {
      "entropy": 5.735134601593018,
      "epoch": 1.931491927640537,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004623450875110597,
      "loss": 5.0654,
      "mean_token_accuracy": 0.1998404085636139,
      "num_tokens": 43065371.0,
      "step": 24825
    },
    {
      "entropy": 5.732225799560547,
      "epoch": 1.9318809570122544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046232973699361147,
      "loss": 5.0069,
      "mean_token_accuracy": 0.19985468536615372,
      "num_tokens": 43073486.0,
      "step": 24830
    },
    {
      "entropy": 5.636778211593628,
      "epoch": 1.932269986383972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004623143836337326,
      "loss": 5.0505,
      "mean_token_accuracy": 0.20184364765882493,
      "num_tokens": 43083028.0,
      "step": 24835
    },
    {
      "entropy": 5.636121416091919,
      "epoch": 1.9326590157556895,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046229902743165607,
      "loss": 5.0425,
      "mean_token_accuracy": 0.19869482964277269,
      "num_tokens": 43091134.0,
      "step": 24840
    },
    {
      "entropy": 5.708029317855835,
      "epoch": 1.933048045127407,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004622836683876149,
      "loss": 5.0848,
      "mean_token_accuracy": 0.19872661232948302,
      "num_tokens": 43099849.0,
      "step": 24845
    },
    {
      "entropy": 5.71333065032959,
      "epoch": 1.9334370744991247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004622683065018422,
      "loss": 5.0355,
      "mean_token_accuracy": 0.1981332778930664,
      "num_tokens": 43108311.0,
      "step": 24850
    },
    {
      "entropy": 5.674883460998535,
      "epoch": 1.9338261038708422,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046225294177457105,
      "loss": 4.9317,
      "mean_token_accuracy": 0.20503869354724885,
      "num_tokens": 43116179.0,
      "step": 24855
    },
    {
      "entropy": 5.600819826126099,
      "epoch": 1.9342151332425597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046223757420603445,
      "loss": 4.9646,
      "mean_token_accuracy": 0.20399170368909836,
      "num_tokens": 43125728.0,
      "step": 24860
    },
    {
      "entropy": 5.598449850082398,
      "epoch": 1.9346041626142774,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004622222037964658,
      "loss": 4.9376,
      "mean_token_accuracy": 0.20348993241786956,
      "num_tokens": 43134512.0,
      "step": 24865
    },
    {
      "entropy": 5.622919511795044,
      "epoch": 1.934993191985995,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046220683054609815,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20116004347801208,
      "num_tokens": 43144831.0,
      "step": 24870
    },
    {
      "entropy": 5.68464150428772,
      "epoch": 1.9353822213577125,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046219145445516483,
      "loss": 5.0808,
      "mean_token_accuracy": 0.1960323229432106,
      "num_tokens": 43153614.0,
      "step": 24875
    },
    {
      "entropy": 5.69285020828247,
      "epoch": 1.93577125072943,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046217607552389905,
      "loss": 5.0655,
      "mean_token_accuracy": 0.19818117022514342,
      "num_tokens": 43162004.0,
      "step": 24880
    },
    {
      "entropy": 5.670533895492554,
      "epoch": 1.9361602801011477,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046216069375253435,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20691214203834535,
      "num_tokens": 43170255.0,
      "step": 24885
    },
    {
      "entropy": 5.633712339401245,
      "epoch": 1.9365493094728652,
      "grad_norm": 1.125,
      "learning_rate": 0.000462145309141304,
      "loss": 5.0684,
      "mean_token_accuracy": 0.1982306033372879,
      "num_tokens": 43179036.0,
      "step": 24890
    },
    {
      "entropy": 5.723203325271607,
      "epoch": 1.9369383388445827,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004621299216904414,
      "loss": 5.0823,
      "mean_token_accuracy": 0.1950405389070511,
      "num_tokens": 43187207.0,
      "step": 24895
    },
    {
      "entropy": 5.6816870212554935,
      "epoch": 1.9373273682163004,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046211453140018006,
      "loss": 5.0717,
      "mean_token_accuracy": 0.19018362015485762,
      "num_tokens": 43195600.0,
      "step": 24900
    },
    {
      "entropy": 5.709857368469239,
      "epoch": 1.937716397588018,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004620991382707537,
      "loss": 5.0779,
      "mean_token_accuracy": 0.2035515621304512,
      "num_tokens": 43204069.0,
      "step": 24905
    },
    {
      "entropy": 5.692591142654419,
      "epoch": 1.9381054269597353,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046208374230239556,
      "loss": 5.1354,
      "mean_token_accuracy": 0.19016434848308564,
      "num_tokens": 43212998.0,
      "step": 24910
    },
    {
      "entropy": 5.737407112121582,
      "epoch": 1.938494456331453,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004620683434953394,
      "loss": 5.086,
      "mean_token_accuracy": 0.1937735676765442,
      "num_tokens": 43222411.0,
      "step": 24915
    },
    {
      "entropy": 5.699862766265869,
      "epoch": 1.9388834857031707,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046205294184981896,
      "loss": 5.0396,
      "mean_token_accuracy": 0.19747995287179948,
      "num_tokens": 43231265.0,
      "step": 24920
    },
    {
      "entropy": 5.7016222953796385,
      "epoch": 1.9392725150748882,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046203753736606787,
      "loss": 5.0394,
      "mean_token_accuracy": 0.19650323390960694,
      "num_tokens": 43240609.0,
      "step": 24925
    },
    {
      "entropy": 5.713491201400757,
      "epoch": 1.9396615444466057,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004620221300443197,
      "loss": 5.0544,
      "mean_token_accuracy": 0.20287737548351287,
      "num_tokens": 43249381.0,
      "step": 24930
    },
    {
      "entropy": 5.702598142623901,
      "epoch": 1.9400505738183234,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004620067198848086,
      "loss": 5.0275,
      "mean_token_accuracy": 0.1996012806892395,
      "num_tokens": 43257653.0,
      "step": 24935
    },
    {
      "entropy": 5.71220326423645,
      "epoch": 1.9404396031900408,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046199130688776805,
      "loss": 5.1243,
      "mean_token_accuracy": 0.19815213978290558,
      "num_tokens": 43267111.0,
      "step": 24940
    },
    {
      "entropy": 5.661815547943116,
      "epoch": 1.9408286325617583,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046197589105343204,
      "loss": 4.9933,
      "mean_token_accuracy": 0.20314147025346757,
      "num_tokens": 43276207.0,
      "step": 24945
    },
    {
      "entropy": 5.737192058563233,
      "epoch": 1.941217661933476,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004619604723820345,
      "loss": 5.0419,
      "mean_token_accuracy": 0.20260156095027923,
      "num_tokens": 43285031.0,
      "step": 24950
    },
    {
      "entropy": 5.72438268661499,
      "epoch": 1.9416066913051937,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004619450508738094,
      "loss": 5.0738,
      "mean_token_accuracy": 0.19643659442663192,
      "num_tokens": 43292802.0,
      "step": 24955
    },
    {
      "entropy": 5.721855974197387,
      "epoch": 1.941995720676911,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046192962652899056,
      "loss": 5.0207,
      "mean_token_accuracy": 0.20338866114616394,
      "num_tokens": 43302349.0,
      "step": 24960
    },
    {
      "entropy": 5.661811447143554,
      "epoch": 1.9423847500486286,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046191419934781236,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20217992961406708,
      "num_tokens": 43311455.0,
      "step": 24965
    },
    {
      "entropy": 5.671417093276977,
      "epoch": 1.9427737794203463,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004618987693305086,
      "loss": 5.0214,
      "mean_token_accuracy": 0.20294463485479355,
      "num_tokens": 43319924.0,
      "step": 24970
    },
    {
      "entropy": 5.708370018005371,
      "epoch": 1.9431628087920638,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004618833364773135,
      "loss": 5.1489,
      "mean_token_accuracy": 0.19678598195314406,
      "num_tokens": 43328772.0,
      "step": 24975
    },
    {
      "entropy": 5.66936502456665,
      "epoch": 1.9435518381637813,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046186790078846127,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20237229019403458,
      "num_tokens": 43338862.0,
      "step": 24980
    },
    {
      "entropy": 5.696899843215943,
      "epoch": 1.943940867535499,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046185246226418603,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19844810962677,
      "num_tokens": 43347292.0,
      "step": 24985
    },
    {
      "entropy": 5.616959047317505,
      "epoch": 1.9443298969072165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046183702090472206,
      "loss": 4.9908,
      "mean_token_accuracy": 0.2068087115883827,
      "num_tokens": 43356676.0,
      "step": 24990
    },
    {
      "entropy": 5.723744297027588,
      "epoch": 1.944718926278934,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004618215767103037,
      "loss": 5.1452,
      "mean_token_accuracy": 0.19212857335805894,
      "num_tokens": 43364746.0,
      "step": 24995
    },
    {
      "entropy": 5.699723911285401,
      "epoch": 1.9451079556506516,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004618061296811653,
      "loss": 4.9689,
      "mean_token_accuracy": 0.20378757268190384,
      "num_tokens": 43373541.0,
      "step": 25000
    },
    {
      "entropy": 5.688391780853271,
      "epoch": 1.9454969850223693,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046179067981754124,
      "loss": 4.9814,
      "mean_token_accuracy": 0.19692103266716005,
      "num_tokens": 43382524.0,
      "step": 25005
    },
    {
      "entropy": 5.605011510848999,
      "epoch": 1.9458860143940866,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046177522711966584,
      "loss": 5.0189,
      "mean_token_accuracy": 0.20577212423086166,
      "num_tokens": 43391875.0,
      "step": 25010
    },
    {
      "entropy": 5.709514570236206,
      "epoch": 1.9462750437658043,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046175977158777377,
      "loss": 5.1163,
      "mean_token_accuracy": 0.193670791387558,
      "num_tokens": 43399942.0,
      "step": 25015
    },
    {
      "entropy": 5.727441740036011,
      "epoch": 1.946664073137522,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004617443132220993,
      "loss": 5.0863,
      "mean_token_accuracy": 0.19419944584369658,
      "num_tokens": 43409371.0,
      "step": 25020
    },
    {
      "entropy": 5.68810133934021,
      "epoch": 1.9470531025092395,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046172885202287717,
      "loss": 5.068,
      "mean_token_accuracy": 0.2046906530857086,
      "num_tokens": 43418181.0,
      "step": 25025
    },
    {
      "entropy": 5.623637390136719,
      "epoch": 1.947442131880957,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046171338799034194,
      "loss": 5.0198,
      "mean_token_accuracy": 0.20464629977941512,
      "num_tokens": 43426004.0,
      "step": 25030
    },
    {
      "entropy": 5.652096462249756,
      "epoch": 1.9478311612526746,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004616979211247282,
      "loss": 4.9283,
      "mean_token_accuracy": 0.20263815373182298,
      "num_tokens": 43433732.0,
      "step": 25035
    },
    {
      "entropy": 5.694263124465943,
      "epoch": 1.9482201906243921,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046168245142627065,
      "loss": 5.0665,
      "mean_token_accuracy": 0.2010141909122467,
      "num_tokens": 43442374.0,
      "step": 25040
    },
    {
      "entropy": 5.723499917984009,
      "epoch": 1.9486092199961096,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004616669788952041,
      "loss": 5.1035,
      "mean_token_accuracy": 0.19832152575254441,
      "num_tokens": 43451335.0,
      "step": 25045
    },
    {
      "entropy": 5.728225612640381,
      "epoch": 1.9489982493678273,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046165150353176315,
      "loss": 5.1091,
      "mean_token_accuracy": 0.1952285036444664,
      "num_tokens": 43459520.0,
      "step": 25050
    },
    {
      "entropy": 5.669061183929443,
      "epoch": 1.949387278739545,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004616360253361828,
      "loss": 5.0269,
      "mean_token_accuracy": 0.19924459159374236,
      "num_tokens": 43467695.0,
      "step": 25055
    },
    {
      "entropy": 5.6503904342651365,
      "epoch": 1.9497763081112622,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046162054430869777,
      "loss": 5.0821,
      "mean_token_accuracy": 0.19960678070783616,
      "num_tokens": 43476161.0,
      "step": 25060
    },
    {
      "entropy": 5.6617834091186525,
      "epoch": 1.95016533748298,
      "grad_norm": 1.125,
      "learning_rate": 0.000461605060449543,
      "loss": 4.9942,
      "mean_token_accuracy": 0.19702244251966478,
      "num_tokens": 43485106.0,
      "step": 25065
    },
    {
      "entropy": 5.719282722473144,
      "epoch": 1.9505543668546976,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046158957375895353,
      "loss": 5.0888,
      "mean_token_accuracy": 0.1899719089269638,
      "num_tokens": 43494232.0,
      "step": 25070
    },
    {
      "entropy": 5.7387049198150635,
      "epoch": 1.950943396226415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004615740842371643,
      "loss": 4.9995,
      "mean_token_accuracy": 0.2024102047085762,
      "num_tokens": 43502556.0,
      "step": 25075
    },
    {
      "entropy": 5.633154964447021,
      "epoch": 1.9513324255981326,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046155859188441023,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2012600228190422,
      "num_tokens": 43511594.0,
      "step": 25080
    },
    {
      "entropy": 5.637977743148804,
      "epoch": 1.9517214549698503,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004615430967009265,
      "loss": 5.093,
      "mean_token_accuracy": 0.19673862755298616,
      "num_tokens": 43520532.0,
      "step": 25085
    },
    {
      "entropy": 5.700028991699218,
      "epoch": 1.9521104843415678,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046152759868694815,
      "loss": 5.1048,
      "mean_token_accuracy": 0.19827057868242265,
      "num_tokens": 43529286.0,
      "step": 25090
    },
    {
      "entropy": 5.72367844581604,
      "epoch": 1.9524995137132852,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004615120978427104,
      "loss": 5.0569,
      "mean_token_accuracy": 0.19502451717853547,
      "num_tokens": 43537983.0,
      "step": 25095
    },
    {
      "entropy": 5.682291316986084,
      "epoch": 1.952888543085003,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004614965941684485,
      "loss": 5.007,
      "mean_token_accuracy": 0.20533406883478164,
      "num_tokens": 43546550.0,
      "step": 25100
    },
    {
      "entropy": 5.7151782512664795,
      "epoch": 1.9532775724567206,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004614810876643975,
      "loss": 5.0969,
      "mean_token_accuracy": 0.2019813746213913,
      "num_tokens": 43555006.0,
      "step": 25105
    },
    {
      "entropy": 5.7196849346160885,
      "epoch": 1.953666601828438,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046146557833079286,
      "loss": 5.1581,
      "mean_token_accuracy": 0.19339825510978698,
      "num_tokens": 43563925.0,
      "step": 25110
    },
    {
      "entropy": 5.740710496902466,
      "epoch": 1.9540556312001556,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004614500661678698,
      "loss": 5.0532,
      "mean_token_accuracy": 0.19742967337369918,
      "num_tokens": 43572269.0,
      "step": 25115
    },
    {
      "entropy": 5.706193733215332,
      "epoch": 1.9544446605718733,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004614345511758639,
      "loss": 5.0691,
      "mean_token_accuracy": 0.19874023497104645,
      "num_tokens": 43579908.0,
      "step": 25120
    },
    {
      "entropy": 5.659958171844482,
      "epoch": 1.9548336899435907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046141903335501034,
      "loss": 5.015,
      "mean_token_accuracy": 0.19763632863759995,
      "num_tokens": 43589328.0,
      "step": 25125
    },
    {
      "entropy": 5.683829927444458,
      "epoch": 1.9552227193153082,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004614035127055447,
      "loss": 5.0596,
      "mean_token_accuracy": 0.20008797943592072,
      "num_tokens": 43597840.0,
      "step": 25130
    },
    {
      "entropy": 5.664583683013916,
      "epoch": 1.955611748687026,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004613879892277024,
      "loss": 5.0113,
      "mean_token_accuracy": 0.20468807816505433,
      "num_tokens": 43606496.0,
      "step": 25135
    },
    {
      "entropy": 5.727624893188477,
      "epoch": 1.9560007780587434,
      "grad_norm": 1.109375,
      "learning_rate": 0.000461372462921719,
      "loss": 5.0211,
      "mean_token_accuracy": 0.20284863114356994,
      "num_tokens": 43614411.0,
      "step": 25140
    },
    {
      "entropy": 5.612778854370117,
      "epoch": 1.9563898074304609,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004613569337878302,
      "loss": 4.9732,
      "mean_token_accuracy": 0.2039729431271553,
      "num_tokens": 43622842.0,
      "step": 25145
    },
    {
      "entropy": 5.680815315246582,
      "epoch": 1.9567788368021786,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004613414018262715,
      "loss": 5.081,
      "mean_token_accuracy": 0.19264834970235825,
      "num_tokens": 43631535.0,
      "step": 25150
    },
    {
      "entropy": 5.724313402175904,
      "epoch": 1.9571678661738963,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004613258670372786,
      "loss": 5.075,
      "mean_token_accuracy": 0.19988464266061784,
      "num_tokens": 43640197.0,
      "step": 25155
    },
    {
      "entropy": 5.632492113113403,
      "epoch": 1.9575568955456137,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004613103294210873,
      "loss": 5.0124,
      "mean_token_accuracy": 0.2055905982851982,
      "num_tokens": 43648685.0,
      "step": 25160
    },
    {
      "entropy": 5.63893690109253,
      "epoch": 1.9579459249173312,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046129478897793323,
      "loss": 5.0327,
      "mean_token_accuracy": 0.19616321921348573,
      "num_tokens": 43657526.0,
      "step": 25165
    },
    {
      "entropy": 5.733870220184326,
      "epoch": 1.958334954289049,
      "grad_norm": 1.125,
      "learning_rate": 0.00046127924570805236,
      "loss": 5.0521,
      "mean_token_accuracy": 0.20517999231815337,
      "num_tokens": 43666119.0,
      "step": 25170
    },
    {
      "entropy": 5.7572027206420895,
      "epoch": 1.9587239836607664,
      "grad_norm": 1.125,
      "learning_rate": 0.0004612636996116803,
      "loss": 5.0335,
      "mean_token_accuracy": 0.19316034466028215,
      "num_tokens": 43674634.0,
      "step": 25175
    },
    {
      "entropy": 5.587348175048828,
      "epoch": 1.9591130130324839,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004612481506890532,
      "loss": 4.9749,
      "mean_token_accuracy": 0.2059198796749115,
      "num_tokens": 43683329.0,
      "step": 25180
    },
    {
      "entropy": 5.652428865432739,
      "epoch": 1.9595020424042016,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046123259894040677,
      "loss": 4.997,
      "mean_token_accuracy": 0.20335134565830232,
      "num_tokens": 43692598.0,
      "step": 25185
    },
    {
      "entropy": 5.6406090259552,
      "epoch": 1.959891071775919,
      "grad_norm": 1.25,
      "learning_rate": 0.000461217044365977,
      "loss": 4.9794,
      "mean_token_accuracy": 0.19794338643550874,
      "num_tokens": 43702260.0,
      "step": 25190
    },
    {
      "entropy": 5.741105222702027,
      "epoch": 1.9602801011476365,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004612014869660002,
      "loss": 5.0847,
      "mean_token_accuracy": 0.19749262034893036,
      "num_tokens": 43710003.0,
      "step": 25195
    },
    {
      "entropy": 5.699049425125122,
      "epoch": 1.9606691305193542,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046118592674071197,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2033785253763199,
      "num_tokens": 43718687.0,
      "step": 25200
    },
    {
      "entropy": 5.628763008117676,
      "epoch": 1.961058159891072,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004611703636903488,
      "loss": 4.9408,
      "mean_token_accuracy": 0.21014311909675598,
      "num_tokens": 43727051.0,
      "step": 25205
    },
    {
      "entropy": 5.652144813537598,
      "epoch": 1.9614471892627894,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004611547978151466,
      "loss": 4.9824,
      "mean_token_accuracy": 0.20597678273916245,
      "num_tokens": 43735572.0,
      "step": 25210
    },
    {
      "entropy": 5.69415807723999,
      "epoch": 1.9618362186345069,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046113922911534167,
      "loss": 5.0378,
      "mean_token_accuracy": 0.20135923326015473,
      "num_tokens": 43744286.0,
      "step": 25215
    },
    {
      "entropy": 5.683207893371582,
      "epoch": 1.9622252480062246,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004611236575911702,
      "loss": 5.1179,
      "mean_token_accuracy": 0.19950320273637773,
      "num_tokens": 43754042.0,
      "step": 25220
    },
    {
      "entropy": 5.661234331130982,
      "epoch": 1.962614277377942,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046110808324286844,
      "loss": 4.918,
      "mean_token_accuracy": 0.20831528156995774,
      "num_tokens": 43762031.0,
      "step": 25225
    },
    {
      "entropy": 5.694571018218994,
      "epoch": 1.9630033067496595,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046109250607067275,
      "loss": 5.0853,
      "mean_token_accuracy": 0.20041561126708984,
      "num_tokens": 43770546.0,
      "step": 25230
    },
    {
      "entropy": 5.674601268768311,
      "epoch": 1.9633923361213772,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004610769260748196,
      "loss": 5.0659,
      "mean_token_accuracy": 0.2015522912144661,
      "num_tokens": 43779918.0,
      "step": 25235
    },
    {
      "entropy": 5.712134981155396,
      "epoch": 1.9637813654930947,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004610613432555451,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19891495406627654,
      "num_tokens": 43788425.0,
      "step": 25240
    },
    {
      "entropy": 5.687092208862305,
      "epoch": 1.9641703948648122,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046104575761308597,
      "loss": 5.081,
      "mean_token_accuracy": 0.1917901709675789,
      "num_tokens": 43797080.0,
      "step": 25245
    },
    {
      "entropy": 5.659074115753174,
      "epoch": 1.9645594242365298,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004610301691476786,
      "loss": 5.0305,
      "mean_token_accuracy": 0.20866090804338455,
      "num_tokens": 43806026.0,
      "step": 25250
    },
    {
      "entropy": 5.742336416244507,
      "epoch": 1.9649484536082475,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004610145778595594,
      "loss": 5.0714,
      "mean_token_accuracy": 0.20536938160657883,
      "num_tokens": 43814333.0,
      "step": 25255
    },
    {
      "entropy": 5.67303056716919,
      "epoch": 1.965337482979965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004609989837489651,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20021286457777024,
      "num_tokens": 43823235.0,
      "step": 25260
    },
    {
      "entropy": 5.7027748107910154,
      "epoch": 1.9657265123516825,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046098338681613234,
      "loss": 5.0472,
      "mean_token_accuracy": 0.2012014254927635,
      "num_tokens": 43832257.0,
      "step": 25265
    },
    {
      "entropy": 5.693505144119262,
      "epoch": 1.9661155417234002,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004609677870612976,
      "loss": 4.9915,
      "mean_token_accuracy": 0.20833290070295335,
      "num_tokens": 43840359.0,
      "step": 25270
    },
    {
      "entropy": 5.621105766296386,
      "epoch": 1.9665045710951177,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004609521844846978,
      "loss": 5.0995,
      "mean_token_accuracy": 0.19711009413003922,
      "num_tokens": 43849374.0,
      "step": 25275
    },
    {
      "entropy": 5.679486989974976,
      "epoch": 1.9668936004668351,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004609365790865695,
      "loss": 5.0904,
      "mean_token_accuracy": 0.19939954429864884,
      "num_tokens": 43857861.0,
      "step": 25280
    },
    {
      "entropy": 5.728759860992431,
      "epoch": 1.9672826298385528,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046092097086714956,
      "loss": 5.0783,
      "mean_token_accuracy": 0.19538939893245696,
      "num_tokens": 43865964.0,
      "step": 25285
    },
    {
      "entropy": 5.6879819393157955,
      "epoch": 1.9676716592102705,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046090535982667486,
      "loss": 4.9977,
      "mean_token_accuracy": 0.21127932667732238,
      "num_tokens": 43874434.0,
      "step": 25290
    },
    {
      "entropy": 5.68955807685852,
      "epoch": 1.9680606885819878,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046088974596538216,
      "loss": 5.0581,
      "mean_token_accuracy": 0.2008354365825653,
      "num_tokens": 43882691.0,
      "step": 25295
    },
    {
      "entropy": 5.7535323143005375,
      "epoch": 1.9684497179537055,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004608741292835085,
      "loss": 5.0778,
      "mean_token_accuracy": 0.19867088049650192,
      "num_tokens": 43890589.0,
      "step": 25300
    },
    {
      "entropy": 5.736508893966675,
      "epoch": 1.9688387473254232,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004608585097812907,
      "loss": 5.0641,
      "mean_token_accuracy": 0.2058962658047676,
      "num_tokens": 43898908.0,
      "step": 25305
    },
    {
      "entropy": 5.715743064880371,
      "epoch": 1.9692277766971407,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004608428874589659,
      "loss": 5.1035,
      "mean_token_accuracy": 0.20044049620628357,
      "num_tokens": 43907640.0,
      "step": 25310
    },
    {
      "entropy": 5.66007809638977,
      "epoch": 1.9696168060688581,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004608272623167711,
      "loss": 4.9666,
      "mean_token_accuracy": 0.20785339325666427,
      "num_tokens": 43915757.0,
      "step": 25315
    },
    {
      "entropy": 5.745566463470459,
      "epoch": 1.9700058354405758,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046081163435494335,
      "loss": 5.1183,
      "mean_token_accuracy": 0.19631423503160478,
      "num_tokens": 43924910.0,
      "step": 25320
    },
    {
      "entropy": 5.70884256362915,
      "epoch": 1.9703948648122933,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004607960035737198,
      "loss": 5.068,
      "mean_token_accuracy": 0.20402657389640808,
      "num_tokens": 43933122.0,
      "step": 25325
    },
    {
      "entropy": 5.6227329730987545,
      "epoch": 1.9707838941840108,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004607803699733376,
      "loss": 5.03,
      "mean_token_accuracy": 0.20439278334379196,
      "num_tokens": 43941996.0,
      "step": 25330
    },
    {
      "entropy": 5.66074948310852,
      "epoch": 1.9711729235557285,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000460764733554034,
      "loss": 4.9705,
      "mean_token_accuracy": 0.21043404787778855,
      "num_tokens": 43950637.0,
      "step": 25335
    },
    {
      "entropy": 5.715720176696777,
      "epoch": 1.9715619529274462,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046074909431604623,
      "loss": 5.0595,
      "mean_token_accuracy": 0.1962682917714119,
      "num_tokens": 43959252.0,
      "step": 25340
    },
    {
      "entropy": 5.685758447647094,
      "epoch": 1.9719509822991634,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004607334522596116,
      "loss": 5.0253,
      "mean_token_accuracy": 0.1940817043185234,
      "num_tokens": 43968391.0,
      "step": 25345
    },
    {
      "entropy": 5.683940839767456,
      "epoch": 1.9723400116708811,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004607178073849675,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20793296098709108,
      "num_tokens": 43977066.0,
      "step": 25350
    },
    {
      "entropy": 5.701573896408081,
      "epoch": 1.9727290410425988,
      "grad_norm": 1.125,
      "learning_rate": 0.0004607021596923512,
      "loss": 5.0307,
      "mean_token_accuracy": 0.2119439348578453,
      "num_tokens": 43985072.0,
      "step": 25355
    },
    {
      "entropy": 5.7974132061004635,
      "epoch": 1.9731180704143163,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004606865091820003,
      "loss": 5.2184,
      "mean_token_accuracy": 0.18994032144546508,
      "num_tokens": 43993321.0,
      "step": 25360
    },
    {
      "entropy": 5.607664346694946,
      "epoch": 1.9735070997860338,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004606708558541521,
      "loss": 4.9082,
      "mean_token_accuracy": 0.21403990387916566,
      "num_tokens": 44002476.0,
      "step": 25365
    },
    {
      "entropy": 5.690027379989624,
      "epoch": 1.9738961291577515,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004606551997090442,
      "loss": 5.1128,
      "mean_token_accuracy": 0.19725189507007598,
      "num_tokens": 44011790.0,
      "step": 25370
    },
    {
      "entropy": 5.662188100814819,
      "epoch": 1.974285158529469,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004606395407469141,
      "loss": 5.0115,
      "mean_token_accuracy": 0.1975963830947876,
      "num_tokens": 44021086.0,
      "step": 25375
    },
    {
      "entropy": 5.64773006439209,
      "epoch": 1.9746741879011864,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004606238789679995,
      "loss": 4.9785,
      "mean_token_accuracy": 0.20605355650186538,
      "num_tokens": 44029789.0,
      "step": 25380
    },
    {
      "entropy": 5.6224274158477785,
      "epoch": 1.9750632172729041,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004606082143725381,
      "loss": 5.0156,
      "mean_token_accuracy": 0.19660866856575013,
      "num_tokens": 44039508.0,
      "step": 25385
    },
    {
      "entropy": 5.671589279174805,
      "epoch": 1.9754522466446218,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004605925469607673,
      "loss": 5.0206,
      "mean_token_accuracy": 0.2061230480670929,
      "num_tokens": 44048464.0,
      "step": 25390
    },
    {
      "entropy": 5.790804386138916,
      "epoch": 1.975841276016339,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046057687673292506,
      "loss": 5.1257,
      "mean_token_accuracy": 0.19666651785373687,
      "num_tokens": 44057613.0,
      "step": 25395
    },
    {
      "entropy": 5.652014827728271,
      "epoch": 1.9762303053880568,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046056120368924907,
      "loss": 4.9155,
      "mean_token_accuracy": 0.2143044278025627,
      "num_tokens": 44066334.0,
      "step": 25400
    },
    {
      "entropy": 5.678717136383057,
      "epoch": 1.9766193347597745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004605455278299772,
      "loss": 5.1578,
      "mean_token_accuracy": 0.19586342126131057,
      "num_tokens": 44076355.0,
      "step": 25405
    },
    {
      "entropy": 5.734120035171509,
      "epoch": 1.977008364131492,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004605298491553472,
      "loss": 5.0873,
      "mean_token_accuracy": 0.19350172579288483,
      "num_tokens": 44084445.0,
      "step": 25410
    },
    {
      "entropy": 5.709806442260742,
      "epoch": 1.9773973935032094,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004605141676655971,
      "loss": 5.0202,
      "mean_token_accuracy": 0.2078096255660057,
      "num_tokens": 44093711.0,
      "step": 25415
    },
    {
      "entropy": 5.675356197357178,
      "epoch": 1.9777864228749271,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046049848336096485,
      "loss": 5.0339,
      "mean_token_accuracy": 0.1947970747947693,
      "num_tokens": 44102647.0,
      "step": 25420
    },
    {
      "entropy": 5.694596862792968,
      "epoch": 1.9781754522466446,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004604827962416883,
      "loss": 4.9447,
      "mean_token_accuracy": 0.20636706054210663,
      "num_tokens": 44110661.0,
      "step": 25425
    },
    {
      "entropy": 5.759284353256225,
      "epoch": 1.978564481618362,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004604671063080055,
      "loss": 5.1851,
      "mean_token_accuracy": 0.18675361275672914,
      "num_tokens": 44119441.0,
      "step": 25430
    },
    {
      "entropy": 5.672123956680298,
      "epoch": 1.9789535109900798,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004604514135601546,
      "loss": 4.9747,
      "mean_token_accuracy": 0.20359019935131073,
      "num_tokens": 44127774.0,
      "step": 25435
    },
    {
      "entropy": 5.746460390090943,
      "epoch": 1.9793425403617975,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004604357179983736,
      "loss": 5.2082,
      "mean_token_accuracy": 0.1801075428724289,
      "num_tokens": 44137075.0,
      "step": 25440
    },
    {
      "entropy": 5.638615465164184,
      "epoch": 1.9797315697335147,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004604200196229009,
      "loss": 4.9598,
      "mean_token_accuracy": 0.20288174152374266,
      "num_tokens": 44145647.0,
      "step": 25445
    },
    {
      "entropy": 5.658754348754883,
      "epoch": 1.9801205991052324,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004604043184339744,
      "loss": 4.9156,
      "mean_token_accuracy": 0.20674976706504822,
      "num_tokens": 44154267.0,
      "step": 25450
    },
    {
      "entropy": 5.6858209609985355,
      "epoch": 1.9805096284769501,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004603886144318325,
      "loss": 5.0673,
      "mean_token_accuracy": 0.1997123286128044,
      "num_tokens": 44162532.0,
      "step": 25455
    },
    {
      "entropy": 5.721542549133301,
      "epoch": 1.9808986578486676,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046037290761671346,
      "loss": 5.0591,
      "mean_token_accuracy": 0.20254214704036713,
      "num_tokens": 44171218.0,
      "step": 25460
    },
    {
      "entropy": 5.6671037673950195,
      "epoch": 1.981287687220385,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004603571979888556,
      "loss": 5.1139,
      "mean_token_accuracy": 0.1940985918045044,
      "num_tokens": 44180189.0,
      "step": 25465
    },
    {
      "entropy": 5.688149785995483,
      "epoch": 1.9816767165921028,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004603414855484973,
      "loss": 4.9742,
      "mean_token_accuracy": 0.20860636681318284,
      "num_tokens": 44189048.0,
      "step": 25470
    },
    {
      "entropy": 5.699688911437988,
      "epoch": 1.9820657459638202,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000460325770295877,
      "loss": 5.0603,
      "mean_token_accuracy": 0.19870493412017823,
      "num_tokens": 44197995.0,
      "step": 25475
    },
    {
      "entropy": 5.722847890853882,
      "epoch": 1.9824547753355377,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046031005223123297,
      "loss": 5.166,
      "mean_token_accuracy": 0.18669108152389527,
      "num_tokens": 44207042.0,
      "step": 25480
    },
    {
      "entropy": 5.6642577171325685,
      "epoch": 1.9828438047072554,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000460294331354804,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21300921142101287,
      "num_tokens": 44214678.0,
      "step": 25485
    },
    {
      "entropy": 5.617475605010986,
      "epoch": 1.983232834078973,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004602786076668283,
      "loss": 4.9891,
      "mean_token_accuracy": 0.20738008320331575,
      "num_tokens": 44223752.0,
      "step": 25490
    },
    {
      "entropy": 5.706676864624024,
      "epoch": 1.9836218634506906,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046026288116754477,
      "loss": 5.0557,
      "mean_token_accuracy": 0.19884685426950455,
      "num_tokens": 44232109.0,
      "step": 25495
    },
    {
      "entropy": 5.778555107116699,
      "epoch": 1.984010892822408,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004602471518571919,
      "loss": 5.1791,
      "mean_token_accuracy": 0.187676078081131,
      "num_tokens": 44239957.0,
      "step": 25500
    },
    {
      "entropy": 5.767152786254883,
      "epoch": 1.9843999221941258,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004602314197360083,
      "loss": 5.0474,
      "mean_token_accuracy": 0.20314751118421553,
      "num_tokens": 44248582.0,
      "step": 25505
    },
    {
      "entropy": 5.757918500900269,
      "epoch": 1.9847889515658432,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004602156848042328,
      "loss": 5.0455,
      "mean_token_accuracy": 0.19783039391040802,
      "num_tokens": 44256711.0,
      "step": 25510
    },
    {
      "entropy": 5.6873878002166744,
      "epoch": 1.9851779809375607,
      "grad_norm": 1.1875,
      "learning_rate": 0.000460199947062104,
      "loss": 5.0179,
      "mean_token_accuracy": 0.20318072140216828,
      "num_tokens": 44265140.0,
      "step": 25515
    },
    {
      "entropy": 5.737044477462769,
      "epoch": 1.9855670103092784,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046018420650986074,
      "loss": 5.0187,
      "mean_token_accuracy": 0.19934685230255128,
      "num_tokens": 44273560.0,
      "step": 25520
    },
    {
      "entropy": 5.635279893875122,
      "epoch": 1.9859560396809959,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000460168463147742,
      "loss": 4.9339,
      "mean_token_accuracy": 0.20691336840391159,
      "num_tokens": 44282155.0,
      "step": 25525
    },
    {
      "entropy": 5.690180063247681,
      "epoch": 1.9863450690527134,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004601527169759865,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20464201271533966,
      "num_tokens": 44291237.0,
      "step": 25530
    },
    {
      "entropy": 5.681266117095947,
      "epoch": 1.986734098424431,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004601369679948333,
      "loss": 5.0057,
      "mean_token_accuracy": 0.19905872792005538,
      "num_tokens": 44299886.0,
      "step": 25535
    },
    {
      "entropy": 5.692330741882325,
      "epoch": 1.9871231277961487,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046012121620452127,
      "loss": 5.1277,
      "mean_token_accuracy": 0.20163533538579942,
      "num_tokens": 44308247.0,
      "step": 25540
    },
    {
      "entropy": 5.717720460891724,
      "epoch": 1.9875121571678662,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004601054616052893,
      "loss": 5.0096,
      "mean_token_accuracy": 0.2031998246908188,
      "num_tokens": 44316301.0,
      "step": 25545
    },
    {
      "entropy": 5.810669040679931,
      "epoch": 1.9879011865395837,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046008970419737674,
      "loss": 5.1456,
      "mean_token_accuracy": 0.18856994509696962,
      "num_tokens": 44324758.0,
      "step": 25550
    },
    {
      "entropy": 5.75964126586914,
      "epoch": 1.9882902159113014,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004600739439810225,
      "loss": 5.1349,
      "mean_token_accuracy": 0.1874127745628357,
      "num_tokens": 44333139.0,
      "step": 25555
    },
    {
      "entropy": 5.744812822341919,
      "epoch": 1.9886792452830189,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046005818095646564,
      "loss": 5.0819,
      "mean_token_accuracy": 0.2012840449810028,
      "num_tokens": 44341992.0,
      "step": 25560
    },
    {
      "entropy": 5.684985208511352,
      "epoch": 1.9890682746547363,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046004241512394544,
      "loss": 5.051,
      "mean_token_accuracy": 0.19588932543992996,
      "num_tokens": 44350947.0,
      "step": 25565
    },
    {
      "entropy": 5.678867483139038,
      "epoch": 1.989457304026454,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004600266464837012,
      "loss": 5.0154,
      "mean_token_accuracy": 0.2034503012895584,
      "num_tokens": 44358934.0,
      "step": 25570
    },
    {
      "entropy": 5.663506078720093,
      "epoch": 1.9898463333981715,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004600108750359721,
      "loss": 4.9743,
      "mean_token_accuracy": 0.20285267978906632,
      "num_tokens": 44367117.0,
      "step": 25575
    },
    {
      "entropy": 5.7010821342468265,
      "epoch": 1.990235362769889,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045999510078099736,
      "loss": 5.0393,
      "mean_token_accuracy": 0.19596458971500397,
      "num_tokens": 44375591.0,
      "step": 25580
    },
    {
      "entropy": 5.6329389095306395,
      "epoch": 1.9906243921416067,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045997932371901645,
      "loss": 4.9312,
      "mean_token_accuracy": 0.21049901992082595,
      "num_tokens": 44384602.0,
      "step": 25585
    },
    {
      "entropy": 5.609166431427002,
      "epoch": 1.9910134215133244,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004599635438502687,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21531407684087753,
      "num_tokens": 44393451.0,
      "step": 25590
    },
    {
      "entropy": 5.630014276504516,
      "epoch": 1.9914024508850419,
      "grad_norm": 1.125,
      "learning_rate": 0.00045994776117499363,
      "loss": 4.9728,
      "mean_token_accuracy": 0.2018023505806923,
      "num_tokens": 44401638.0,
      "step": 25595
    },
    {
      "entropy": 5.661988639831543,
      "epoch": 1.9917914802567593,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045993197569343063,
      "loss": 5.001,
      "mean_token_accuracy": 0.20364686995744705,
      "num_tokens": 44410337.0,
      "step": 25600
    },
    {
      "entropy": 5.681255960464478,
      "epoch": 1.992180509628477,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045991618740581926,
      "loss": 5.0636,
      "mean_token_accuracy": 0.2019265279173851,
      "num_tokens": 44418178.0,
      "step": 25605
    },
    {
      "entropy": 5.740093803405761,
      "epoch": 1.9925695390001945,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004599003963123991,
      "loss": 5.0764,
      "mean_token_accuracy": 0.19247499257326126,
      "num_tokens": 44427056.0,
      "step": 25610
    },
    {
      "entropy": 5.769051361083984,
      "epoch": 1.992958568371912,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045988460241340966,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20148291140794755,
      "num_tokens": 44434687.0,
      "step": 25615
    },
    {
      "entropy": 5.699707174301148,
      "epoch": 1.9933475977436297,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045986880570909075,
      "loss": 5.0328,
      "mean_token_accuracy": 0.20362118631601334,
      "num_tokens": 44442763.0,
      "step": 25620
    },
    {
      "entropy": 5.558286046981811,
      "epoch": 1.9937366271153472,
      "grad_norm": 1.125,
      "learning_rate": 0.00045985300619968186,
      "loss": 4.9175,
      "mean_token_accuracy": 0.2075865775346756,
      "num_tokens": 44451850.0,
      "step": 25625
    },
    {
      "entropy": 5.602642250061035,
      "epoch": 1.9941256564870646,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045983720388542286,
      "loss": 4.9737,
      "mean_token_accuracy": 0.21108553558588028,
      "num_tokens": 44459915.0,
      "step": 25630
    },
    {
      "entropy": 5.606112337112426,
      "epoch": 1.9945146858587823,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004598213987665535,
      "loss": 5.036,
      "mean_token_accuracy": 0.20402218252420426,
      "num_tokens": 44469252.0,
      "step": 25635
    },
    {
      "entropy": 5.778771114349365,
      "epoch": 1.9949037152305,
      "grad_norm": 1.125,
      "learning_rate": 0.00045980559084331354,
      "loss": 5.1011,
      "mean_token_accuracy": 0.19799538552761078,
      "num_tokens": 44477748.0,
      "step": 25640
    },
    {
      "entropy": 5.662566614151001,
      "epoch": 1.9952927446022175,
      "grad_norm": 1.078125,
      "learning_rate": 0.000459789780115943,
      "loss": 5.0422,
      "mean_token_accuracy": 0.19278250634670258,
      "num_tokens": 44485796.0,
      "step": 25645
    },
    {
      "entropy": 5.611935758590699,
      "epoch": 1.995681773973935,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045977396658468156,
      "loss": 5.0482,
      "mean_token_accuracy": 0.19376108795404434,
      "num_tokens": 44494604.0,
      "step": 25650
    },
    {
      "entropy": 5.655222320556641,
      "epoch": 1.9960708033456527,
      "grad_norm": 1.125,
      "learning_rate": 0.0004597581502497693,
      "loss": 5.0303,
      "mean_token_accuracy": 0.19579687863588333,
      "num_tokens": 44503728.0,
      "step": 25655
    },
    {
      "entropy": 5.707696533203125,
      "epoch": 1.9964598327173702,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004597423311114462,
      "loss": 5.0072,
      "mean_token_accuracy": 0.20464685261249543,
      "num_tokens": 44512201.0,
      "step": 25660
    },
    {
      "entropy": 5.715597152709961,
      "epoch": 1.9968488620890876,
      "grad_norm": 1.25,
      "learning_rate": 0.0004597265091699522,
      "loss": 5.036,
      "mean_token_accuracy": 0.19904591143131256,
      "num_tokens": 44520122.0,
      "step": 25665
    },
    {
      "entropy": 5.605073499679565,
      "epoch": 1.9972378914608053,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004597106844255276,
      "loss": 5.038,
      "mean_token_accuracy": 0.19938948899507522,
      "num_tokens": 44528928.0,
      "step": 25670
    },
    {
      "entropy": 5.731636619567871,
      "epoch": 1.9976269208325228,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045969485687841227,
      "loss": 5.1668,
      "mean_token_accuracy": 0.19572292417287826,
      "num_tokens": 44537330.0,
      "step": 25675
    },
    {
      "entropy": 5.720711374282837,
      "epoch": 1.9980159502042403,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045967902652884645,
      "loss": 5.0309,
      "mean_token_accuracy": 0.20138122886419296,
      "num_tokens": 44545748.0,
      "step": 25680
    },
    {
      "entropy": 5.6698570728302,
      "epoch": 1.998404979575958,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004596631933770704,
      "loss": 5.1346,
      "mean_token_accuracy": 0.19554666429758072,
      "num_tokens": 44554562.0,
      "step": 25685
    },
    {
      "entropy": 5.651662397384643,
      "epoch": 1.9987940089476757,
      "grad_norm": 1.125,
      "learning_rate": 0.00045964735742332427,
      "loss": 4.9729,
      "mean_token_accuracy": 0.20400928407907487,
      "num_tokens": 44563544.0,
      "step": 25690
    },
    {
      "entropy": 5.717379760742188,
      "epoch": 1.9991830383193931,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004596315186678484,
      "loss": 5.0522,
      "mean_token_accuracy": 0.1999122068285942,
      "num_tokens": 44571179.0,
      "step": 25695
    },
    {
      "entropy": 5.720243692398071,
      "epoch": 1.9995720676911106,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045961567711088307,
      "loss": 5.0894,
      "mean_token_accuracy": 0.20062737315893173,
      "num_tokens": 44580237.0,
      "step": 25700
    },
    {
      "entropy": 5.675827741622925,
      "epoch": 1.9999610970628283,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045959983275266873,
      "loss": 5.0272,
      "mean_token_accuracy": 0.20159684717655182,
      "num_tokens": 44588362.0,
      "step": 25705
    },
    {
      "entropy": 5.671978367699517,
      "epoch": 2.000311223497374,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045958398559344573,
      "loss": 4.9589,
      "mean_token_accuracy": 0.20681650274329716,
      "num_tokens": 44596037.0,
      "step": 25710
    },
    {
      "entropy": 5.734713172912597,
      "epoch": 2.0007002528690916,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045956813563345454,
      "loss": 5.0188,
      "mean_token_accuracy": 0.20624375194311143,
      "num_tokens": 44604913.0,
      "step": 25715
    },
    {
      "entropy": 5.71753339767456,
      "epoch": 2.0010892822408093,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004595522828729357,
      "loss": 5.0555,
      "mean_token_accuracy": 0.20264316648244857,
      "num_tokens": 44613523.0,
      "step": 25720
    },
    {
      "entropy": 5.654740905761718,
      "epoch": 2.0014783116125265,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004595364273121296,
      "loss": 4.8569,
      "mean_token_accuracy": 0.21545921862125397,
      "num_tokens": 44622259.0,
      "step": 25725
    },
    {
      "entropy": 5.616589832305908,
      "epoch": 2.0018673409842442,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004595205689512771,
      "loss": 4.9408,
      "mean_token_accuracy": 0.2076679602265358,
      "num_tokens": 44630793.0,
      "step": 25730
    },
    {
      "entropy": 5.678444528579712,
      "epoch": 2.002256370355962,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045950470779061855,
      "loss": 4.9575,
      "mean_token_accuracy": 0.20859505981206894,
      "num_tokens": 44638986.0,
      "step": 25735
    },
    {
      "entropy": 5.705001640319824,
      "epoch": 2.0026453997276796,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045948884383039475,
      "loss": 5.0253,
      "mean_token_accuracy": 0.19875845611095427,
      "num_tokens": 44647491.0,
      "step": 25740
    },
    {
      "entropy": 5.613849735260009,
      "epoch": 2.003034429099397,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045947297707084637,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2067682147026062,
      "num_tokens": 44655431.0,
      "step": 25745
    },
    {
      "entropy": 5.634070634841919,
      "epoch": 2.0034234584711146,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004594571075122142,
      "loss": 4.9621,
      "mean_token_accuracy": 0.20248010754585266,
      "num_tokens": 44664398.0,
      "step": 25750
    },
    {
      "entropy": 5.693981504440307,
      "epoch": 2.0038124878428323,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045944123515473905,
      "loss": 4.9518,
      "mean_token_accuracy": 0.2038433238863945,
      "num_tokens": 44673372.0,
      "step": 25755
    },
    {
      "entropy": 5.611620903015137,
      "epoch": 2.0042015172145495,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045942535999866175,
      "loss": 4.8688,
      "mean_token_accuracy": 0.21319093853235244,
      "num_tokens": 44681736.0,
      "step": 25760
    },
    {
      "entropy": 5.6853429794311525,
      "epoch": 2.0045905465862672,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004594094820442231,
      "loss": 5.0136,
      "mean_token_accuracy": 0.20086731463670732,
      "num_tokens": 44691858.0,
      "step": 25765
    },
    {
      "entropy": 5.746147680282593,
      "epoch": 2.004979575957985,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000459393601291664,
      "loss": 5.0466,
      "mean_token_accuracy": 0.19366568475961685,
      "num_tokens": 44701002.0,
      "step": 25770
    },
    {
      "entropy": 5.686776065826416,
      "epoch": 2.005368605329702,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045937771774122563,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21595702320337296,
      "num_tokens": 44709158.0,
      "step": 25775
    },
    {
      "entropy": 5.638172149658203,
      "epoch": 2.00575763470142,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004593618313931488,
      "loss": 4.9622,
      "mean_token_accuracy": 0.20781536400318146,
      "num_tokens": 44717495.0,
      "step": 25780
    },
    {
      "entropy": 5.687889480590821,
      "epoch": 2.0061466640731376,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045934594224767463,
      "loss": 4.9503,
      "mean_token_accuracy": 0.20594877898693084,
      "num_tokens": 44726040.0,
      "step": 25785
    },
    {
      "entropy": 5.675553703308106,
      "epoch": 2.0065356934448553,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045933005030504424,
      "loss": 4.9884,
      "mean_token_accuracy": 0.1997014030814171,
      "num_tokens": 44735098.0,
      "step": 25790
    },
    {
      "entropy": 5.649155521392823,
      "epoch": 2.0069247228165725,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045931415556549863,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20108432918787003,
      "num_tokens": 44743764.0,
      "step": 25795
    },
    {
      "entropy": 5.643327617645264,
      "epoch": 2.0073137521882902,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004592982580292792,
      "loss": 4.8748,
      "mean_token_accuracy": 0.20692960619926454,
      "num_tokens": 44751554.0,
      "step": 25800
    },
    {
      "entropy": 5.734205865859986,
      "epoch": 2.007702781560008,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045928235769662697,
      "loss": 5.0574,
      "mean_token_accuracy": 0.19601832032203675,
      "num_tokens": 44760158.0,
      "step": 25805
    },
    {
      "entropy": 5.66917781829834,
      "epoch": 2.008091810931725,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045926645456778327,
      "loss": 4.9414,
      "mean_token_accuracy": 0.20913155674934386,
      "num_tokens": 44769209.0,
      "step": 25810
    },
    {
      "entropy": 5.6618571281433105,
      "epoch": 2.008480840303443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045925054864298946,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20589143335819243,
      "num_tokens": 44777991.0,
      "step": 25815
    },
    {
      "entropy": 5.707407665252686,
      "epoch": 2.0088698696751606,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045923463992248684,
      "loss": 5.0302,
      "mean_token_accuracy": 0.19930059015750884,
      "num_tokens": 44786492.0,
      "step": 25820
    },
    {
      "entropy": 5.710167074203492,
      "epoch": 2.009258899046878,
      "grad_norm": 1.125,
      "learning_rate": 0.00045921872840651675,
      "loss": 5.0604,
      "mean_token_accuracy": 0.20184142887592316,
      "num_tokens": 44795021.0,
      "step": 25825
    },
    {
      "entropy": 5.6584185600280765,
      "epoch": 2.0096479284185955,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045920281409532064,
      "loss": 4.9671,
      "mean_token_accuracy": 0.20276207625865936,
      "num_tokens": 44803267.0,
      "step": 25830
    },
    {
      "entropy": 5.714813375473023,
      "epoch": 2.010036957790313,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004591868969891401,
      "loss": 5.0269,
      "mean_token_accuracy": 0.20033709108829498,
      "num_tokens": 44811954.0,
      "step": 25835
    },
    {
      "entropy": 5.643784618377685,
      "epoch": 2.010425987162031,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004591709770882165,
      "loss": 4.8838,
      "mean_token_accuracy": 0.2096843644976616,
      "num_tokens": 44820748.0,
      "step": 25840
    },
    {
      "entropy": 5.6935258388519285,
      "epoch": 2.010815016533748,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004591550543927915,
      "loss": 5.0806,
      "mean_token_accuracy": 0.19925837367773055,
      "num_tokens": 44829400.0,
      "step": 25845
    },
    {
      "entropy": 5.733881187438965,
      "epoch": 2.011204045905466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004591391289031067,
      "loss": 4.997,
      "mean_token_accuracy": 0.2024434193968773,
      "num_tokens": 44838663.0,
      "step": 25850
    },
    {
      "entropy": 5.700822496414185,
      "epoch": 2.0115930752771836,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004591232006194036,
      "loss": 4.9219,
      "mean_token_accuracy": 0.21281079053878785,
      "num_tokens": 44847365.0,
      "step": 25855
    },
    {
      "entropy": 5.696553325653076,
      "epoch": 2.011982104648901,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045910726954192404,
      "loss": 4.9795,
      "mean_token_accuracy": 0.20659883618354796,
      "num_tokens": 44856093.0,
      "step": 25860
    },
    {
      "entropy": 5.61371603012085,
      "epoch": 2.0123711340206185,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004590913356709097,
      "loss": 4.9443,
      "mean_token_accuracy": 0.20985369086265565,
      "num_tokens": 44865116.0,
      "step": 25865
    },
    {
      "entropy": 5.654669570922851,
      "epoch": 2.012760163392336,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045907539900660237,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20194179862737655,
      "num_tokens": 44874037.0,
      "step": 25870
    },
    {
      "entropy": 5.649487161636353,
      "epoch": 2.013149192764054,
      "grad_norm": 1.25,
      "learning_rate": 0.0004590594595492438,
      "loss": 4.8503,
      "mean_token_accuracy": 0.21097995042800904,
      "num_tokens": 44882603.0,
      "step": 25875
    },
    {
      "entropy": 5.667109632492066,
      "epoch": 2.013538222135771,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045904351729907593,
      "loss": 4.8998,
      "mean_token_accuracy": 0.21082854717969896,
      "num_tokens": 44890645.0,
      "step": 25880
    },
    {
      "entropy": 5.732386732101441,
      "epoch": 2.013927251507489,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045902757225634066,
      "loss": 5.0209,
      "mean_token_accuracy": 0.20473106652498246,
      "num_tokens": 44898972.0,
      "step": 25885
    },
    {
      "entropy": 5.632081985473633,
      "epoch": 2.0143162808792066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004590116244212799,
      "loss": 5.0007,
      "mean_token_accuracy": 0.21055980771780014,
      "num_tokens": 44907427.0,
      "step": 25890
    },
    {
      "entropy": 5.697575187683105,
      "epoch": 2.014705310250924,
      "grad_norm": 1.25,
      "learning_rate": 0.0004589956737941355,
      "loss": 5.0422,
      "mean_token_accuracy": 0.19956227391958237,
      "num_tokens": 44916465.0,
      "step": 25895
    },
    {
      "entropy": 5.7018509864807125,
      "epoch": 2.0150943396226415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004589797203751497,
      "loss": 4.959,
      "mean_token_accuracy": 0.21271492838859557,
      "num_tokens": 44925102.0,
      "step": 25900
    },
    {
      "entropy": 5.651805591583252,
      "epoch": 2.015483368994359,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004589637641645645,
      "loss": 4.9309,
      "mean_token_accuracy": 0.21065156310796737,
      "num_tokens": 44933061.0,
      "step": 25905
    },
    {
      "entropy": 5.5902876377105715,
      "epoch": 2.0158723983660765,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045894780516262193,
      "loss": 4.9805,
      "mean_token_accuracy": 0.20562071949243546,
      "num_tokens": 44942342.0,
      "step": 25910
    },
    {
      "entropy": 5.677871322631836,
      "epoch": 2.016261427737794,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004589318433695642,
      "loss": 4.9725,
      "mean_token_accuracy": 0.2132783353328705,
      "num_tokens": 44951163.0,
      "step": 25915
    },
    {
      "entropy": 5.632634162902832,
      "epoch": 2.016650457109512,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004589158787856336,
      "loss": 4.9429,
      "mean_token_accuracy": 0.20722294002771377,
      "num_tokens": 44960208.0,
      "step": 25920
    },
    {
      "entropy": 5.613522672653199,
      "epoch": 2.0170394864812295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004588999114110721,
      "loss": 4.8495,
      "mean_token_accuracy": 0.21180173009634018,
      "num_tokens": 44969051.0,
      "step": 25925
    },
    {
      "entropy": 5.619331979751587,
      "epoch": 2.017428515852947,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004588839412461223,
      "loss": 4.895,
      "mean_token_accuracy": 0.21775769293308259,
      "num_tokens": 44977980.0,
      "step": 25930
    },
    {
      "entropy": 5.628773593902588,
      "epoch": 2.0178175452246645,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004588679682910264,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20029470175504685,
      "num_tokens": 44986520.0,
      "step": 25935
    },
    {
      "entropy": 5.675970983505249,
      "epoch": 2.018206574596382,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004588519925460266,
      "loss": 4.9658,
      "mean_token_accuracy": 0.20854251384735106,
      "num_tokens": 44995011.0,
      "step": 25940
    },
    {
      "entropy": 5.600999116897583,
      "epoch": 2.0185956039680995,
      "grad_norm": 1.125,
      "learning_rate": 0.0004588360140113655,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21334500908851622,
      "num_tokens": 45004611.0,
      "step": 25945
    },
    {
      "entropy": 5.694470453262329,
      "epoch": 2.018984633339817,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004588200326872855,
      "loss": 4.9996,
      "mean_token_accuracy": 0.19941993951797485,
      "num_tokens": 45013832.0,
      "step": 25950
    },
    {
      "entropy": 5.692751932144165,
      "epoch": 2.019373662711535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004588040485740291,
      "loss": 4.938,
      "mean_token_accuracy": 0.20973325073719024,
      "num_tokens": 45022751.0,
      "step": 25955
    },
    {
      "entropy": 5.710015678405762,
      "epoch": 2.019762692083252,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004587880616718387,
      "loss": 5.0067,
      "mean_token_accuracy": 0.20685900747776031,
      "num_tokens": 45030826.0,
      "step": 25960
    },
    {
      "entropy": 5.694577550888061,
      "epoch": 2.02015172145497,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004587720719809572,
      "loss": 4.9807,
      "mean_token_accuracy": 0.20188210159540176,
      "num_tokens": 45039625.0,
      "step": 25965
    },
    {
      "entropy": 5.645271158218383,
      "epoch": 2.0205407508266875,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004587560795016269,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21033827662467958,
      "num_tokens": 45048229.0,
      "step": 25970
    },
    {
      "entropy": 5.710571336746216,
      "epoch": 2.020929780198405,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004587400842340905,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19907742440700532,
      "num_tokens": 45057414.0,
      "step": 25975
    },
    {
      "entropy": 5.616314888000488,
      "epoch": 2.0213188095701224,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045872408617859083,
      "loss": 4.8948,
      "mean_token_accuracy": 0.206028513610363,
      "num_tokens": 45065785.0,
      "step": 25980
    },
    {
      "entropy": 5.604463863372803,
      "epoch": 2.02170783894184,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045870808533537066,
      "loss": 4.9172,
      "mean_token_accuracy": 0.20810531228780746,
      "num_tokens": 45074896.0,
      "step": 25985
    },
    {
      "entropy": 5.593366909027099,
      "epoch": 2.022096868313558,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045869208170467256,
      "loss": 4.9534,
      "mean_token_accuracy": 0.20079635828733444,
      "num_tokens": 45083554.0,
      "step": 25990
    },
    {
      "entropy": 5.705989122390747,
      "epoch": 2.022485897685275,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004586760752867396,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2041795775294304,
      "num_tokens": 45092204.0,
      "step": 25995
    },
    {
      "entropy": 5.691451597213745,
      "epoch": 2.022874927056993,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045866006608181456,
      "loss": 4.9848,
      "mean_token_accuracy": 0.19816917479038237,
      "num_tokens": 45101665.0,
      "step": 26000
    },
    {
      "entropy": 5.663026142120361,
      "epoch": 2.0232639564287105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004586440540901403,
      "loss": 5.0325,
      "mean_token_accuracy": 0.19990342557430268,
      "num_tokens": 45111093.0,
      "step": 26005
    },
    {
      "entropy": 5.673209238052368,
      "epoch": 2.0236529858004277,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045862803931195976,
      "loss": 4.9594,
      "mean_token_accuracy": 0.20590735971927643,
      "num_tokens": 45118826.0,
      "step": 26010
    },
    {
      "entropy": 5.697975730895996,
      "epoch": 2.0240420151721454,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004586120217475161,
      "loss": 4.9947,
      "mean_token_accuracy": 0.20848352313041688,
      "num_tokens": 45128265.0,
      "step": 26015
    },
    {
      "entropy": 5.714987516403198,
      "epoch": 2.024431044543863,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004585960013970522,
      "loss": 5.0309,
      "mean_token_accuracy": 0.19488996118307114,
      "num_tokens": 45137922.0,
      "step": 26020
    },
    {
      "entropy": 5.611620473861694,
      "epoch": 2.024820073915581,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004585799782608112,
      "loss": 4.8571,
      "mean_token_accuracy": 0.21922712475061418,
      "num_tokens": 45146242.0,
      "step": 26025
    },
    {
      "entropy": 5.692760896682739,
      "epoch": 2.025209103287298,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045856395233903624,
      "loss": 5.0341,
      "mean_token_accuracy": 0.19816716462373735,
      "num_tokens": 45154897.0,
      "step": 26030
    },
    {
      "entropy": 5.678602266311645,
      "epoch": 2.025598132659016,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004585479236319705,
      "loss": 4.9918,
      "mean_token_accuracy": 0.2063049018383026,
      "num_tokens": 45164786.0,
      "step": 26035
    },
    {
      "entropy": 5.702360534667969,
      "epoch": 2.0259871620307335,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045853189213985714,
      "loss": 5.0272,
      "mean_token_accuracy": 0.19565706849098205,
      "num_tokens": 45173652.0,
      "step": 26040
    },
    {
      "entropy": 5.672584581375122,
      "epoch": 2.0263761914024507,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045851585786293943,
      "loss": 4.902,
      "mean_token_accuracy": 0.21253832131624223,
      "num_tokens": 45182399.0,
      "step": 26045
    },
    {
      "entropy": 5.6637773513793945,
      "epoch": 2.0267652207741684,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045849982080146067,
      "loss": 4.9684,
      "mean_token_accuracy": 0.20364944636821747,
      "num_tokens": 45191587.0,
      "step": 26050
    },
    {
      "entropy": 5.638995122909546,
      "epoch": 2.027154250145886,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004584837809556642,
      "loss": 4.944,
      "mean_token_accuracy": 0.20783798694610595,
      "num_tokens": 45200032.0,
      "step": 26055
    },
    {
      "entropy": 5.634393167495728,
      "epoch": 2.0275432795176034,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045846773832579346,
      "loss": 4.9616,
      "mean_token_accuracy": 0.2055723026394844,
      "num_tokens": 45209205.0,
      "step": 26060
    },
    {
      "entropy": 5.721559238433838,
      "epoch": 2.027932308889321,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045845169291209176,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20819129943847656,
      "num_tokens": 45217603.0,
      "step": 26065
    },
    {
      "entropy": 5.669111919403076,
      "epoch": 2.0283213382610388,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045843564471480263,
      "loss": 4.9232,
      "mean_token_accuracy": 0.2025085687637329,
      "num_tokens": 45225971.0,
      "step": 26070
    },
    {
      "entropy": 5.698772192001343,
      "epoch": 2.0287103676327565,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004584195937341695,
      "loss": 4.9969,
      "mean_token_accuracy": 0.20161072462797164,
      "num_tokens": 45234494.0,
      "step": 26075
    },
    {
      "entropy": 5.617690944671631,
      "epoch": 2.0290993970044737,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045840353997043606,
      "loss": 4.8215,
      "mean_token_accuracy": 0.2159263789653778,
      "num_tokens": 45242685.0,
      "step": 26080
    },
    {
      "entropy": 5.650600004196167,
      "epoch": 2.0294884263761914,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004583874834238458,
      "loss": 4.9071,
      "mean_token_accuracy": 0.2079970806837082,
      "num_tokens": 45252282.0,
      "step": 26085
    },
    {
      "entropy": 5.685449504852295,
      "epoch": 2.029877455747909,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004583714240946423,
      "loss": 5.0674,
      "mean_token_accuracy": 0.19315167367458344,
      "num_tokens": 45261390.0,
      "step": 26090
    },
    {
      "entropy": 5.718031501770019,
      "epoch": 2.0302664851196264,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045835536198306936,
      "loss": 5.078,
      "mean_token_accuracy": 0.19824135452508926,
      "num_tokens": 45270244.0,
      "step": 26095
    },
    {
      "entropy": 5.705060958862305,
      "epoch": 2.030655514491344,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045833929708937067,
      "loss": 4.99,
      "mean_token_accuracy": 0.1969300001859665,
      "num_tokens": 45279404.0,
      "step": 26100
    },
    {
      "entropy": 5.692380475997925,
      "epoch": 2.0310445438630618,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004583232294137899,
      "loss": 4.957,
      "mean_token_accuracy": 0.21014004945755005,
      "num_tokens": 45288064.0,
      "step": 26105
    },
    {
      "entropy": 5.68449559211731,
      "epoch": 2.031433573234779,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004583071589565709,
      "loss": 4.9963,
      "mean_token_accuracy": 0.20225270539522172,
      "num_tokens": 45296528.0,
      "step": 26110
    },
    {
      "entropy": 5.6436333656311035,
      "epoch": 2.0318226026064967,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004582910857179576,
      "loss": 4.9103,
      "mean_token_accuracy": 0.20844800472259523,
      "num_tokens": 45305524.0,
      "step": 26115
    },
    {
      "entropy": 5.7331455707550045,
      "epoch": 2.0322116319782144,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004582750096981938,
      "loss": 5.0968,
      "mean_token_accuracy": 0.19437944144010544,
      "num_tokens": 45314386.0,
      "step": 26120
    },
    {
      "entropy": 5.68181471824646,
      "epoch": 2.032600661349932,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004582589308975234,
      "loss": 4.9866,
      "mean_token_accuracy": 0.20569924265146255,
      "num_tokens": 45322901.0,
      "step": 26125
    },
    {
      "entropy": 5.6107419490814205,
      "epoch": 2.0329896907216494,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045824284931619044,
      "loss": 4.92,
      "mean_token_accuracy": 0.20761809945106507,
      "num_tokens": 45331509.0,
      "step": 26130
    },
    {
      "entropy": 5.6627459049224855,
      "epoch": 2.033378720093367,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045822676495443893,
      "loss": 4.9699,
      "mean_token_accuracy": 0.2066069394350052,
      "num_tokens": 45340356.0,
      "step": 26135
    },
    {
      "entropy": 5.753885221481323,
      "epoch": 2.0337677494650848,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045821067781251284,
      "loss": 5.0075,
      "mean_token_accuracy": 0.20187541842460632,
      "num_tokens": 45349890.0,
      "step": 26140
    },
    {
      "entropy": 5.671606969833374,
      "epoch": 2.034156778836802,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045819458789065633,
      "loss": 4.9676,
      "mean_token_accuracy": 0.20906978994607925,
      "num_tokens": 45359249.0,
      "step": 26145
    },
    {
      "entropy": 5.65541410446167,
      "epoch": 2.0345458082085197,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004581784951891135,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21289274096488953,
      "num_tokens": 45367832.0,
      "step": 26150
    },
    {
      "entropy": 5.662096929550171,
      "epoch": 2.0349348375802374,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004581623997081286,
      "loss": 4.9549,
      "mean_token_accuracy": 0.20920063853263854,
      "num_tokens": 45376316.0,
      "step": 26155
    },
    {
      "entropy": 5.710555791854858,
      "epoch": 2.0353238669519547,
      "grad_norm": 1.125,
      "learning_rate": 0.0004581463014479459,
      "loss": 5.0221,
      "mean_token_accuracy": 0.19455120116472244,
      "num_tokens": 45385801.0,
      "step": 26160
    },
    {
      "entropy": 5.6823070526123045,
      "epoch": 2.0357128963236724,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004581302004088095,
      "loss": 4.984,
      "mean_token_accuracy": 0.20982643812894822,
      "num_tokens": 45394669.0,
      "step": 26165
    },
    {
      "entropy": 5.644961738586426,
      "epoch": 2.03610192569539,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004581140965909638,
      "loss": 4.945,
      "mean_token_accuracy": 0.2074610635638237,
      "num_tokens": 45402896.0,
      "step": 26170
    },
    {
      "entropy": 5.637491273880005,
      "epoch": 2.0364909550671078,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004580979899946531,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20109013319015503,
      "num_tokens": 45411047.0,
      "step": 26175
    },
    {
      "entropy": 5.71692385673523,
      "epoch": 2.036879984438825,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004580818806201219,
      "loss": 5.0923,
      "mean_token_accuracy": 0.19554199278354645,
      "num_tokens": 45420661.0,
      "step": 26180
    },
    {
      "entropy": 5.729820346832275,
      "epoch": 2.0372690138105427,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045806576846761453,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21313071995973587,
      "num_tokens": 45429199.0,
      "step": 26185
    },
    {
      "entropy": 5.69167308807373,
      "epoch": 2.0376580431822604,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045804965353737556,
      "loss": 4.9362,
      "mean_token_accuracy": 0.20775840878486634,
      "num_tokens": 45437504.0,
      "step": 26190
    },
    {
      "entropy": 5.589030981063843,
      "epoch": 2.0380470725539777,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004580335358296494,
      "loss": 4.8991,
      "mean_token_accuracy": 0.20800567418336868,
      "num_tokens": 45445696.0,
      "step": 26195
    },
    {
      "entropy": 5.662701892852783,
      "epoch": 2.0384361019256954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045801741534468065,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20205149352550505,
      "num_tokens": 45455058.0,
      "step": 26200
    },
    {
      "entropy": 5.829818964004517,
      "epoch": 2.038825131297413,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004580012920827139,
      "loss": 5.072,
      "mean_token_accuracy": 0.20382075756788254,
      "num_tokens": 45464193.0,
      "step": 26205
    },
    {
      "entropy": 5.601741695404053,
      "epoch": 2.0392141606691303,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004579851660439939,
      "loss": 4.8451,
      "mean_token_accuracy": 0.2141188085079193,
      "num_tokens": 45471922.0,
      "step": 26210
    },
    {
      "entropy": 5.674674701690674,
      "epoch": 2.039603190040848,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045796903722876523,
      "loss": 5.0304,
      "mean_token_accuracy": 0.19680144488811493,
      "num_tokens": 45480181.0,
      "step": 26215
    },
    {
      "entropy": 5.8024567604064945,
      "epoch": 2.0399922194125657,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004579529056372726,
      "loss": 5.0355,
      "mean_token_accuracy": 0.20499414950609207,
      "num_tokens": 45488970.0,
      "step": 26220
    },
    {
      "entropy": 5.793949937820434,
      "epoch": 2.0403812487842834,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004579367712697609,
      "loss": 4.9593,
      "mean_token_accuracy": 0.20311334282159804,
      "num_tokens": 45497066.0,
      "step": 26225
    },
    {
      "entropy": 5.6367613792419435,
      "epoch": 2.0407702781560006,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045792063412647475,
      "loss": 4.9373,
      "mean_token_accuracy": 0.2043283000588417,
      "num_tokens": 45506399.0,
      "step": 26230
    },
    {
      "entropy": 5.648182439804077,
      "epoch": 2.0411593075277183,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045790449420765925,
      "loss": 4.9665,
      "mean_token_accuracy": 0.20089273899793625,
      "num_tokens": 45516625.0,
      "step": 26235
    },
    {
      "entropy": 5.75351128578186,
      "epoch": 2.041548336899436,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045788835151355914,
      "loss": 5.0613,
      "mean_token_accuracy": 0.19787285923957826,
      "num_tokens": 45525539.0,
      "step": 26240
    },
    {
      "entropy": 5.7503454208374025,
      "epoch": 2.0419373662711533,
      "grad_norm": 1.125,
      "learning_rate": 0.00045787220604441933,
      "loss": 4.9804,
      "mean_token_accuracy": 0.2094811752438545,
      "num_tokens": 45535388.0,
      "step": 26245
    },
    {
      "entropy": 5.701186466217041,
      "epoch": 2.042326395642871,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045785605780048497,
      "loss": 5.0224,
      "mean_token_accuracy": 0.2008216053247452,
      "num_tokens": 45543729.0,
      "step": 26250
    },
    {
      "entropy": 5.702984809875488,
      "epoch": 2.0427154250145887,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045783990678200086,
      "loss": 5.0288,
      "mean_token_accuracy": 0.2026811271905899,
      "num_tokens": 45551887.0,
      "step": 26255
    },
    {
      "entropy": 5.783022403717041,
      "epoch": 2.043104454386306,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045782375298921227,
      "loss": 5.0823,
      "mean_token_accuracy": 0.19830229729413987,
      "num_tokens": 45560267.0,
      "step": 26260
    },
    {
      "entropy": 5.711447811126709,
      "epoch": 2.0434934837580236,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004578075964223642,
      "loss": 4.9902,
      "mean_token_accuracy": 0.1970672056078911,
      "num_tokens": 45569459.0,
      "step": 26265
    },
    {
      "entropy": 5.770064115524292,
      "epoch": 2.0438825131297413,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004577914370817018,
      "loss": 4.9751,
      "mean_token_accuracy": 0.1991291642189026,
      "num_tokens": 45577713.0,
      "step": 26270
    },
    {
      "entropy": 5.645679521560669,
      "epoch": 2.044271542501459,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045777527496747034,
      "loss": 4.9147,
      "mean_token_accuracy": 0.2066315308213234,
      "num_tokens": 45586521.0,
      "step": 26275
    },
    {
      "entropy": 5.677498483657837,
      "epoch": 2.0446605718731763,
      "grad_norm": 1.25,
      "learning_rate": 0.00045775911007991493,
      "loss": 4.9637,
      "mean_token_accuracy": 0.20217486321926117,
      "num_tokens": 45595210.0,
      "step": 26280
    },
    {
      "entropy": 5.689458227157592,
      "epoch": 2.045049601244894,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045774294241928093,
      "loss": 5.0127,
      "mean_token_accuracy": 0.2055308476090431,
      "num_tokens": 45603472.0,
      "step": 26285
    },
    {
      "entropy": 5.681685447692871,
      "epoch": 2.0454386306166117,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004577267719858137,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19471936970949172,
      "num_tokens": 45612149.0,
      "step": 26290
    },
    {
      "entropy": 5.779747676849365,
      "epoch": 2.045827659988329,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045771059877975853,
      "loss": 5.0731,
      "mean_token_accuracy": 0.19781418293714523,
      "num_tokens": 45619903.0,
      "step": 26295
    },
    {
      "entropy": 5.692175817489624,
      "epoch": 2.0462166893600466,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004576944228013608,
      "loss": 4.9637,
      "mean_token_accuracy": 0.21066200584173203,
      "num_tokens": 45628012.0,
      "step": 26300
    },
    {
      "entropy": 5.642592239379883,
      "epoch": 2.0466057187317643,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000457678244050866,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20328520089387894,
      "num_tokens": 45636587.0,
      "step": 26305
    },
    {
      "entropy": 5.67674036026001,
      "epoch": 2.046994748103482,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004576620625285196,
      "loss": 4.9494,
      "mean_token_accuracy": 0.20184803009033203,
      "num_tokens": 45645189.0,
      "step": 26310
    },
    {
      "entropy": 5.750026893615723,
      "epoch": 2.0473837774751993,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045764587823456717,
      "loss": 5.0311,
      "mean_token_accuracy": 0.2033383622765541,
      "num_tokens": 45654059.0,
      "step": 26315
    },
    {
      "entropy": 5.695265674591065,
      "epoch": 2.047772806846917,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045762969116925424,
      "loss": 4.9006,
      "mean_token_accuracy": 0.21536493450403213,
      "num_tokens": 45662125.0,
      "step": 26320
    },
    {
      "entropy": 5.7020679950714115,
      "epoch": 2.0481618362186347,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045761350133282643,
      "loss": 4.9797,
      "mean_token_accuracy": 0.20785591900348663,
      "num_tokens": 45670068.0,
      "step": 26325
    },
    {
      "entropy": 5.656160736083985,
      "epoch": 2.048550865590352,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045759730872552937,
      "loss": 4.8675,
      "mean_token_accuracy": 0.21632105708122254,
      "num_tokens": 45678717.0,
      "step": 26330
    },
    {
      "entropy": 5.645941638946534,
      "epoch": 2.0489398949620696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004575811133476088,
      "loss": 4.9587,
      "mean_token_accuracy": 0.20723410546779633,
      "num_tokens": 45686882.0,
      "step": 26335
    },
    {
      "entropy": 5.601016664505005,
      "epoch": 2.0493289243337873,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045756491519931047,
      "loss": 4.8898,
      "mean_token_accuracy": 0.20890021324157715,
      "num_tokens": 45695460.0,
      "step": 26340
    },
    {
      "entropy": 5.756913805007935,
      "epoch": 2.0497179537055046,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004575487142808801,
      "loss": 5.0525,
      "mean_token_accuracy": 0.19704596996307372,
      "num_tokens": 45703287.0,
      "step": 26345
    },
    {
      "entropy": 5.727641439437866,
      "epoch": 2.0501069830772223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004575325105925636,
      "loss": 5.0713,
      "mean_token_accuracy": 0.19566494226455688,
      "num_tokens": 45712830.0,
      "step": 26350
    },
    {
      "entropy": 5.6460315704345705,
      "epoch": 2.05049601244894,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045751630413460665,
      "loss": 5.0651,
      "mean_token_accuracy": 0.19860530644655228,
      "num_tokens": 45720701.0,
      "step": 26355
    },
    {
      "entropy": 5.6865331649780275,
      "epoch": 2.0508850418206577,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004575000949072554,
      "loss": 4.9264,
      "mean_token_accuracy": 0.2069900169968605,
      "num_tokens": 45728663.0,
      "step": 26360
    },
    {
      "entropy": 5.6829897403717045,
      "epoch": 2.051274071192375,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004574838829107557,
      "loss": 4.9779,
      "mean_token_accuracy": 0.2015938863158226,
      "num_tokens": 45737661.0,
      "step": 26365
    },
    {
      "entropy": 5.599342250823975,
      "epoch": 2.0516631005640926,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004574676681453535,
      "loss": 4.9297,
      "mean_token_accuracy": 0.2127887412905693,
      "num_tokens": 45746343.0,
      "step": 26370
    },
    {
      "entropy": 5.622148895263672,
      "epoch": 2.0520521299358103,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045745145061129485,
      "loss": 4.9654,
      "mean_token_accuracy": 0.2032032698392868,
      "num_tokens": 45755025.0,
      "step": 26375
    },
    {
      "entropy": 5.705486536026001,
      "epoch": 2.0524411593075276,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045743523030882584,
      "loss": 5.0355,
      "mean_token_accuracy": 0.20282019525766373,
      "num_tokens": 45763107.0,
      "step": 26380
    },
    {
      "entropy": 5.6822922706604,
      "epoch": 2.0528301886792453,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004574190072381926,
      "loss": 4.9203,
      "mean_token_accuracy": 0.21141140460968016,
      "num_tokens": 45771754.0,
      "step": 26385
    },
    {
      "entropy": 5.688706350326538,
      "epoch": 2.053219218050963,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004574027813996413,
      "loss": 5.0893,
      "mean_token_accuracy": 0.20015982836484908,
      "num_tokens": 45779282.0,
      "step": 26390
    },
    {
      "entropy": 5.707024145126343,
      "epoch": 2.05360824742268,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004573865527934181,
      "loss": 5.0427,
      "mean_token_accuracy": 0.19662826657295226,
      "num_tokens": 45788002.0,
      "step": 26395
    },
    {
      "entropy": 5.763675212860107,
      "epoch": 2.053997276794398,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004573703214197692,
      "loss": 5.0319,
      "mean_token_accuracy": 0.1990533947944641,
      "num_tokens": 45796563.0,
      "step": 26400
    },
    {
      "entropy": 5.687012004852295,
      "epoch": 2.0543863061661156,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045735408727894095,
      "loss": 4.9572,
      "mean_token_accuracy": 0.20754703283309936,
      "num_tokens": 45806311.0,
      "step": 26405
    },
    {
      "entropy": 5.682385015487671,
      "epoch": 2.0547753355378333,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045733785037117977,
      "loss": 4.9483,
      "mean_token_accuracy": 0.20658033937215806,
      "num_tokens": 45814007.0,
      "step": 26410
    },
    {
      "entropy": 5.670385646820068,
      "epoch": 2.0551643649095506,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004573216106967319,
      "loss": 4.9801,
      "mean_token_accuracy": 0.19763838350772858,
      "num_tokens": 45822492.0,
      "step": 26415
    },
    {
      "entropy": 5.641687440872192,
      "epoch": 2.0555533942812683,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045730536825584365,
      "loss": 4.918,
      "mean_token_accuracy": 0.20933483242988588,
      "num_tokens": 45830853.0,
      "step": 26420
    },
    {
      "entropy": 5.739394187927246,
      "epoch": 2.055942423652986,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045728912304876176,
      "loss": 5.0138,
      "mean_token_accuracy": 0.20124027729034424,
      "num_tokens": 45839948.0,
      "step": 26425
    },
    {
      "entropy": 5.793256711959839,
      "epoch": 2.056331453024703,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004572728750757326,
      "loss": 5.1247,
      "mean_token_accuracy": 0.1831857144832611,
      "num_tokens": 45848264.0,
      "step": 26430
    },
    {
      "entropy": 5.611777877807617,
      "epoch": 2.056720482396421,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004572566243370025,
      "loss": 4.892,
      "mean_token_accuracy": 0.21454827040433883,
      "num_tokens": 45856523.0,
      "step": 26435
    },
    {
      "entropy": 5.642489337921143,
      "epoch": 2.0571095117681386,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004572403708328183,
      "loss": 5.0056,
      "mean_token_accuracy": 0.20898969769477843,
      "num_tokens": 45864736.0,
      "step": 26440
    },
    {
      "entropy": 5.671368217468261,
      "epoch": 2.057498541139856,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004572241145634266,
      "loss": 5.0037,
      "mean_token_accuracy": 0.2016543686389923,
      "num_tokens": 45873979.0,
      "step": 26445
    },
    {
      "entropy": 5.643639373779297,
      "epoch": 2.0578875705115736,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000457207855529074,
      "loss": 4.959,
      "mean_token_accuracy": 0.20861584991216658,
      "num_tokens": 45882545.0,
      "step": 26450
    },
    {
      "entropy": 5.669561815261841,
      "epoch": 2.0582765998832913,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045719159373000713,
      "loss": 5.0473,
      "mean_token_accuracy": 0.19620343595743178,
      "num_tokens": 45890830.0,
      "step": 26455
    },
    {
      "entropy": 5.773867225646972,
      "epoch": 2.058665629255009,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004571753291664729,
      "loss": 4.9935,
      "mean_token_accuracy": 0.20281164348125458,
      "num_tokens": 45899457.0,
      "step": 26460
    },
    {
      "entropy": 5.685356998443604,
      "epoch": 2.059054658626726,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004571590618387179,
      "loss": 4.9994,
      "mean_token_accuracy": 0.20380024760961532,
      "num_tokens": 45907998.0,
      "step": 26465
    },
    {
      "entropy": 5.645242547988891,
      "epoch": 2.059443687998444,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004571427917469892,
      "loss": 4.928,
      "mean_token_accuracy": 0.2026064619421959,
      "num_tokens": 45916102.0,
      "step": 26470
    },
    {
      "entropy": 5.657299900054932,
      "epoch": 2.0598327173701616,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045712651889153345,
      "loss": 4.9613,
      "mean_token_accuracy": 0.20668953508138657,
      "num_tokens": 45924347.0,
      "step": 26475
    },
    {
      "entropy": 5.7314183712005615,
      "epoch": 2.060221746741879,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045711024327259764,
      "loss": 4.9837,
      "mean_token_accuracy": 0.2074410766363144,
      "num_tokens": 45932702.0,
      "step": 26480
    },
    {
      "entropy": 5.61371865272522,
      "epoch": 2.0606107761135966,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045709396489042884,
      "loss": 4.9459,
      "mean_token_accuracy": 0.20138026028871536,
      "num_tokens": 45941235.0,
      "step": 26485
    },
    {
      "entropy": 5.681282043457031,
      "epoch": 2.0609998054853143,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045707768374527385,
      "loss": 4.94,
      "mean_token_accuracy": 0.21190550923347473,
      "num_tokens": 45949582.0,
      "step": 26490
    },
    {
      "entropy": 5.695214796066284,
      "epoch": 2.0613888348570315,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004570613998373799,
      "loss": 5.0095,
      "mean_token_accuracy": 0.20235701650381088,
      "num_tokens": 45957920.0,
      "step": 26495
    },
    {
      "entropy": 5.6700811862945555,
      "epoch": 2.061777864228749,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045704511316699395,
      "loss": 5.0506,
      "mean_token_accuracy": 0.20360081791877746,
      "num_tokens": 45966302.0,
      "step": 26500
    },
    {
      "entropy": 5.733704376220703,
      "epoch": 2.062166893600467,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045702882373436317,
      "loss": 5.0719,
      "mean_token_accuracy": 0.19766143411397935,
      "num_tokens": 45975481.0,
      "step": 26505
    },
    {
      "entropy": 5.816517686843872,
      "epoch": 2.0625559229721846,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004570125315397347,
      "loss": 5.062,
      "mean_token_accuracy": 0.20263160914182662,
      "num_tokens": 45984466.0,
      "step": 26510
    },
    {
      "entropy": 5.662425756454468,
      "epoch": 2.062944952343902,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004569962365833558,
      "loss": 4.8943,
      "mean_token_accuracy": 0.20768968611955643,
      "num_tokens": 45992862.0,
      "step": 26515
    },
    {
      "entropy": 5.63205771446228,
      "epoch": 2.0633339817156195,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045697993886547374,
      "loss": 4.9946,
      "mean_token_accuracy": 0.19968704879283905,
      "num_tokens": 46001211.0,
      "step": 26520
    },
    {
      "entropy": 5.721051979064941,
      "epoch": 2.0637230110873372,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045696363838633566,
      "loss": 5.0073,
      "mean_token_accuracy": 0.19828609973192216,
      "num_tokens": 46009574.0,
      "step": 26525
    },
    {
      "entropy": 5.71382737159729,
      "epoch": 2.0641120404590545,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045694733514618904,
      "loss": 4.9422,
      "mean_token_accuracy": 0.20257085859775542,
      "num_tokens": 46017686.0,
      "step": 26530
    },
    {
      "entropy": 5.688066673278809,
      "epoch": 2.064501069830772,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004569310291452812,
      "loss": 4.9663,
      "mean_token_accuracy": 0.2029442995786667,
      "num_tokens": 46026842.0,
      "step": 26535
    },
    {
      "entropy": 5.644578695297241,
      "epoch": 2.06489009920249,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004569147203838595,
      "loss": 4.9604,
      "mean_token_accuracy": 0.20868255496025084,
      "num_tokens": 46036308.0,
      "step": 26540
    },
    {
      "entropy": 5.702411651611328,
      "epoch": 2.065279128574207,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045689840886217157,
      "loss": 4.9733,
      "mean_token_accuracy": 0.20491039454936982,
      "num_tokens": 46045424.0,
      "step": 26545
    },
    {
      "entropy": 5.667631721496582,
      "epoch": 2.065668157945925,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045688209458046475,
      "loss": 4.9508,
      "mean_token_accuracy": 0.20506266504526138,
      "num_tokens": 46054386.0,
      "step": 26550
    },
    {
      "entropy": 5.633437442779541,
      "epoch": 2.0660571873176425,
      "grad_norm": 1.125,
      "learning_rate": 0.00045686577753898663,
      "loss": 4.8846,
      "mean_token_accuracy": 0.214173524081707,
      "num_tokens": 46063617.0,
      "step": 26555
    },
    {
      "entropy": 5.702445220947266,
      "epoch": 2.0664462166893602,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045684945773798483,
      "loss": 5.0766,
      "mean_token_accuracy": 0.20049827843904494,
      "num_tokens": 46072915.0,
      "step": 26560
    },
    {
      "entropy": 5.725633859634399,
      "epoch": 2.0668352460610775,
      "grad_norm": 1.21875,
      "learning_rate": 0.000456833135177707,
      "loss": 5.101,
      "mean_token_accuracy": 0.19491964280605317,
      "num_tokens": 46081890.0,
      "step": 26565
    },
    {
      "entropy": 5.770382881164551,
      "epoch": 2.067224275432795,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004568168098584007,
      "loss": 5.0652,
      "mean_token_accuracy": 0.19314909130334854,
      "num_tokens": 46090445.0,
      "step": 26570
    },
    {
      "entropy": 5.76362042427063,
      "epoch": 2.067613304804513,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004568004817803137,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20208129584789275,
      "num_tokens": 46099866.0,
      "step": 26575
    },
    {
      "entropy": 5.674625205993652,
      "epoch": 2.06800233417623,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004567841509436937,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20954311192035674,
      "num_tokens": 46108036.0,
      "step": 26580
    },
    {
      "entropy": 5.678498792648315,
      "epoch": 2.068391363547948,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004567678173487887,
      "loss": 4.8992,
      "mean_token_accuracy": 0.21291859298944474,
      "num_tokens": 46116087.0,
      "step": 26585
    },
    {
      "entropy": 5.643871879577636,
      "epoch": 2.0687803929196655,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004567514809958462,
      "loss": 4.9825,
      "mean_token_accuracy": 0.21221491247415541,
      "num_tokens": 46124934.0,
      "step": 26590
    },
    {
      "entropy": 5.700538682937622,
      "epoch": 2.069169422291383,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004567351418851144,
      "loss": 4.9921,
      "mean_token_accuracy": 0.2074885591864586,
      "num_tokens": 46133691.0,
      "step": 26595
    },
    {
      "entropy": 5.729164123535156,
      "epoch": 2.0695584516631005,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045671880001684113,
      "loss": 5.0537,
      "mean_token_accuracy": 0.19493894577026366,
      "num_tokens": 46141933.0,
      "step": 26600
    },
    {
      "entropy": 5.672248888015747,
      "epoch": 2.069947481034818,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045670245539127413,
      "loss": 4.9336,
      "mean_token_accuracy": 0.20921938270330429,
      "num_tokens": 46150656.0,
      "step": 26605
    },
    {
      "entropy": 5.620133686065674,
      "epoch": 2.070336510406536,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045668610800866173,
      "loss": 4.9543,
      "mean_token_accuracy": 0.2051005929708481,
      "num_tokens": 46160353.0,
      "step": 26610
    },
    {
      "entropy": 5.684297752380371,
      "epoch": 2.070725539778253,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045666975786925177,
      "loss": 4.9904,
      "mean_token_accuracy": 0.2084883764386177,
      "num_tokens": 46168810.0,
      "step": 26615
    },
    {
      "entropy": 5.699772834777832,
      "epoch": 2.071114569149971,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004566534049732923,
      "loss": 4.9695,
      "mean_token_accuracy": 0.20911960750818254,
      "num_tokens": 46177593.0,
      "step": 26620
    },
    {
      "entropy": 5.688872289657593,
      "epoch": 2.0715035985216885,
      "grad_norm": 1.25,
      "learning_rate": 0.00045663704932103166,
      "loss": 4.9698,
      "mean_token_accuracy": 0.21247036904096603,
      "num_tokens": 46186027.0,
      "step": 26625
    },
    {
      "entropy": 5.7123737812042235,
      "epoch": 2.071892627893406,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045662069091271785,
      "loss": 5.0808,
      "mean_token_accuracy": 0.19122884124517442,
      "num_tokens": 46194705.0,
      "step": 26630
    },
    {
      "entropy": 5.698910570144653,
      "epoch": 2.0722816572651235,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045660432974859924,
      "loss": 4.9227,
      "mean_token_accuracy": 0.20927760750055313,
      "num_tokens": 46203498.0,
      "step": 26635
    },
    {
      "entropy": 5.721298789978027,
      "epoch": 2.072670686636841,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045658796582892383,
      "loss": 4.9745,
      "mean_token_accuracy": 0.21304211765527725,
      "num_tokens": 46211918.0,
      "step": 26640
    },
    {
      "entropy": 5.742507600784302,
      "epoch": 2.073059716008559,
      "grad_norm": 1.125,
      "learning_rate": 0.00045657159915394013,
      "loss": 5.0695,
      "mean_token_accuracy": 0.20158637464046478,
      "num_tokens": 46220689.0,
      "step": 26645
    },
    {
      "entropy": 5.711783456802368,
      "epoch": 2.073448745380276,
      "grad_norm": 1.125,
      "learning_rate": 0.0004565552297238964,
      "loss": 5.0406,
      "mean_token_accuracy": 0.1979992210865021,
      "num_tokens": 46229645.0,
      "step": 26650
    },
    {
      "entropy": 5.754027795791626,
      "epoch": 2.073837774751994,
      "grad_norm": 1.3125,
      "learning_rate": 0.000456538857539041,
      "loss": 5.0118,
      "mean_token_accuracy": 0.1969640538096428,
      "num_tokens": 46237691.0,
      "step": 26655
    },
    {
      "entropy": 5.702052974700928,
      "epoch": 2.0742268041237115,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045652248259962254,
      "loss": 4.9701,
      "mean_token_accuracy": 0.20612976849079132,
      "num_tokens": 46245756.0,
      "step": 26660
    },
    {
      "entropy": 5.753409910202026,
      "epoch": 2.0746158334954288,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004565061049058892,
      "loss": 5.0689,
      "mean_token_accuracy": 0.19615186154842376,
      "num_tokens": 46255233.0,
      "step": 26665
    },
    {
      "entropy": 5.705156421661377,
      "epoch": 2.0750048628671465,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045648972445808963,
      "loss": 5.0111,
      "mean_token_accuracy": 0.19983147978782653,
      "num_tokens": 46263569.0,
      "step": 26670
    },
    {
      "entropy": 5.616665649414062,
      "epoch": 2.075393892238864,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045647334125647235,
      "loss": 4.8987,
      "mean_token_accuracy": 0.20818208903074265,
      "num_tokens": 46273010.0,
      "step": 26675
    },
    {
      "entropy": 5.754445362091064,
      "epoch": 2.0757829216105814,
      "grad_norm": 1.109375,
      "learning_rate": 0.000456456955301286,
      "loss": 4.9747,
      "mean_token_accuracy": 0.21063933074474334,
      "num_tokens": 46281566.0,
      "step": 26680
    },
    {
      "entropy": 5.602251291275024,
      "epoch": 2.076171950982299,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004564405665927791,
      "loss": 4.8693,
      "mean_token_accuracy": 0.21099644154310226,
      "num_tokens": 46290267.0,
      "step": 26685
    },
    {
      "entropy": 5.634386110305786,
      "epoch": 2.076560980354017,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045642417513120043,
      "loss": 4.8995,
      "mean_token_accuracy": 0.21202029436826705,
      "num_tokens": 46298694.0,
      "step": 26690
    },
    {
      "entropy": 5.680882310867309,
      "epoch": 2.076950009725734,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045640778091679876,
      "loss": 4.8625,
      "mean_token_accuracy": 0.21378698199987411,
      "num_tokens": 46307358.0,
      "step": 26695
    },
    {
      "entropy": 5.6812952041625975,
      "epoch": 2.0773390390974518,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004563913839498226,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2135591298341751,
      "num_tokens": 46316351.0,
      "step": 26700
    },
    {
      "entropy": 5.6801787376403805,
      "epoch": 2.0777280684691695,
      "grad_norm": 1.171875,
      "learning_rate": 0.000456374984230521,
      "loss": 4.9746,
      "mean_token_accuracy": 0.20548408180475236,
      "num_tokens": 46325591.0,
      "step": 26705
    },
    {
      "entropy": 5.690999555587768,
      "epoch": 2.078117097840887,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004563585817591427,
      "loss": 4.9555,
      "mean_token_accuracy": 0.20443084537982942,
      "num_tokens": 46334595.0,
      "step": 26710
    },
    {
      "entropy": 5.771510267257691,
      "epoch": 2.0785061272126044,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004563421765359365,
      "loss": 5.0369,
      "mean_token_accuracy": 0.19937919825315475,
      "num_tokens": 46343587.0,
      "step": 26715
    },
    {
      "entropy": 5.8022724151611325,
      "epoch": 2.078895156584322,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045632576856115156,
      "loss": 5.0261,
      "mean_token_accuracy": 0.20182194858789443,
      "num_tokens": 46351900.0,
      "step": 26720
    },
    {
      "entropy": 5.722235107421875,
      "epoch": 2.07928418595604,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045630935783503657,
      "loss": 5.0231,
      "mean_token_accuracy": 0.2011237323284149,
      "num_tokens": 46359854.0,
      "step": 26725
    },
    {
      "entropy": 5.712496328353882,
      "epoch": 2.079673215327757,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004562929443578407,
      "loss": 5.1116,
      "mean_token_accuracy": 0.19846736788749694,
      "num_tokens": 46368930.0,
      "step": 26730
    },
    {
      "entropy": 5.6774827003479,
      "epoch": 2.0800622446994748,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004562765281298129,
      "loss": 4.8986,
      "mean_token_accuracy": 0.20788912922143937,
      "num_tokens": 46377522.0,
      "step": 26735
    },
    {
      "entropy": 5.669716310501099,
      "epoch": 2.0804512740711925,
      "grad_norm": 1.25,
      "learning_rate": 0.0004562601091512024,
      "loss": 4.9971,
      "mean_token_accuracy": 0.19960644990205764,
      "num_tokens": 46386368.0,
      "step": 26740
    },
    {
      "entropy": 5.695429706573487,
      "epoch": 2.08084030344291,
      "grad_norm": 1.25,
      "learning_rate": 0.0004562436874222582,
      "loss": 4.9553,
      "mean_token_accuracy": 0.2035237267613411,
      "num_tokens": 46395363.0,
      "step": 26745
    },
    {
      "entropy": 5.748808670043945,
      "epoch": 2.0812293328146274,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045622726294322955,
      "loss": 5.0225,
      "mean_token_accuracy": 0.19950494766235352,
      "num_tokens": 46403998.0,
      "step": 26750
    },
    {
      "entropy": 5.576841592788696,
      "epoch": 2.081618362186345,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045621083571436563,
      "loss": 4.8201,
      "mean_token_accuracy": 0.222538498044014,
      "num_tokens": 46412138.0,
      "step": 26755
    },
    {
      "entropy": 5.631717395782471,
      "epoch": 2.082007391558063,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004561944057359157,
      "loss": 4.9819,
      "mean_token_accuracy": 0.20640279799699784,
      "num_tokens": 46420761.0,
      "step": 26760
    },
    {
      "entropy": 5.601476192474365,
      "epoch": 2.08239642092978,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004561779730081291,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21046752184629441,
      "num_tokens": 46428938.0,
      "step": 26765
    },
    {
      "entropy": 5.7185474872589115,
      "epoch": 2.0827854503014978,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004561615375312551,
      "loss": 4.9554,
      "mean_token_accuracy": 0.21121462434530258,
      "num_tokens": 46436552.0,
      "step": 26770
    },
    {
      "entropy": 5.633351278305054,
      "epoch": 2.0831744796732155,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045614509930554304,
      "loss": 4.9118,
      "mean_token_accuracy": 0.2094930127263069,
      "num_tokens": 46445223.0,
      "step": 26775
    },
    {
      "entropy": 5.680253314971924,
      "epoch": 2.0835635090449327,
      "grad_norm": 1.25,
      "learning_rate": 0.00045612865833124253,
      "loss": 5.0323,
      "mean_token_accuracy": 0.20136726796627044,
      "num_tokens": 46453232.0,
      "step": 26780
    },
    {
      "entropy": 5.559950256347657,
      "epoch": 2.0839525384166504,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004561122146086029,
      "loss": 4.8388,
      "mean_token_accuracy": 0.21345621794462205,
      "num_tokens": 46462562.0,
      "step": 26785
    },
    {
      "entropy": 5.675860929489136,
      "epoch": 2.084341567788368,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004560957681378737,
      "loss": 5.0317,
      "mean_token_accuracy": 0.1949341729283333,
      "num_tokens": 46470599.0,
      "step": 26790
    },
    {
      "entropy": 5.745585489273071,
      "epoch": 2.084730597160086,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045607931891930445,
      "loss": 4.9897,
      "mean_token_accuracy": 0.2033182993531227,
      "num_tokens": 46478707.0,
      "step": 26795
    },
    {
      "entropy": 5.657948923110962,
      "epoch": 2.085119626531803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004560628669531447,
      "loss": 4.9773,
      "mean_token_accuracy": 0.2046966314315796,
      "num_tokens": 46487044.0,
      "step": 26800
    },
    {
      "entropy": 5.686334657669067,
      "epoch": 2.0855086559035207,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045604641223964416,
      "loss": 4.985,
      "mean_token_accuracy": 0.19873972088098527,
      "num_tokens": 46495278.0,
      "step": 26805
    },
    {
      "entropy": 5.644309186935425,
      "epoch": 2.0858976852752384,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045602995477905247,
      "loss": 4.9989,
      "mean_token_accuracy": 0.2050021097064018,
      "num_tokens": 46503545.0,
      "step": 26810
    },
    {
      "entropy": 5.654935693740844,
      "epoch": 2.0862867146469557,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004560134945716194,
      "loss": 5.0312,
      "mean_token_accuracy": 0.21401808112859727,
      "num_tokens": 46513058.0,
      "step": 26815
    },
    {
      "entropy": 5.708903217315674,
      "epoch": 2.0866757440186734,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045599703161759464,
      "loss": 4.9381,
      "mean_token_accuracy": 0.20220120400190353,
      "num_tokens": 46520651.0,
      "step": 26820
    },
    {
      "entropy": 5.651180028915405,
      "epoch": 2.087064773390391,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045598056591722805,
      "loss": 4.92,
      "mean_token_accuracy": 0.20638586580753326,
      "num_tokens": 46528899.0,
      "step": 26825
    },
    {
      "entropy": 5.731946802139282,
      "epoch": 2.0874538027621083,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045596409747076936,
      "loss": 5.1213,
      "mean_token_accuracy": 0.18965098559856414,
      "num_tokens": 46537921.0,
      "step": 26830
    },
    {
      "entropy": 5.713159465789795,
      "epoch": 2.087842832133826,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004559476262784686,
      "loss": 4.9551,
      "mean_token_accuracy": 0.20966471135616302,
      "num_tokens": 46547432.0,
      "step": 26835
    },
    {
      "entropy": 5.694389772415161,
      "epoch": 2.0882318615055437,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004559311523405755,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20187440067529677,
      "num_tokens": 46555406.0,
      "step": 26840
    },
    {
      "entropy": 5.727861070632935,
      "epoch": 2.0886208908772614,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004559146756573402,
      "loss": 5.0013,
      "mean_token_accuracy": 0.2088017523288727,
      "num_tokens": 46564428.0,
      "step": 26845
    },
    {
      "entropy": 5.704367351531983,
      "epoch": 2.0890099202489787,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045589819622901274,
      "loss": 4.9285,
      "mean_token_accuracy": 0.2126939758658409,
      "num_tokens": 46573222.0,
      "step": 26850
    },
    {
      "entropy": 5.666756296157837,
      "epoch": 2.0893989496206964,
      "grad_norm": 1.234375,
      "learning_rate": 0.000455881714055843,
      "loss": 4.9345,
      "mean_token_accuracy": 0.2110202580690384,
      "num_tokens": 46581951.0,
      "step": 26855
    },
    {
      "entropy": 5.728613185882568,
      "epoch": 2.089787978992414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045586522913808114,
      "loss": 5.0557,
      "mean_token_accuracy": 0.21197190880775452,
      "num_tokens": 46590658.0,
      "step": 26860
    },
    {
      "entropy": 5.694840002059936,
      "epoch": 2.0901770083641313,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004558487414759773,
      "loss": 4.9027,
      "mean_token_accuracy": 0.21503257155418395,
      "num_tokens": 46599068.0,
      "step": 26865
    },
    {
      "entropy": 5.655412244796753,
      "epoch": 2.090566037735849,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045583225106978175,
      "loss": 4.9914,
      "mean_token_accuracy": 0.20396257936954498,
      "num_tokens": 46608024.0,
      "step": 26870
    },
    {
      "entropy": 5.643037366867065,
      "epoch": 2.0909550671075667,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004558157579197446,
      "loss": 4.9302,
      "mean_token_accuracy": 0.20895154327154158,
      "num_tokens": 46616558.0,
      "step": 26875
    },
    {
      "entropy": 5.7506249904632565,
      "epoch": 2.091344096479284,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045579926202611603,
      "loss": 5.1379,
      "mean_token_accuracy": 0.19193762093782424,
      "num_tokens": 46624943.0,
      "step": 26880
    },
    {
      "entropy": 5.673505687713623,
      "epoch": 2.0917331258510017,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004557827633891465,
      "loss": 4.9833,
      "mean_token_accuracy": 0.19999424517154693,
      "num_tokens": 46633753.0,
      "step": 26885
    },
    {
      "entropy": 5.711205911636353,
      "epoch": 2.0921221552227194,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004557662620090863,
      "loss": 4.9935,
      "mean_token_accuracy": 0.20564150661230088,
      "num_tokens": 46642210.0,
      "step": 26890
    },
    {
      "entropy": 5.691160678863525,
      "epoch": 2.092511184594437,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045574975788618564,
      "loss": 4.9513,
      "mean_token_accuracy": 0.20996245741844177,
      "num_tokens": 46650327.0,
      "step": 26895
    },
    {
      "entropy": 5.6484599113464355,
      "epoch": 2.0929002139661543,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004557332510206953,
      "loss": 4.8833,
      "mean_token_accuracy": 0.21219709515571594,
      "num_tokens": 46659453.0,
      "step": 26900
    },
    {
      "entropy": 5.736992979049683,
      "epoch": 2.093289243337872,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004557167414128654,
      "loss": 4.9679,
      "mean_token_accuracy": 0.20799987614154816,
      "num_tokens": 46667378.0,
      "step": 26905
    },
    {
      "entropy": 5.725967359542847,
      "epoch": 2.0936782727095897,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004557002290629467,
      "loss": 5.1324,
      "mean_token_accuracy": 0.1900894433259964,
      "num_tokens": 46675949.0,
      "step": 26910
    },
    {
      "entropy": 5.770013904571533,
      "epoch": 2.094067302081307,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045568371397118954,
      "loss": 5.1541,
      "mean_token_accuracy": 0.19959221333265303,
      "num_tokens": 46684808.0,
      "step": 26915
    },
    {
      "entropy": 5.716790723800659,
      "epoch": 2.0944563314530247,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004556671961378446,
      "loss": 4.9666,
      "mean_token_accuracy": 0.2076625183224678,
      "num_tokens": 46693209.0,
      "step": 26920
    },
    {
      "entropy": 5.633689975738525,
      "epoch": 2.0948453608247424,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045565067556316264,
      "loss": 4.943,
      "mean_token_accuracy": 0.2135712593793869,
      "num_tokens": 46702405.0,
      "step": 26925
    },
    {
      "entropy": 5.5798985958099365,
      "epoch": 2.0952343901964596,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004556341522473941,
      "loss": 4.9592,
      "mean_token_accuracy": 0.2059960275888443,
      "num_tokens": 46711568.0,
      "step": 26930
    },
    {
      "entropy": 5.655941534042358,
      "epoch": 2.0956234195681773,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004556176261907899,
      "loss": 4.98,
      "mean_token_accuracy": 0.20742260813713073,
      "num_tokens": 46719886.0,
      "step": 26935
    },
    {
      "entropy": 5.6564287662506105,
      "epoch": 2.096012448939895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004556010973936006,
      "loss": 4.8356,
      "mean_token_accuracy": 0.21430304795503616,
      "num_tokens": 46728293.0,
      "step": 26940
    },
    {
      "entropy": 5.683703279495239,
      "epoch": 2.0964014783116127,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004555845658560772,
      "loss": 4.9879,
      "mean_token_accuracy": 0.2014298677444458,
      "num_tokens": 46736730.0,
      "step": 26945
    },
    {
      "entropy": 5.687503290176392,
      "epoch": 2.09679050768333,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045556803157847036,
      "loss": 5.0408,
      "mean_token_accuracy": 0.1930103436112404,
      "num_tokens": 46745563.0,
      "step": 26950
    },
    {
      "entropy": 5.722416162490845,
      "epoch": 2.0971795370550477,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045555149456103113,
      "loss": 4.988,
      "mean_token_accuracy": 0.20209750384092331,
      "num_tokens": 46754285.0,
      "step": 26955
    },
    {
      "entropy": 5.706587600708008,
      "epoch": 2.0975685664267654,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004555349548040102,
      "loss": 4.9867,
      "mean_token_accuracy": 0.21139516979455947,
      "num_tokens": 46763114.0,
      "step": 26960
    },
    {
      "entropy": 5.704325723648071,
      "epoch": 2.0979575957984826,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004555184123076589,
      "loss": 4.9986,
      "mean_token_accuracy": 0.2040882334113121,
      "num_tokens": 46771150.0,
      "step": 26965
    },
    {
      "entropy": 5.6477532386779785,
      "epoch": 2.0983466251702003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045550186707222785,
      "loss": 4.8857,
      "mean_token_accuracy": 0.21533036679029466,
      "num_tokens": 46779955.0,
      "step": 26970
    },
    {
      "entropy": 5.678753089904785,
      "epoch": 2.098735654541918,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004554853190979683,
      "loss": 5.0204,
      "mean_token_accuracy": 0.19841573685407637,
      "num_tokens": 46788769.0,
      "step": 26975
    },
    {
      "entropy": 5.72996416091919,
      "epoch": 2.0991246839136353,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00045546876838513134,
      "loss": 5.0257,
      "mean_token_accuracy": 0.20052011162042618,
      "num_tokens": 46797649.0,
      "step": 26980
    },
    {
      "entropy": 5.693415212631225,
      "epoch": 2.099513713285353,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045545221493396805,
      "loss": 4.9371,
      "mean_token_accuracy": 0.20562161654233932,
      "num_tokens": 46806140.0,
      "step": 26985
    },
    {
      "entropy": 5.697494268417358,
      "epoch": 2.0999027426570707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045543565874472963,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2161271169781685,
      "num_tokens": 46815476.0,
      "step": 26990
    },
    {
      "entropy": 5.6711503028869625,
      "epoch": 2.1002917720287884,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045541909981766725,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20086161494255067,
      "num_tokens": 46824171.0,
      "step": 26995
    },
    {
      "entropy": 5.609083890914917,
      "epoch": 2.1006808014005056,
      "grad_norm": 1.25,
      "learning_rate": 0.0004554025381530322,
      "loss": 4.8806,
      "mean_token_accuracy": 0.2124312177300453,
      "num_tokens": 46832474.0,
      "step": 27000
    },
    {
      "epoch": 2.1006808014005056,
      "eval_entropy": 5.377945898861927,
      "eval_loss": 5.092503070831299,
      "eval_mean_token_accuracy": 0.209100198200738,
      "eval_num_tokens": 46832474.0,
      "eval_runtime": 21.5964,
      "eval_samples_per_second": 1924.304,
      "eval_steps_per_second": 240.55,
      "step": 27000
    },
    {
      "entropy": 5.627761554718018,
      "epoch": 2.1010698307722233,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004553859737510758,
      "loss": 4.8619,
      "mean_token_accuracy": 0.2110622063279152,
      "num_tokens": 46840626.0,
      "step": 27005
    },
    {
      "entropy": 5.606659460067749,
      "epoch": 2.101458860143941,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004553694066120493,
      "loss": 4.8525,
      "mean_token_accuracy": 0.21379860639572143,
      "num_tokens": 46849626.0,
      "step": 27010
    },
    {
      "entropy": 5.741579961776734,
      "epoch": 2.1018478895156583,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045535283673620426,
      "loss": 5.0076,
      "mean_token_accuracy": 0.20166471153497695,
      "num_tokens": 46858520.0,
      "step": 27015
    },
    {
      "entropy": 5.684970951080322,
      "epoch": 2.102236918887376,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045533626412379194,
      "loss": 4.9927,
      "mean_token_accuracy": 0.202072574198246,
      "num_tokens": 46866532.0,
      "step": 27020
    },
    {
      "entropy": 5.711877155303955,
      "epoch": 2.1026259482590937,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004553196887750638,
      "loss": 5.0273,
      "mean_token_accuracy": 0.2013051524758339,
      "num_tokens": 46875815.0,
      "step": 27025
    },
    {
      "entropy": 5.689473009109497,
      "epoch": 2.103014977630811,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004553031106902714,
      "loss": 4.9548,
      "mean_token_accuracy": 0.20021456778049468,
      "num_tokens": 46884708.0,
      "step": 27030
    },
    {
      "entropy": 5.697423124313355,
      "epoch": 2.1034040070025286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045528652986966627,
      "loss": 4.8838,
      "mean_token_accuracy": 0.2125650018453598,
      "num_tokens": 46894089.0,
      "step": 27035
    },
    {
      "entropy": 5.7602277278900145,
      "epoch": 2.1037930363742463,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045526994631350006,
      "loss": 5.0568,
      "mean_token_accuracy": 0.20414750128984452,
      "num_tokens": 46903318.0,
      "step": 27040
    },
    {
      "entropy": 5.648242330551147,
      "epoch": 2.104182065745964,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004552533600220243,
      "loss": 4.8922,
      "mean_token_accuracy": 0.21144102960824968,
      "num_tokens": 46912138.0,
      "step": 27045
    },
    {
      "entropy": 5.717019987106323,
      "epoch": 2.1045710951176813,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00045523677099549073,
      "loss": 4.9611,
      "mean_token_accuracy": 0.19795125871896743,
      "num_tokens": 46919544.0,
      "step": 27050
    },
    {
      "entropy": 5.699784469604492,
      "epoch": 2.104960124489399,
      "grad_norm": 1.21875,
      "learning_rate": 0.000455220179234151,
      "loss": 5.0549,
      "mean_token_accuracy": 0.2009477749466896,
      "num_tokens": 46928760.0,
      "step": 27055
    },
    {
      "entropy": 5.677061080932617,
      "epoch": 2.1053491538611167,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004552035847382569,
      "loss": 5.0,
      "mean_token_accuracy": 0.20254934579133987,
      "num_tokens": 46936860.0,
      "step": 27060
    },
    {
      "entropy": 5.673934412002564,
      "epoch": 2.105738183232834,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045518698750806024,
      "loss": 4.9893,
      "mean_token_accuracy": 0.19814175367355347,
      "num_tokens": 46945157.0,
      "step": 27065
    },
    {
      "entropy": 5.76425461769104,
      "epoch": 2.1061272126045516,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045517038754381287,
      "loss": 4.9977,
      "mean_token_accuracy": 0.19859624207019805,
      "num_tokens": 46954006.0,
      "step": 27070
    },
    {
      "entropy": 5.653291893005371,
      "epoch": 2.1065162419762693,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004551537848457666,
      "loss": 4.9437,
      "mean_token_accuracy": 0.2098340079188347,
      "num_tokens": 46962437.0,
      "step": 27075
    },
    {
      "entropy": 5.690201282501221,
      "epoch": 2.106905271347987,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004551371794141734,
      "loss": 5.0561,
      "mean_token_accuracy": 0.2047122150659561,
      "num_tokens": 46972379.0,
      "step": 27080
    },
    {
      "entropy": 5.740162515640259,
      "epoch": 2.1072943007197042,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004551205712492852,
      "loss": 5.0277,
      "mean_token_accuracy": 0.19806330054998397,
      "num_tokens": 46981410.0,
      "step": 27085
    },
    {
      "entropy": 5.691622066497803,
      "epoch": 2.107683330091422,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004551039603513541,
      "loss": 4.8922,
      "mean_token_accuracy": 0.20493658185005187,
      "num_tokens": 46989662.0,
      "step": 27090
    },
    {
      "entropy": 5.731349229812622,
      "epoch": 2.1080723594631396,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045508734672063204,
      "loss": 5.0324,
      "mean_token_accuracy": 0.20330643504858018,
      "num_tokens": 46998726.0,
      "step": 27095
    },
    {
      "entropy": 5.74033932685852,
      "epoch": 2.108461388834857,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004550707303573711,
      "loss": 5.0494,
      "mean_token_accuracy": 0.20305478423833848,
      "num_tokens": 47006980.0,
      "step": 27100
    },
    {
      "entropy": 5.663844919204712,
      "epoch": 2.1088504182065746,
      "grad_norm": 1.25,
      "learning_rate": 0.00045505411126182347,
      "loss": 5.0353,
      "mean_token_accuracy": 0.19266543984413148,
      "num_tokens": 47015693.0,
      "step": 27105
    },
    {
      "entropy": 5.655839347839356,
      "epoch": 2.1092394475782923,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004550374894342412,
      "loss": 4.9483,
      "mean_token_accuracy": 0.2141800418496132,
      "num_tokens": 47024256.0,
      "step": 27110
    },
    {
      "entropy": 5.73844633102417,
      "epoch": 2.1096284769500095,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004550208648748767,
      "loss": 5.0481,
      "mean_token_accuracy": 0.1997669205069542,
      "num_tokens": 47032962.0,
      "step": 27115
    },
    {
      "entropy": 5.712097549438477,
      "epoch": 2.1100175063217272,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000455004237583982,
      "loss": 5.0107,
      "mean_token_accuracy": 0.19942270666360856,
      "num_tokens": 47041507.0,
      "step": 27120
    },
    {
      "entropy": 5.620849466323852,
      "epoch": 2.110406535693445,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004549876075618096,
      "loss": 4.9283,
      "mean_token_accuracy": 0.20913327485322952,
      "num_tokens": 47050088.0,
      "step": 27125
    },
    {
      "entropy": 5.760932493209839,
      "epoch": 2.110795565065162,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004549709748086117,
      "loss": 5.0662,
      "mean_token_accuracy": 0.20364027917385102,
      "num_tokens": 47058900.0,
      "step": 27130
    },
    {
      "entropy": 5.735073375701904,
      "epoch": 2.11118459443688,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045495433932464063,
      "loss": 4.9608,
      "mean_token_accuracy": 0.2026571273803711,
      "num_tokens": 47067689.0,
      "step": 27135
    },
    {
      "entropy": 5.677677965164184,
      "epoch": 2.1115736238085976,
      "grad_norm": 1.09375,
      "learning_rate": 0.000454937701110149,
      "loss": 4.9365,
      "mean_token_accuracy": 0.20697788000106812,
      "num_tokens": 47076309.0,
      "step": 27140
    },
    {
      "entropy": 5.675854015350342,
      "epoch": 2.1119626531803153,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045492106016538917,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21431101262569427,
      "num_tokens": 47084462.0,
      "step": 27145
    },
    {
      "entropy": 5.684068822860718,
      "epoch": 2.1123516825520325,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045490441649061354,
      "loss": 5.0495,
      "mean_token_accuracy": 0.2020340234041214,
      "num_tokens": 47093586.0,
      "step": 27150
    },
    {
      "entropy": 5.674016904830933,
      "epoch": 2.1127407119237502,
      "grad_norm": 1.25,
      "learning_rate": 0.0004548877700860747,
      "loss": 4.9372,
      "mean_token_accuracy": 0.20793830901384353,
      "num_tokens": 47101296.0,
      "step": 27155
    },
    {
      "entropy": 5.7256042003631595,
      "epoch": 2.113129741295468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045487112095202544,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20082684308290483,
      "num_tokens": 47110033.0,
      "step": 27160
    },
    {
      "entropy": 5.678921222686768,
      "epoch": 2.113518770667185,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004548544690887181,
      "loss": 5.0512,
      "mean_token_accuracy": 0.19783466011285783,
      "num_tokens": 47119292.0,
      "step": 27165
    },
    {
      "entropy": 5.724234628677368,
      "epoch": 2.113907800038903,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004548378144964054,
      "loss": 5.0362,
      "mean_token_accuracy": 0.20080995559692383,
      "num_tokens": 47128081.0,
      "step": 27170
    },
    {
      "entropy": 5.776790380477905,
      "epoch": 2.1142968294106206,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045482115717534024,
      "loss": 5.0421,
      "mean_token_accuracy": 0.20386772006750106,
      "num_tokens": 47136639.0,
      "step": 27175
    },
    {
      "entropy": 5.725074768066406,
      "epoch": 2.1146858587823383,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004548044971257752,
      "loss": 5.0844,
      "mean_token_accuracy": 0.19528531581163405,
      "num_tokens": 47145830.0,
      "step": 27180
    },
    {
      "entropy": 5.668376016616821,
      "epoch": 2.1150748881540555,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045478783434796307,
      "loss": 5.0241,
      "mean_token_accuracy": 0.19877138286828994,
      "num_tokens": 47154695.0,
      "step": 27185
    },
    {
      "entropy": 5.711630487442017,
      "epoch": 2.1154639175257732,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045477116884215675,
      "loss": 5.028,
      "mean_token_accuracy": 0.20138407051563262,
      "num_tokens": 47163818.0,
      "step": 27190
    },
    {
      "entropy": 5.675554370880127,
      "epoch": 2.115852946897491,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000454754500608609,
      "loss": 4.9212,
      "mean_token_accuracy": 0.21019237637519836,
      "num_tokens": 47172426.0,
      "step": 27195
    },
    {
      "entropy": 5.7159984588623045,
      "epoch": 2.116241976269208,
      "grad_norm": 1.125,
      "learning_rate": 0.0004547378296475729,
      "loss": 5.1174,
      "mean_token_accuracy": 0.19318286031484605,
      "num_tokens": 47182034.0,
      "step": 27200
    },
    {
      "entropy": 5.687224578857422,
      "epoch": 2.116631005640926,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045472115595930124,
      "loss": 4.999,
      "mean_token_accuracy": 0.20143968760967254,
      "num_tokens": 47190968.0,
      "step": 27205
    },
    {
      "entropy": 5.609368085861206,
      "epoch": 2.1170200350126436,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045470447954404716,
      "loss": 4.9021,
      "mean_token_accuracy": 0.2096202254295349,
      "num_tokens": 47199917.0,
      "step": 27210
    },
    {
      "entropy": 5.672963333129883,
      "epoch": 2.117409064384361,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045468780040206364,
      "loss": 4.994,
      "mean_token_accuracy": 0.2040281981229782,
      "num_tokens": 47208467.0,
      "step": 27215
    },
    {
      "entropy": 5.6280388832092285,
      "epoch": 2.1177980937560785,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004546711185336037,
      "loss": 4.9761,
      "mean_token_accuracy": 0.2038068652153015,
      "num_tokens": 47216784.0,
      "step": 27220
    },
    {
      "entropy": 5.736069822311402,
      "epoch": 2.1181871231277962,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045465443393892047,
      "loss": 5.0335,
      "mean_token_accuracy": 0.2083165466785431,
      "num_tokens": 47226215.0,
      "step": 27225
    },
    {
      "entropy": 5.714542198181152,
      "epoch": 2.118576152499514,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004546377466182673,
      "loss": 4.962,
      "mean_token_accuracy": 0.21154684871435164,
      "num_tokens": 47234952.0,
      "step": 27230
    },
    {
      "entropy": 5.724227952957153,
      "epoch": 2.118965181871231,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004546210565718972,
      "loss": 5.0439,
      "mean_token_accuracy": 0.19571791738271713,
      "num_tokens": 47243897.0,
      "step": 27235
    },
    {
      "entropy": 5.6996622562408445,
      "epoch": 2.119354211242949,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045460436380006337,
      "loss": 4.9687,
      "mean_token_accuracy": 0.2075258880853653,
      "num_tokens": 47252171.0,
      "step": 27240
    },
    {
      "entropy": 5.700238132476807,
      "epoch": 2.1197432406146666,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004545876683030192,
      "loss": 4.957,
      "mean_token_accuracy": 0.2079739138484001,
      "num_tokens": 47260494.0,
      "step": 27245
    },
    {
      "entropy": 5.757265949249268,
      "epoch": 2.120132269986384,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004545709700810181,
      "loss": 5.1334,
      "mean_token_accuracy": 0.19683487862348556,
      "num_tokens": 47269485.0,
      "step": 27250
    },
    {
      "entropy": 5.57166428565979,
      "epoch": 2.1205212993581015,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004545542691343133,
      "loss": 4.8213,
      "mean_token_accuracy": 0.2149672254920006,
      "num_tokens": 47278275.0,
      "step": 27255
    },
    {
      "entropy": 5.666736650466919,
      "epoch": 2.120910328729819,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004545375654631582,
      "loss": 4.9443,
      "mean_token_accuracy": 0.20634870380163192,
      "num_tokens": 47287267.0,
      "step": 27260
    },
    {
      "entropy": 5.7010571479797365,
      "epoch": 2.1212993581015365,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045452085906780645,
      "loss": 5.0707,
      "mean_token_accuracy": 0.20313130617141723,
      "num_tokens": 47295785.0,
      "step": 27265
    },
    {
      "entropy": 5.734347724914551,
      "epoch": 2.121688387473254,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045450414994851123,
      "loss": 5.0025,
      "mean_token_accuracy": 0.20578503757715225,
      "num_tokens": 47303951.0,
      "step": 27270
    },
    {
      "entropy": 5.683846855163575,
      "epoch": 2.122077416844972,
      "grad_norm": 1.25,
      "learning_rate": 0.00045448743810552636,
      "loss": 4.9791,
      "mean_token_accuracy": 0.2063542753458023,
      "num_tokens": 47312226.0,
      "step": 27275
    },
    {
      "entropy": 5.673722267150879,
      "epoch": 2.1224664462166896,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045447072353910537,
      "loss": 5.0588,
      "mean_token_accuracy": 0.203763385117054,
      "num_tokens": 47320657.0,
      "step": 27280
    },
    {
      "entropy": 5.654352235794067,
      "epoch": 2.122855475588407,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004544540062495016,
      "loss": 4.9663,
      "mean_token_accuracy": 0.21708998680114747,
      "num_tokens": 47329063.0,
      "step": 27285
    },
    {
      "entropy": 5.681779861450195,
      "epoch": 2.1232445049601245,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045443728623696905,
      "loss": 4.9354,
      "mean_token_accuracy": 0.20447685569524765,
      "num_tokens": 47338304.0,
      "step": 27290
    },
    {
      "entropy": 5.691348648071289,
      "epoch": 2.123633534331842,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045442056350176123,
      "loss": 5.0573,
      "mean_token_accuracy": 0.19748046845197678,
      "num_tokens": 47347248.0,
      "step": 27295
    },
    {
      "entropy": 5.686807632446289,
      "epoch": 2.1240225637035595,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000454403838044132,
      "loss": 5.0524,
      "mean_token_accuracy": 0.2009979620575905,
      "num_tokens": 47356586.0,
      "step": 27300
    },
    {
      "entropy": 5.723837995529175,
      "epoch": 2.124411593075277,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000454387109864335,
      "loss": 4.9227,
      "mean_token_accuracy": 0.2070169672369957,
      "num_tokens": 47365195.0,
      "step": 27305
    },
    {
      "entropy": 5.722400712966919,
      "epoch": 2.124800622446995,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045437037896262425,
      "loss": 5.1387,
      "mean_token_accuracy": 0.18968493044376372,
      "num_tokens": 47374154.0,
      "step": 27310
    },
    {
      "entropy": 5.743993425369263,
      "epoch": 2.125189651818712,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004543536453392534,
      "loss": 4.9295,
      "mean_token_accuracy": 0.2105773866176605,
      "num_tokens": 47383253.0,
      "step": 27315
    },
    {
      "entropy": 5.7482421875,
      "epoch": 2.12557868119043,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045433690899447643,
      "loss": 5.003,
      "mean_token_accuracy": 0.2056488424539566,
      "num_tokens": 47391387.0,
      "step": 27320
    },
    {
      "entropy": 5.732699203491211,
      "epoch": 2.1259677105621475,
      "grad_norm": 1.125,
      "learning_rate": 0.00045432016992854734,
      "loss": 5.0384,
      "mean_token_accuracy": 0.20466195046901703,
      "num_tokens": 47400294.0,
      "step": 27325
    },
    {
      "entropy": 5.707462644577026,
      "epoch": 2.126356739933865,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004543034281417201,
      "loss": 4.9573,
      "mean_token_accuracy": 0.20490468442440032,
      "num_tokens": 47409024.0,
      "step": 27330
    },
    {
      "entropy": 5.716181087493896,
      "epoch": 2.1267457693055825,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004542866836342488,
      "loss": 4.9422,
      "mean_token_accuracy": 0.20737902522087098,
      "num_tokens": 47416974.0,
      "step": 27335
    },
    {
      "entropy": 5.703472566604614,
      "epoch": 2.1271347986773,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004542699364063873,
      "loss": 5.0463,
      "mean_token_accuracy": 0.20386471152305602,
      "num_tokens": 47425934.0,
      "step": 27340
    },
    {
      "entropy": 5.62651104927063,
      "epoch": 2.127523828049018,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004542531864583899,
      "loss": 5.0414,
      "mean_token_accuracy": 0.2014046400785446,
      "num_tokens": 47434764.0,
      "step": 27345
    },
    {
      "entropy": 5.783756351470947,
      "epoch": 2.127912857420735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004542364337905108,
      "loss": 5.0694,
      "mean_token_accuracy": 0.19161145836114885,
      "num_tokens": 47443435.0,
      "step": 27350
    },
    {
      "entropy": 5.751344442367554,
      "epoch": 2.128301886792453,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045421967840300403,
      "loss": 5.1019,
      "mean_token_accuracy": 0.1964049145579338,
      "num_tokens": 47452109.0,
      "step": 27355
    },
    {
      "entropy": 5.7265543937683105,
      "epoch": 2.1286909161641705,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004542029202961239,
      "loss": 4.9467,
      "mean_token_accuracy": 0.2081325888633728,
      "num_tokens": 47460319.0,
      "step": 27360
    },
    {
      "entropy": 5.762038469314575,
      "epoch": 2.1290799455358878,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045418615947012464,
      "loss": 5.1033,
      "mean_token_accuracy": 0.19344992637634278,
      "num_tokens": 47468806.0,
      "step": 27365
    },
    {
      "entropy": 5.718782520294189,
      "epoch": 2.1294689749076054,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004541693959252607,
      "loss": 5.0311,
      "mean_token_accuracy": 0.20244591385126115,
      "num_tokens": 47477121.0,
      "step": 27370
    },
    {
      "entropy": 5.561141729354858,
      "epoch": 2.129858004279323,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045415262966178635,
      "loss": 4.8542,
      "mean_token_accuracy": 0.21646149456501007,
      "num_tokens": 47485379.0,
      "step": 27375
    },
    {
      "entropy": 5.686192512512207,
      "epoch": 2.130247033651041,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004541358606799559,
      "loss": 5.0313,
      "mean_token_accuracy": 0.20052382349967957,
      "num_tokens": 47493402.0,
      "step": 27380
    },
    {
      "entropy": 5.764338111877441,
      "epoch": 2.130636063022758,
      "grad_norm": 1.40625,
      "learning_rate": 0.000454119088980024,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20788115859031678,
      "num_tokens": 47503148.0,
      "step": 27385
    },
    {
      "entropy": 5.715002536773682,
      "epoch": 2.131025092394476,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004541023145622449,
      "loss": 4.957,
      "mean_token_accuracy": 0.20351082533597947,
      "num_tokens": 47511681.0,
      "step": 27390
    },
    {
      "entropy": 5.646071338653565,
      "epoch": 2.1314141217661935,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045408553742687336,
      "loss": 4.9329,
      "mean_token_accuracy": 0.2098097398877144,
      "num_tokens": 47520240.0,
      "step": 27395
    },
    {
      "entropy": 5.704679489135742,
      "epoch": 2.1318031511379107,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004540687575741638,
      "loss": 4.9945,
      "mean_token_accuracy": 0.20268191695213317,
      "num_tokens": 47528553.0,
      "step": 27400
    },
    {
      "entropy": 5.721995401382446,
      "epoch": 2.1321921805096284,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045405197500437077,
      "loss": 5.0235,
      "mean_token_accuracy": 0.20466084331274031,
      "num_tokens": 47537735.0,
      "step": 27405
    },
    {
      "entropy": 5.656980514526367,
      "epoch": 2.132581209881346,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045403518971774915,
      "loss": 4.9134,
      "mean_token_accuracy": 0.2093452602624893,
      "num_tokens": 47545251.0,
      "step": 27410
    },
    {
      "entropy": 5.713734912872314,
      "epoch": 2.132970239253064,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045401840171455343,
      "loss": 5.0079,
      "mean_token_accuracy": 0.19672364443540574,
      "num_tokens": 47553962.0,
      "step": 27415
    },
    {
      "entropy": 5.700325155258179,
      "epoch": 2.133359268624781,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004540016109950384,
      "loss": 5.0413,
      "mean_token_accuracy": 0.20518169105052947,
      "num_tokens": 47562401.0,
      "step": 27420
    },
    {
      "entropy": 5.719853401184082,
      "epoch": 2.133748297996499,
      "grad_norm": 1.25,
      "learning_rate": 0.0004539848175594589,
      "loss": 4.9886,
      "mean_token_accuracy": 0.2078239843249321,
      "num_tokens": 47570097.0,
      "step": 27425
    },
    {
      "entropy": 5.722873258590698,
      "epoch": 2.1341373273682165,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004539680214080695,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19602590352296828,
      "num_tokens": 47578705.0,
      "step": 27430
    },
    {
      "entropy": 5.6723301887512205,
      "epoch": 2.1345263567399337,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045395122254112536,
      "loss": 5.0235,
      "mean_token_accuracy": 0.19966764599084855,
      "num_tokens": 47588075.0,
      "step": 27435
    },
    {
      "entropy": 5.649790525436401,
      "epoch": 2.1349153861116514,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004539344209588812,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2063824102282524,
      "num_tokens": 47596906.0,
      "step": 27440
    },
    {
      "entropy": 5.678345775604248,
      "epoch": 2.135304415483369,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045391761666159204,
      "loss": 4.9776,
      "mean_token_accuracy": 0.21034736186265945,
      "num_tokens": 47605427.0,
      "step": 27445
    },
    {
      "entropy": 5.781550598144531,
      "epoch": 2.1356934448550864,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045390080964951294,
      "loss": 5.1121,
      "mean_token_accuracy": 0.19671715497970582,
      "num_tokens": 47614249.0,
      "step": 27450
    },
    {
      "entropy": 5.672934436798096,
      "epoch": 2.136082474226804,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045388399992289864,
      "loss": 4.9333,
      "mean_token_accuracy": 0.20291397124528884,
      "num_tokens": 47623200.0,
      "step": 27455
    },
    {
      "entropy": 5.725113725662231,
      "epoch": 2.136471503598522,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045386718748200436,
      "loss": 5.0333,
      "mean_token_accuracy": 0.2016819268465042,
      "num_tokens": 47632392.0,
      "step": 27460
    },
    {
      "entropy": 5.8106883525848385,
      "epoch": 2.136860532970239,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045385037232708534,
      "loss": 5.1098,
      "mean_token_accuracy": 0.20248483419418334,
      "num_tokens": 47641648.0,
      "step": 27465
    },
    {
      "entropy": 5.7682336330413815,
      "epoch": 2.1372495623419567,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004538335544583964,
      "loss": 4.9852,
      "mean_token_accuracy": 0.20656303614377974,
      "num_tokens": 47651113.0,
      "step": 27470
    },
    {
      "entropy": 5.757300996780396,
      "epoch": 2.1376385917136744,
      "grad_norm": 1.25,
      "learning_rate": 0.00045381673387619306,
      "loss": 5.1525,
      "mean_token_accuracy": 0.195793953537941,
      "num_tokens": 47659959.0,
      "step": 27475
    },
    {
      "entropy": 5.681690835952759,
      "epoch": 2.138027621085392,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045379991058073035,
      "loss": 5.0732,
      "mean_token_accuracy": 0.19897158443927765,
      "num_tokens": 47668738.0,
      "step": 27480
    },
    {
      "entropy": 5.606641960144043,
      "epoch": 2.1384166504571094,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004537830845722636,
      "loss": 4.9181,
      "mean_token_accuracy": 0.211308616399765,
      "num_tokens": 47677268.0,
      "step": 27485
    },
    {
      "entropy": 5.673650026321411,
      "epoch": 2.138805679828827,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004537662558510481,
      "loss": 4.912,
      "mean_token_accuracy": 0.20991974771022798,
      "num_tokens": 47685339.0,
      "step": 27490
    },
    {
      "entropy": 5.721701431274414,
      "epoch": 2.1391947092005448,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045374942441733925,
      "loss": 5.0032,
      "mean_token_accuracy": 0.2025295227766037,
      "num_tokens": 47694729.0,
      "step": 27495
    },
    {
      "entropy": 5.672301292419434,
      "epoch": 2.139583738572262,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004537325902713923,
      "loss": 4.9777,
      "mean_token_accuracy": 0.20211876183748245,
      "num_tokens": 47703555.0,
      "step": 27500
    },
    {
      "entropy": 5.716456317901612,
      "epoch": 2.1399727679439797,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004537157534134629,
      "loss": 5.0313,
      "mean_token_accuracy": 0.1967541590332985,
      "num_tokens": 47713115.0,
      "step": 27505
    },
    {
      "entropy": 5.7185218334198,
      "epoch": 2.1403617973156974,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045369891384380623,
      "loss": 5.0326,
      "mean_token_accuracy": 0.2072561040520668,
      "num_tokens": 47722263.0,
      "step": 27510
    },
    {
      "entropy": 5.718893814086914,
      "epoch": 2.140750826687415,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004536820715626781,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20263720899820328,
      "num_tokens": 47730037.0,
      "step": 27515
    },
    {
      "entropy": 5.65296688079834,
      "epoch": 2.1411398560591324,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045366522657033385,
      "loss": 4.9154,
      "mean_token_accuracy": 0.2206691473722458,
      "num_tokens": 47738968.0,
      "step": 27520
    },
    {
      "entropy": 5.698263883590698,
      "epoch": 2.14152888543085,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045364837886702924,
      "loss": 4.9855,
      "mean_token_accuracy": 0.2060439646244049,
      "num_tokens": 47747515.0,
      "step": 27525
    },
    {
      "entropy": 5.667622709274292,
      "epoch": 2.1419179148025678,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004536315284530198,
      "loss": 4.963,
      "mean_token_accuracy": 0.2136923313140869,
      "num_tokens": 47756440.0,
      "step": 27530
    },
    {
      "entropy": 5.6969531059265135,
      "epoch": 2.142306944174285,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004536146753285612,
      "loss": 5.0288,
      "mean_token_accuracy": 0.1925949364900589,
      "num_tokens": 47765823.0,
      "step": 27535
    },
    {
      "entropy": 5.716357088088989,
      "epoch": 2.1426959735460027,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004535978194939093,
      "loss": 4.9839,
      "mean_token_accuracy": 0.20726005584001542,
      "num_tokens": 47775423.0,
      "step": 27540
    },
    {
      "entropy": 5.710713577270508,
      "epoch": 2.1430850029177204,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045358096094931966,
      "loss": 5.0465,
      "mean_token_accuracy": 0.19402572363615037,
      "num_tokens": 47784276.0,
      "step": 27545
    },
    {
      "entropy": 5.679698657989502,
      "epoch": 2.1434740322894377,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004535640996950482,
      "loss": 4.9451,
      "mean_token_accuracy": 0.21284402310848236,
      "num_tokens": 47792608.0,
      "step": 27550
    },
    {
      "entropy": 5.659192371368408,
      "epoch": 2.1438630616611554,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004535472357313507,
      "loss": 5.0316,
      "mean_token_accuracy": 0.20178054124116898,
      "num_tokens": 47802356.0,
      "step": 27555
    },
    {
      "entropy": 5.701833009719849,
      "epoch": 2.144252091032873,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045353036905848316,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20415971726179122,
      "num_tokens": 47811307.0,
      "step": 27560
    },
    {
      "entropy": 5.791537523269653,
      "epoch": 2.1446411204045903,
      "grad_norm": 1.125,
      "learning_rate": 0.0004535134996767014,
      "loss": 5.1092,
      "mean_token_accuracy": 0.195357084274292,
      "num_tokens": 47820572.0,
      "step": 27565
    },
    {
      "entropy": 5.742964267730713,
      "epoch": 2.145030149776308,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004534966275862614,
      "loss": 5.0789,
      "mean_token_accuracy": 0.20082349479198455,
      "num_tokens": 47829053.0,
      "step": 27570
    },
    {
      "entropy": 5.766876602172852,
      "epoch": 2.1454191791480257,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004534797527874192,
      "loss": 4.9838,
      "mean_token_accuracy": 0.20230368822813033,
      "num_tokens": 47838244.0,
      "step": 27575
    },
    {
      "entropy": 5.692704486846924,
      "epoch": 2.1458082085197434,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045346287528043083,
      "loss": 4.9235,
      "mean_token_accuracy": 0.2118486151099205,
      "num_tokens": 47846908.0,
      "step": 27580
    },
    {
      "entropy": 5.706371879577636,
      "epoch": 2.1461972378914607,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045344599506555244,
      "loss": 5.0736,
      "mean_token_accuracy": 0.20030611902475357,
      "num_tokens": 47855824.0,
      "step": 27585
    },
    {
      "entropy": 5.67712435722351,
      "epoch": 2.1465862672631784,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045342911214304,
      "loss": 4.9183,
      "mean_token_accuracy": 0.20333437174558638,
      "num_tokens": 47864641.0,
      "step": 27590
    },
    {
      "entropy": 5.742670106887817,
      "epoch": 2.146975296634896,
      "grad_norm": 1.25,
      "learning_rate": 0.0004534122265131498,
      "loss": 5.0939,
      "mean_token_accuracy": 0.19816931188106537,
      "num_tokens": 47873721.0,
      "step": 27595
    },
    {
      "entropy": 5.701150512695312,
      "epoch": 2.1473643260066133,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045339533817613807,
      "loss": 4.9352,
      "mean_token_accuracy": 0.21271092295646668,
      "num_tokens": 47882485.0,
      "step": 27600
    },
    {
      "entropy": 5.655875396728516,
      "epoch": 2.147753355378331,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045337844713226106,
      "loss": 4.8788,
      "mean_token_accuracy": 0.20662842839956283,
      "num_tokens": 47891574.0,
      "step": 27605
    },
    {
      "entropy": 5.635758781433106,
      "epoch": 2.1481423847500487,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045336155338177497,
      "loss": 4.8731,
      "mean_token_accuracy": 0.21360377222299576,
      "num_tokens": 47900123.0,
      "step": 27610
    },
    {
      "entropy": 5.6526069164276125,
      "epoch": 2.1485314141217664,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004533446569249362,
      "loss": 4.9978,
      "mean_token_accuracy": 0.20859355628490447,
      "num_tokens": 47908342.0,
      "step": 27615
    },
    {
      "entropy": 5.666867113113403,
      "epoch": 2.1489204434934837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045332775776200116,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21033590883016587,
      "num_tokens": 47917365.0,
      "step": 27620
    },
    {
      "entropy": 5.664396047592163,
      "epoch": 2.1493094728652014,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045331085589322626,
      "loss": 4.9375,
      "mean_token_accuracy": 0.20193236023187638,
      "num_tokens": 47925839.0,
      "step": 27625
    },
    {
      "entropy": 5.649382591247559,
      "epoch": 2.149698502236919,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004532939513188679,
      "loss": 4.8774,
      "mean_token_accuracy": 0.2188877433538437,
      "num_tokens": 47934408.0,
      "step": 27630
    },
    {
      "entropy": 5.695407199859619,
      "epoch": 2.1500875316086363,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004532770440391826,
      "loss": 5.0192,
      "mean_token_accuracy": 0.197219780087471,
      "num_tokens": 47943486.0,
      "step": 27635
    },
    {
      "entropy": 5.742846393585205,
      "epoch": 2.150476560980354,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004532601340544269,
      "loss": 5.0478,
      "mean_token_accuracy": 0.2034695565700531,
      "num_tokens": 47952809.0,
      "step": 27640
    },
    {
      "entropy": 5.729238986968994,
      "epoch": 2.1508655903520717,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004532432213648574,
      "loss": 4.9385,
      "mean_token_accuracy": 0.2084517851471901,
      "num_tokens": 47961104.0,
      "step": 27645
    },
    {
      "entropy": 5.677875089645386,
      "epoch": 2.151254619723789,
      "grad_norm": 1.25,
      "learning_rate": 0.00045322630597073067,
      "loss": 4.9793,
      "mean_token_accuracy": 0.19941056966781617,
      "num_tokens": 47969442.0,
      "step": 27650
    },
    {
      "entropy": 5.597454595565796,
      "epoch": 2.1516436490955066,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045320938787230355,
      "loss": 4.8951,
      "mean_token_accuracy": 0.2114834189414978,
      "num_tokens": 47978703.0,
      "step": 27655
    },
    {
      "entropy": 5.736273527145386,
      "epoch": 2.1520326784672243,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045319246706983257,
      "loss": 5.0271,
      "mean_token_accuracy": 0.20214398950338364,
      "num_tokens": 47986886.0,
      "step": 27660
    },
    {
      "entropy": 5.7328636169433596,
      "epoch": 2.152421707838942,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045317554356357446,
      "loss": 4.9328,
      "mean_token_accuracy": 0.21054258942604065,
      "num_tokens": 47995278.0,
      "step": 27665
    },
    {
      "entropy": 5.681626224517823,
      "epoch": 2.1528107372106593,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045315861735378615,
      "loss": 4.9773,
      "mean_token_accuracy": 0.2038572072982788,
      "num_tokens": 48004351.0,
      "step": 27670
    },
    {
      "entropy": 5.722432374954224,
      "epoch": 2.153199766582377,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045314168844072435,
      "loss": 4.9813,
      "mean_token_accuracy": 0.20663430243730546,
      "num_tokens": 48013346.0,
      "step": 27675
    },
    {
      "entropy": 5.693291234970093,
      "epoch": 2.1535887959540947,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045312475682464604,
      "loss": 5.0587,
      "mean_token_accuracy": 0.2035825252532959,
      "num_tokens": 48023371.0,
      "step": 27680
    },
    {
      "entropy": 5.642207241058349,
      "epoch": 2.153977825325812,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045310782250580794,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2094508081674576,
      "num_tokens": 48032095.0,
      "step": 27685
    },
    {
      "entropy": 5.684958410263062,
      "epoch": 2.1543668546975296,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004530908854844672,
      "loss": 4.9224,
      "mean_token_accuracy": 0.20611189305782318,
      "num_tokens": 48040818.0,
      "step": 27690
    },
    {
      "entropy": 5.664567470550537,
      "epoch": 2.1547558840692473,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045307394576088076,
      "loss": 4.9085,
      "mean_token_accuracy": 0.21852250546216964,
      "num_tokens": 48049737.0,
      "step": 27695
    },
    {
      "entropy": 5.595551872253418,
      "epoch": 2.1551449134409646,
      "grad_norm": 1.125,
      "learning_rate": 0.0004530570033353056,
      "loss": 4.8788,
      "mean_token_accuracy": 0.2068950951099396,
      "num_tokens": 48057659.0,
      "step": 27700
    },
    {
      "entropy": 5.655428886413574,
      "epoch": 2.1555339428126823,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045304005820799874,
      "loss": 4.8456,
      "mean_token_accuracy": 0.2075860545039177,
      "num_tokens": 48066578.0,
      "step": 27705
    },
    {
      "entropy": 5.688319873809815,
      "epoch": 2.1559229721844,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004530231103792175,
      "loss": 5.0395,
      "mean_token_accuracy": 0.20110439360141755,
      "num_tokens": 48076051.0,
      "step": 27710
    },
    {
      "entropy": 5.7266932964324955,
      "epoch": 2.1563120015561177,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004530061598492188,
      "loss": 5.054,
      "mean_token_accuracy": 0.20010719150304795,
      "num_tokens": 48084722.0,
      "step": 27715
    },
    {
      "entropy": 5.647017955780029,
      "epoch": 2.156701030927835,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045298920661826004,
      "loss": 4.8658,
      "mean_token_accuracy": 0.21423972845077516,
      "num_tokens": 48092890.0,
      "step": 27720
    },
    {
      "entropy": 5.720371866226197,
      "epoch": 2.1570900602995526,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004529722506865984,
      "loss": 4.998,
      "mean_token_accuracy": 0.20528857707977294,
      "num_tokens": 48101146.0,
      "step": 27725
    },
    {
      "entropy": 5.70270299911499,
      "epoch": 2.1574790896712703,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000452955292054491,
      "loss": 5.0043,
      "mean_token_accuracy": 0.20414438396692275,
      "num_tokens": 48109974.0,
      "step": 27730
    },
    {
      "entropy": 5.728003740310669,
      "epoch": 2.1578681190429876,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045293833072219535,
      "loss": 5.061,
      "mean_token_accuracy": 0.2004196286201477,
      "num_tokens": 48118325.0,
      "step": 27735
    },
    {
      "entropy": 5.746850919723511,
      "epoch": 2.1582571484147053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045292136668996876,
      "loss": 5.0917,
      "mean_token_accuracy": 0.19573197066783904,
      "num_tokens": 48128255.0,
      "step": 27740
    },
    {
      "entropy": 5.699099397659301,
      "epoch": 2.158646177786423,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004529043999580686,
      "loss": 4.9114,
      "mean_token_accuracy": 0.2118326410651207,
      "num_tokens": 48136763.0,
      "step": 27745
    },
    {
      "entropy": 5.652976036071777,
      "epoch": 2.1590352071581402,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045288743052675234,
      "loss": 4.9159,
      "mean_token_accuracy": 0.21486248522996904,
      "num_tokens": 48146371.0,
      "step": 27750
    },
    {
      "entropy": 5.694534492492676,
      "epoch": 2.159424236529858,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045287045839627744,
      "loss": 4.9901,
      "mean_token_accuracy": 0.21100818663835524,
      "num_tokens": 48155180.0,
      "step": 27755
    },
    {
      "entropy": 5.617306852340699,
      "epoch": 2.1598132659015756,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045285348356690155,
      "loss": 4.9559,
      "mean_token_accuracy": 0.20671525001525878,
      "num_tokens": 48163728.0,
      "step": 27760
    },
    {
      "entropy": 5.658135080337525,
      "epoch": 2.1602022952732933,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000452836506038882,
      "loss": 4.9737,
      "mean_token_accuracy": 0.2010297641158104,
      "num_tokens": 48171844.0,
      "step": 27765
    },
    {
      "entropy": 5.759475517272949,
      "epoch": 2.1605913246450106,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045281952581247654,
      "loss": 5.0525,
      "mean_token_accuracy": 0.2034651219844818,
      "num_tokens": 48180617.0,
      "step": 27770
    },
    {
      "entropy": 5.69092288017273,
      "epoch": 2.1609803540167283,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045280254288794286,
      "loss": 4.9342,
      "mean_token_accuracy": 0.20536487996578218,
      "num_tokens": 48188875.0,
      "step": 27775
    },
    {
      "entropy": 5.591668033599854,
      "epoch": 2.161369383388446,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045278555726553855,
      "loss": 4.88,
      "mean_token_accuracy": 0.20635617673397064,
      "num_tokens": 48197280.0,
      "step": 27780
    },
    {
      "entropy": 5.701783990859985,
      "epoch": 2.1617584127601632,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045276856894552143,
      "loss": 5.0697,
      "mean_token_accuracy": 0.20198207944631577,
      "num_tokens": 48206907.0,
      "step": 27785
    },
    {
      "entropy": 5.661320352554322,
      "epoch": 2.162147442131881,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004527515779281492,
      "loss": 4.8941,
      "mean_token_accuracy": 0.21254518181085585,
      "num_tokens": 48215334.0,
      "step": 27790
    },
    {
      "entropy": 5.675133943557739,
      "epoch": 2.1625364715035986,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045273458421367976,
      "loss": 5.0164,
      "mean_token_accuracy": 0.20372989028692245,
      "num_tokens": 48224667.0,
      "step": 27795
    },
    {
      "entropy": 5.694805145263672,
      "epoch": 2.162925500875316,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004527175878023708,
      "loss": 5.0729,
      "mean_token_accuracy": 0.20298838019371032,
      "num_tokens": 48233237.0,
      "step": 27800
    },
    {
      "entropy": 5.701748943328857,
      "epoch": 2.1633145302470336,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004527005886944803,
      "loss": 4.9758,
      "mean_token_accuracy": 0.20144418478012086,
      "num_tokens": 48241048.0,
      "step": 27805
    },
    {
      "entropy": 5.781676864624023,
      "epoch": 2.1637035596187513,
      "grad_norm": 1.125,
      "learning_rate": 0.00045268358689026626,
      "loss": 5.0762,
      "mean_token_accuracy": 0.20464471280574797,
      "num_tokens": 48250013.0,
      "step": 27810
    },
    {
      "entropy": 5.775678730010986,
      "epoch": 2.164092588990469,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004526665823899866,
      "loss": 5.0554,
      "mean_token_accuracy": 0.20379042327404023,
      "num_tokens": 48258566.0,
      "step": 27815
    },
    {
      "entropy": 5.6985467910766605,
      "epoch": 2.164481618362186,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045264957519389936,
      "loss": 4.9191,
      "mean_token_accuracy": 0.20683735311031343,
      "num_tokens": 48266843.0,
      "step": 27820
    },
    {
      "entropy": 5.644751358032226,
      "epoch": 2.164870647733904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045263256530226253,
      "loss": 4.8997,
      "mean_token_accuracy": 0.2079235076904297,
      "num_tokens": 48276138.0,
      "step": 27825
    },
    {
      "entropy": 5.627683544158936,
      "epoch": 2.1652596771056216,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004526155527153343,
      "loss": 4.8477,
      "mean_token_accuracy": 0.21218783408403397,
      "num_tokens": 48284452.0,
      "step": 27830
    },
    {
      "entropy": 5.741854667663574,
      "epoch": 2.165648706477339,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004525985374333727,
      "loss": 5.0011,
      "mean_token_accuracy": 0.20259683579206467,
      "num_tokens": 48292871.0,
      "step": 27835
    },
    {
      "entropy": 5.670397710800171,
      "epoch": 2.1660377358490566,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000452581519456636,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20041822046041488,
      "num_tokens": 48300642.0,
      "step": 27840
    },
    {
      "entropy": 5.712048673629761,
      "epoch": 2.1664267652207743,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045256449878538243,
      "loss": 5.0652,
      "mean_token_accuracy": 0.19402231723070146,
      "num_tokens": 48309229.0,
      "step": 27845
    },
    {
      "entropy": 5.771245098114013,
      "epoch": 2.166815794592492,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045254747541987017,
      "loss": 4.9991,
      "mean_token_accuracy": 0.1950952872633934,
      "num_tokens": 48317474.0,
      "step": 27850
    },
    {
      "entropy": 5.757614374160767,
      "epoch": 2.167204823964209,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004525304493603576,
      "loss": 5.0661,
      "mean_token_accuracy": 0.203395713865757,
      "num_tokens": 48325810.0,
      "step": 27855
    },
    {
      "entropy": 5.708132982254028,
      "epoch": 2.167593853335927,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045251342060710295,
      "loss": 4.9907,
      "mean_token_accuracy": 0.20279304534196854,
      "num_tokens": 48334348.0,
      "step": 27860
    },
    {
      "entropy": 5.607769870758057,
      "epoch": 2.1679828827076446,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004524963891603647,
      "loss": 4.8415,
      "mean_token_accuracy": 0.20977737605571747,
      "num_tokens": 48342565.0,
      "step": 27865
    },
    {
      "entropy": 5.6819226264953615,
      "epoch": 2.168371912079362,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045247935502040136,
      "loss": 5.087,
      "mean_token_accuracy": 0.1961832895874977,
      "num_tokens": 48351398.0,
      "step": 27870
    },
    {
      "entropy": 5.667052841186523,
      "epoch": 2.1687609414510796,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004524623181874712,
      "loss": 4.9843,
      "mean_token_accuracy": 0.20053137689828873,
      "num_tokens": 48360275.0,
      "step": 27875
    },
    {
      "entropy": 5.687873888015747,
      "epoch": 2.1691499708227973,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004524452786618329,
      "loss": 4.9426,
      "mean_token_accuracy": 0.20312148332595825,
      "num_tokens": 48369516.0,
      "step": 27880
    },
    {
      "entropy": 5.661909532546997,
      "epoch": 2.1695390001945145,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045242823644374484,
      "loss": 4.951,
      "mean_token_accuracy": 0.2072077289223671,
      "num_tokens": 48378412.0,
      "step": 27885
    },
    {
      "entropy": 5.714970302581787,
      "epoch": 2.169928029566232,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004524111915334658,
      "loss": 5.022,
      "mean_token_accuracy": 0.20247285813093185,
      "num_tokens": 48386566.0,
      "step": 27890
    },
    {
      "entropy": 5.757609033584595,
      "epoch": 2.17031705893795,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045239414393125424,
      "loss": 5.0151,
      "mean_token_accuracy": 0.19871056973934173,
      "num_tokens": 48394727.0,
      "step": 27895
    },
    {
      "entropy": 5.767094707489013,
      "epoch": 2.170706088309667,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004523770936373689,
      "loss": 5.0734,
      "mean_token_accuracy": 0.1999158039689064,
      "num_tokens": 48404440.0,
      "step": 27900
    },
    {
      "entropy": 5.691420316696167,
      "epoch": 2.171095117681385,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004523600406520685,
      "loss": 4.9397,
      "mean_token_accuracy": 0.20646082609891891,
      "num_tokens": 48413441.0,
      "step": 27905
    },
    {
      "entropy": 5.741039657592774,
      "epoch": 2.1714841470531026,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004523429849756118,
      "loss": 5.0655,
      "mean_token_accuracy": 0.19967739433050155,
      "num_tokens": 48422393.0,
      "step": 27910
    },
    {
      "entropy": 5.712139701843261,
      "epoch": 2.1718731764248203,
      "grad_norm": 1.25,
      "learning_rate": 0.0004523259266082576,
      "loss": 5.0062,
      "mean_token_accuracy": 0.21397970914840697,
      "num_tokens": 48431151.0,
      "step": 27915
    },
    {
      "entropy": 5.684113931655884,
      "epoch": 2.1722622057965375,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045230886555026455,
      "loss": 4.9466,
      "mean_token_accuracy": 0.21216524839401246,
      "num_tokens": 48439453.0,
      "step": 27920
    },
    {
      "entropy": 5.748486948013306,
      "epoch": 2.172651235168255,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045229180180189176,
      "loss": 5.027,
      "mean_token_accuracy": 0.20650583505630493,
      "num_tokens": 48448055.0,
      "step": 27925
    },
    {
      "entropy": 5.710920524597168,
      "epoch": 2.173040264539973,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045227473536339817,
      "loss": 4.9027,
      "mean_token_accuracy": 0.21643728613853455,
      "num_tokens": 48456015.0,
      "step": 27930
    },
    {
      "entropy": 5.666434478759766,
      "epoch": 2.17342929391169,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004522576662350425,
      "loss": 5.001,
      "mean_token_accuracy": 0.20859433114528655,
      "num_tokens": 48464389.0,
      "step": 27935
    },
    {
      "entropy": 5.707480382919312,
      "epoch": 2.173818323283408,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000452240594417084,
      "loss": 4.945,
      "mean_token_accuracy": 0.20088809728622437,
      "num_tokens": 48472403.0,
      "step": 27940
    },
    {
      "entropy": 5.644100189208984,
      "epoch": 2.1742073526551255,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004522235199097815,
      "loss": 4.9903,
      "mean_token_accuracy": 0.19898601323366166,
      "num_tokens": 48480879.0,
      "step": 27945
    },
    {
      "entropy": 5.644384384155273,
      "epoch": 2.1745963820268432,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004522064427133942,
      "loss": 4.9621,
      "mean_token_accuracy": 0.20367229878902435,
      "num_tokens": 48489722.0,
      "step": 27950
    },
    {
      "entropy": 5.6606639385223385,
      "epoch": 2.1749854113985605,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045218936282818116,
      "loss": 4.9732,
      "mean_token_accuracy": 0.2055150479078293,
      "num_tokens": 48499023.0,
      "step": 27955
    },
    {
      "entropy": 5.743335723876953,
      "epoch": 2.175374440770278,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004521722802544016,
      "loss": 5.0285,
      "mean_token_accuracy": 0.21172675639390945,
      "num_tokens": 48508352.0,
      "step": 27960
    },
    {
      "entropy": 5.6740015029907225,
      "epoch": 2.175763470141996,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045215519499231465,
      "loss": 4.883,
      "mean_token_accuracy": 0.2197446718811989,
      "num_tokens": 48517481.0,
      "step": 27965
    },
    {
      "entropy": 5.672723960876465,
      "epoch": 2.176152499513713,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045213810704217963,
      "loss": 4.9661,
      "mean_token_accuracy": 0.20563762485980988,
      "num_tokens": 48525777.0,
      "step": 27970
    },
    {
      "entropy": 5.629616355895996,
      "epoch": 2.176541528885431,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004521210164042557,
      "loss": 4.8746,
      "mean_token_accuracy": 0.2062548115849495,
      "num_tokens": 48534297.0,
      "step": 27975
    },
    {
      "entropy": 5.708858251571655,
      "epoch": 2.1769305582571485,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004521039230788024,
      "loss": 4.9882,
      "mean_token_accuracy": 0.21162746995687484,
      "num_tokens": 48543425.0,
      "step": 27980
    },
    {
      "entropy": 5.664286327362061,
      "epoch": 2.177319587628866,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004520868270660789,
      "loss": 4.9389,
      "mean_token_accuracy": 0.2029540240764618,
      "num_tokens": 48553114.0,
      "step": 27985
    },
    {
      "entropy": 5.813627672195435,
      "epoch": 2.1777086170005835,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004520697283663446,
      "loss": 5.0163,
      "mean_token_accuracy": 0.20899298042058945,
      "num_tokens": 48561409.0,
      "step": 27990
    },
    {
      "entropy": 5.7239377975463865,
      "epoch": 2.178097646372301,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045205262697985897,
      "loss": 5.0141,
      "mean_token_accuracy": 0.19575997292995453,
      "num_tokens": 48569782.0,
      "step": 27995
    },
    {
      "entropy": 5.716666793823242,
      "epoch": 2.1784866757440184,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004520355229068816,
      "loss": 5.0797,
      "mean_token_accuracy": 0.19586392641067504,
      "num_tokens": 48579283.0,
      "step": 28000
    },
    {
      "entropy": 5.744035005569458,
      "epoch": 2.178875705115736,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045201841614767203,
      "loss": 5.0012,
      "mean_token_accuracy": 0.20357227474451065,
      "num_tokens": 48587649.0,
      "step": 28005
    },
    {
      "entropy": 5.74961748123169,
      "epoch": 2.179264734487454,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045200130670248974,
      "loss": 4.9978,
      "mean_token_accuracy": 0.20005417764186859,
      "num_tokens": 48596227.0,
      "step": 28010
    },
    {
      "entropy": 5.654628849029541,
      "epoch": 2.1796537638591715,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045198419457159424,
      "loss": 4.947,
      "mean_token_accuracy": 0.19708692133426667,
      "num_tokens": 48605364.0,
      "step": 28015
    },
    {
      "entropy": 5.685307884216309,
      "epoch": 2.180042793230889,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004519670797552453,
      "loss": 5.0421,
      "mean_token_accuracy": 0.2038318023085594,
      "num_tokens": 48613794.0,
      "step": 28020
    },
    {
      "entropy": 5.741675233840942,
      "epoch": 2.1804318226026065,
      "grad_norm": 1.375,
      "learning_rate": 0.00045194996225370266,
      "loss": 5.012,
      "mean_token_accuracy": 0.19756045788526536,
      "num_tokens": 48622381.0,
      "step": 28025
    },
    {
      "entropy": 5.808018445968628,
      "epoch": 2.180820851974324,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045193284206722595,
      "loss": 5.1455,
      "mean_token_accuracy": 0.19119613915681838,
      "num_tokens": 48631484.0,
      "step": 28030
    },
    {
      "entropy": 5.693644571304321,
      "epoch": 2.1812098813460414,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004519157191960749,
      "loss": 5.0294,
      "mean_token_accuracy": 0.20057007670402527,
      "num_tokens": 48641082.0,
      "step": 28035
    },
    {
      "entropy": 5.722043466567993,
      "epoch": 2.181598910717759,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004518985936405096,
      "loss": 4.9872,
      "mean_token_accuracy": 0.20346074402332306,
      "num_tokens": 48649974.0,
      "step": 28040
    },
    {
      "entropy": 5.755811309814453,
      "epoch": 2.181987940089477,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004518814654007896,
      "loss": 5.0201,
      "mean_token_accuracy": 0.20758064687252045,
      "num_tokens": 48658633.0,
      "step": 28045
    },
    {
      "entropy": 5.664405822753906,
      "epoch": 2.1823769694611945,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045186433447717486,
      "loss": 4.9898,
      "mean_token_accuracy": 0.2016853481531143,
      "num_tokens": 48667313.0,
      "step": 28050
    },
    {
      "entropy": 5.7164473056793215,
      "epoch": 2.182765998832912,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045184720086992536,
      "loss": 5.0518,
      "mean_token_accuracy": 0.19998390972614288,
      "num_tokens": 48676664.0,
      "step": 28055
    },
    {
      "entropy": 5.70733413696289,
      "epoch": 2.1831550282046295,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045183006457930107,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21213295608758925,
      "num_tokens": 48684770.0,
      "step": 28060
    },
    {
      "entropy": 5.659025430679321,
      "epoch": 2.183544057576347,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004518129256055621,
      "loss": 4.9347,
      "mean_token_accuracy": 0.20799660086631774,
      "num_tokens": 48692965.0,
      "step": 28065
    },
    {
      "entropy": 5.774905443191528,
      "epoch": 2.1839330869480644,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004517957839489683,
      "loss": 5.0308,
      "mean_token_accuracy": 0.20385248214006424,
      "num_tokens": 48702273.0,
      "step": 28070
    },
    {
      "entropy": 5.674905395507812,
      "epoch": 2.184322116319782,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045177863960977986,
      "loss": 4.991,
      "mean_token_accuracy": 0.1967081233859062,
      "num_tokens": 48710884.0,
      "step": 28075
    },
    {
      "entropy": 5.810346031188965,
      "epoch": 2.1847111456915,
      "grad_norm": 1.28125,
      "learning_rate": 0.000451761492588257,
      "loss": 5.1314,
      "mean_token_accuracy": 0.18994418680667877,
      "num_tokens": 48719673.0,
      "step": 28080
    },
    {
      "entropy": 5.689353704452515,
      "epoch": 2.185100175063217,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045174434288465986,
      "loss": 4.9893,
      "mean_token_accuracy": 0.21025285720825196,
      "num_tokens": 48728084.0,
      "step": 28085
    },
    {
      "entropy": 5.74102201461792,
      "epoch": 2.1854892044349348,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004517271904992486,
      "loss": 5.0776,
      "mean_token_accuracy": 0.20154687464237214,
      "num_tokens": 48736621.0,
      "step": 28090
    },
    {
      "entropy": 5.753596067428589,
      "epoch": 2.1858782338066525,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004517100354322835,
      "loss": 5.0088,
      "mean_token_accuracy": 0.20686496943235397,
      "num_tokens": 48744990.0,
      "step": 28095
    },
    {
      "entropy": 5.707619857788086,
      "epoch": 2.18626726317837,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045169287768402497,
      "loss": 5.0353,
      "mean_token_accuracy": 0.19160938858985901,
      "num_tokens": 48753640.0,
      "step": 28100
    },
    {
      "entropy": 5.63163046836853,
      "epoch": 2.1866562925500874,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004516757172547331,
      "loss": 4.9263,
      "mean_token_accuracy": 0.2099306106567383,
      "num_tokens": 48762836.0,
      "step": 28105
    },
    {
      "entropy": 5.714001417160034,
      "epoch": 2.187045321921805,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004516585541446685,
      "loss": 5.0141,
      "mean_token_accuracy": 0.20325562208890915,
      "num_tokens": 48771158.0,
      "step": 28110
    },
    {
      "entropy": 5.703816699981689,
      "epoch": 2.187434351293523,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045164138835409153,
      "loss": 5.0733,
      "mean_token_accuracy": 0.1949499160051346,
      "num_tokens": 48779631.0,
      "step": 28115
    },
    {
      "entropy": 5.707851552963257,
      "epoch": 2.18782338066524,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045162421988326266,
      "loss": 5.0607,
      "mean_token_accuracy": 0.19756405502557756,
      "num_tokens": 48788584.0,
      "step": 28120
    },
    {
      "entropy": 5.695174694061279,
      "epoch": 2.1882124100369578,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004516070487324424,
      "loss": 4.9391,
      "mean_token_accuracy": 0.20184656381607055,
      "num_tokens": 48797403.0,
      "step": 28125
    },
    {
      "entropy": 5.671128988265991,
      "epoch": 2.1886014394086755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045158987490189124,
      "loss": 5.0085,
      "mean_token_accuracy": 0.21112131625413894,
      "num_tokens": 48805448.0,
      "step": 28130
    },
    {
      "entropy": 5.72662091255188,
      "epoch": 2.1889904687803927,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004515726983918697,
      "loss": 5.0436,
      "mean_token_accuracy": 0.20126300156116486,
      "num_tokens": 48814983.0,
      "step": 28135
    },
    {
      "entropy": 5.7558988571167,
      "epoch": 2.1893794981521104,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045155551920263864,
      "loss": 4.994,
      "mean_token_accuracy": 0.2037679061293602,
      "num_tokens": 48823955.0,
      "step": 28140
    },
    {
      "entropy": 5.750918388366699,
      "epoch": 2.189768527523828,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045153833733445857,
      "loss": 5.0369,
      "mean_token_accuracy": 0.20174161344766617,
      "num_tokens": 48832078.0,
      "step": 28145
    },
    {
      "entropy": 5.750604772567749,
      "epoch": 2.190157556895546,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004515211527875903,
      "loss": 4.9704,
      "mean_token_accuracy": 0.20717132836580276,
      "num_tokens": 48840351.0,
      "step": 28150
    },
    {
      "entropy": 5.687026166915894,
      "epoch": 2.190546586267263,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004515039655622943,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21429229229688646,
      "num_tokens": 48848333.0,
      "step": 28155
    },
    {
      "entropy": 5.663086938858032,
      "epoch": 2.1909356156389808,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045148677565883167,
      "loss": 4.9948,
      "mean_token_accuracy": 0.19683879017829894,
      "num_tokens": 48856806.0,
      "step": 28160
    },
    {
      "entropy": 5.741447591781617,
      "epoch": 2.1913246450106985,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004514695830774631,
      "loss": 5.0501,
      "mean_token_accuracy": 0.20508329272270204,
      "num_tokens": 48865848.0,
      "step": 28165
    },
    {
      "entropy": 5.7666182041168215,
      "epoch": 2.1917136743824157,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045145238781844953,
      "loss": 5.0658,
      "mean_token_accuracy": 0.19703116565942763,
      "num_tokens": 48875899.0,
      "step": 28170
    },
    {
      "entropy": 5.718249273300171,
      "epoch": 2.1921027037541334,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045143518988205184,
      "loss": 4.9535,
      "mean_token_accuracy": 0.20782748609781265,
      "num_tokens": 48884026.0,
      "step": 28175
    },
    {
      "entropy": 5.720352125167847,
      "epoch": 2.192491733125851,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045141798926853094,
      "loss": 4.9976,
      "mean_token_accuracy": 0.19865641593933106,
      "num_tokens": 48892640.0,
      "step": 28180
    },
    {
      "entropy": 5.786416959762573,
      "epoch": 2.1928807624975684,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004514007859781478,
      "loss": 5.0596,
      "mean_token_accuracy": 0.19939381927251815,
      "num_tokens": 48901071.0,
      "step": 28185
    },
    {
      "entropy": 5.745792627334595,
      "epoch": 2.193269791869286,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004513835800111635,
      "loss": 4.91,
      "mean_token_accuracy": 0.2157277673482895,
      "num_tokens": 48909034.0,
      "step": 28190
    },
    {
      "entropy": 5.753616666793823,
      "epoch": 2.1936588212410038,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004513663713678392,
      "loss": 5.1,
      "mean_token_accuracy": 0.19856774657964707,
      "num_tokens": 48918399.0,
      "step": 28195
    },
    {
      "entropy": 5.7065513134002686,
      "epoch": 2.1940478506127215,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045134916004843596,
      "loss": 5.0057,
      "mean_token_accuracy": 0.20225861966609954,
      "num_tokens": 48926767.0,
      "step": 28200
    },
    {
      "entropy": 5.722287178039551,
      "epoch": 2.1944368799844387,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004513319460532148,
      "loss": 4.9429,
      "mean_token_accuracy": 0.2059488609433174,
      "num_tokens": 48935666.0,
      "step": 28205
    },
    {
      "entropy": 5.70296893119812,
      "epoch": 2.1948259093561564,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045131472938243707,
      "loss": 5.0093,
      "mean_token_accuracy": 0.20246360450983047,
      "num_tokens": 48944461.0,
      "step": 28210
    },
    {
      "entropy": 5.6407942295074465,
      "epoch": 2.195214938727874,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004512975100363639,
      "loss": 4.9325,
      "mean_token_accuracy": 0.20925680547952652,
      "num_tokens": 48952368.0,
      "step": 28215
    },
    {
      "entropy": 5.574014472961426,
      "epoch": 2.1956039680995914,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045128028801525675,
      "loss": 4.8126,
      "mean_token_accuracy": 0.2130169078707695,
      "num_tokens": 48961866.0,
      "step": 28220
    },
    {
      "entropy": 5.7087925434112545,
      "epoch": 2.195992997471309,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045126306331937674,
      "loss": 4.9766,
      "mean_token_accuracy": 0.20847883224487304,
      "num_tokens": 48970401.0,
      "step": 28225
    },
    {
      "entropy": 5.733750867843628,
      "epoch": 2.1963820268430267,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004512458359489853,
      "loss": 5.0313,
      "mean_token_accuracy": 0.19230654537677766,
      "num_tokens": 48979165.0,
      "step": 28230
    },
    {
      "entropy": 5.69598126411438,
      "epoch": 2.196771056214744,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004512286059043437,
      "loss": 4.912,
      "mean_token_accuracy": 0.21594143509864808,
      "num_tokens": 48987700.0,
      "step": 28235
    },
    {
      "entropy": 5.699603271484375,
      "epoch": 2.1971600855864617,
      "grad_norm": 1.125,
      "learning_rate": 0.00045121137318571364,
      "loss": 4.9985,
      "mean_token_accuracy": 0.20668129920959472,
      "num_tokens": 48996901.0,
      "step": 28240
    },
    {
      "entropy": 5.73549108505249,
      "epoch": 2.1975491149581794,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045119413779335645,
      "loss": 5.0988,
      "mean_token_accuracy": 0.19888806641101836,
      "num_tokens": 49006049.0,
      "step": 28245
    },
    {
      "entropy": 5.784724473953247,
      "epoch": 2.197938144329897,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004511768997275336,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20857109725475312,
      "num_tokens": 49015092.0,
      "step": 28250
    },
    {
      "entropy": 5.706155490875244,
      "epoch": 2.1983271737016143,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004511596589885068,
      "loss": 5.0057,
      "mean_token_accuracy": 0.19969862848520278,
      "num_tokens": 49023388.0,
      "step": 28255
    },
    {
      "entropy": 5.6270498752594,
      "epoch": 2.198716203073332,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004511424155765375,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21338020116090775,
      "num_tokens": 49031931.0,
      "step": 28260
    },
    {
      "entropy": 5.706189966201782,
      "epoch": 2.1991052324450497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045112516949188735,
      "loss": 5.0789,
      "mean_token_accuracy": 0.2013660192489624,
      "num_tokens": 49040584.0,
      "step": 28265
    },
    {
      "entropy": 5.7466577053070065,
      "epoch": 2.199494261816767,
      "grad_norm": 1.125,
      "learning_rate": 0.0004511079207348182,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20657130032777787,
      "num_tokens": 49049107.0,
      "step": 28270
    },
    {
      "entropy": 5.70091495513916,
      "epoch": 2.1998832911884847,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004510906693055916,
      "loss": 4.9596,
      "mean_token_accuracy": 0.20573167651891708,
      "num_tokens": 49058038.0,
      "step": 28275
    },
    {
      "entropy": 5.683161497116089,
      "epoch": 2.2002723205602024,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004510734152044694,
      "loss": 4.9544,
      "mean_token_accuracy": 0.2072075441479683,
      "num_tokens": 49066875.0,
      "step": 28280
    },
    {
      "entropy": 5.707600831985474,
      "epoch": 2.20066134993192,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045105615843171336,
      "loss": 5.0312,
      "mean_token_accuracy": 0.19821940511465072,
      "num_tokens": 49076374.0,
      "step": 28285
    },
    {
      "entropy": 5.710311985015869,
      "epoch": 2.2010503793036373,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045103889898758534,
      "loss": 4.9815,
      "mean_token_accuracy": 0.20235689878463745,
      "num_tokens": 49085729.0,
      "step": 28290
    },
    {
      "entropy": 5.665670394897461,
      "epoch": 2.201439408675355,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004510216368723472,
      "loss": 4.911,
      "mean_token_accuracy": 0.20671570748090745,
      "num_tokens": 49093959.0,
      "step": 28295
    },
    {
      "entropy": 5.669644165039062,
      "epoch": 2.2018284380470727,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045100437208626086,
      "loss": 4.9941,
      "mean_token_accuracy": 0.2049276500940323,
      "num_tokens": 49102412.0,
      "step": 28300
    },
    {
      "entropy": 5.637655258178711,
      "epoch": 2.20221746741879,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045098710462958844,
      "loss": 4.9802,
      "mean_token_accuracy": 0.20358238220214844,
      "num_tokens": 49110655.0,
      "step": 28305
    },
    {
      "entropy": 5.730246829986572,
      "epoch": 2.2026064967905077,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045096983450259166,
      "loss": 4.9931,
      "mean_token_accuracy": 0.20257960855960847,
      "num_tokens": 49119117.0,
      "step": 28310
    },
    {
      "entropy": 5.679993104934693,
      "epoch": 2.2029955261622254,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045095256170553285,
      "loss": 4.923,
      "mean_token_accuracy": 0.20516852736473085,
      "num_tokens": 49127513.0,
      "step": 28315
    },
    {
      "entropy": 5.684822845458984,
      "epoch": 2.2033845555339426,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045093528623867384,
      "loss": 4.9368,
      "mean_token_accuracy": 0.20608424246311188,
      "num_tokens": 49135851.0,
      "step": 28320
    },
    {
      "entropy": 5.7045563697814945,
      "epoch": 2.2037735849056603,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045091800810227706,
      "loss": 5.0663,
      "mean_token_accuracy": 0.19839129745960235,
      "num_tokens": 49144775.0,
      "step": 28325
    },
    {
      "entropy": 5.7332395076751705,
      "epoch": 2.204162614277378,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004509007272966044,
      "loss": 4.9907,
      "mean_token_accuracy": 0.2042761594057083,
      "num_tokens": 49152861.0,
      "step": 28330
    },
    {
      "entropy": 5.719871139526367,
      "epoch": 2.2045516436490953,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004508834438219182,
      "loss": 5.0742,
      "mean_token_accuracy": 0.19557158499956132,
      "num_tokens": 49161875.0,
      "step": 28335
    },
    {
      "entropy": 5.76036262512207,
      "epoch": 2.204940673020813,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004508661576784807,
      "loss": 5.101,
      "mean_token_accuracy": 0.20120255202054976,
      "num_tokens": 49170277.0,
      "step": 28340
    },
    {
      "entropy": 5.69703049659729,
      "epoch": 2.2053297023925307,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004508488688665541,
      "loss": 4.8846,
      "mean_token_accuracy": 0.2119675487279892,
      "num_tokens": 49179078.0,
      "step": 28345
    },
    {
      "entropy": 5.692879104614258,
      "epoch": 2.2057187317642484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004508315773864009,
      "loss": 4.9859,
      "mean_token_accuracy": 0.19961091876029968,
      "num_tokens": 49187649.0,
      "step": 28350
    },
    {
      "entropy": 5.666404390335083,
      "epoch": 2.2061077611359656,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045081428323828337,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21948359906673431,
      "num_tokens": 49195369.0,
      "step": 28355
    },
    {
      "entropy": 5.769012784957885,
      "epoch": 2.2064967905076833,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004507969864224639,
      "loss": 5.0442,
      "mean_token_accuracy": 0.20128490924835205,
      "num_tokens": 49203623.0,
      "step": 28360
    },
    {
      "entropy": 5.723573970794678,
      "epoch": 2.206885819879401,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000450779686939205,
      "loss": 4.9977,
      "mean_token_accuracy": 0.2062707021832466,
      "num_tokens": 49212251.0,
      "step": 28365
    },
    {
      "entropy": 5.721948146820068,
      "epoch": 2.2072748492511183,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045076238478876916,
      "loss": 4.9697,
      "mean_token_accuracy": 0.2032117709517479,
      "num_tokens": 49220580.0,
      "step": 28370
    },
    {
      "entropy": 5.759302806854248,
      "epoch": 2.207663878622836,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004507450799714188,
      "loss": 5.008,
      "mean_token_accuracy": 0.20607277601957322,
      "num_tokens": 49229011.0,
      "step": 28375
    },
    {
      "entropy": 5.692243194580078,
      "epoch": 2.2080529079945537,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045072777248741664,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20925655514001845,
      "num_tokens": 49237082.0,
      "step": 28380
    },
    {
      "entropy": 5.7546450138092045,
      "epoch": 2.2084419373662714,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004507104623370252,
      "loss": 5.0915,
      "mean_token_accuracy": 0.19839746206998826,
      "num_tokens": 49245768.0,
      "step": 28385
    },
    {
      "entropy": 5.699833631515503,
      "epoch": 2.2088309667379886,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004506931495205072,
      "loss": 5.011,
      "mean_token_accuracy": 0.20534223765134813,
      "num_tokens": 49254932.0,
      "step": 28390
    },
    {
      "entropy": 5.6779553413391115,
      "epoch": 2.2092199961097063,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004506758340381253,
      "loss": 4.9982,
      "mean_token_accuracy": 0.19613065272569657,
      "num_tokens": 49263798.0,
      "step": 28395
    },
    {
      "entropy": 5.690609502792358,
      "epoch": 2.209609025481424,
      "grad_norm": 1.25,
      "learning_rate": 0.0004506585158901422,
      "loss": 4.9167,
      "mean_token_accuracy": 0.20441493839025499,
      "num_tokens": 49272045.0,
      "step": 28400
    },
    {
      "entropy": 5.76332836151123,
      "epoch": 2.2099980548531413,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004506411950768207,
      "loss": 5.1068,
      "mean_token_accuracy": 0.19730678349733352,
      "num_tokens": 49281562.0,
      "step": 28405
    },
    {
      "entropy": 5.666595077514648,
      "epoch": 2.210387084224859,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045062387159842363,
      "loss": 4.8893,
      "mean_token_accuracy": 0.214822655916214,
      "num_tokens": 49290113.0,
      "step": 28410
    },
    {
      "entropy": 5.686538171768189,
      "epoch": 2.2107761135965767,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045060654545521386,
      "loss": 4.9995,
      "mean_token_accuracy": 0.20580074638128282,
      "num_tokens": 49298210.0,
      "step": 28415
    },
    {
      "entropy": 5.730928516387939,
      "epoch": 2.211165142968294,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045058921664745436,
      "loss": 5.0214,
      "mean_token_accuracy": 0.19478271007537842,
      "num_tokens": 49306533.0,
      "step": 28420
    },
    {
      "entropy": 5.723982095718384,
      "epoch": 2.2115541723400116,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045057188517540783,
      "loss": 5.0632,
      "mean_token_accuracy": 0.20542512238025665,
      "num_tokens": 49315617.0,
      "step": 28425
    },
    {
      "entropy": 5.762454557418823,
      "epoch": 2.2119432017117293,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004505545510393374,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20437406599521638,
      "num_tokens": 49324351.0,
      "step": 28430
    },
    {
      "entropy": 5.6805655002594,
      "epoch": 2.2123322310834466,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045053721423950617,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20792602747678757,
      "num_tokens": 49333036.0,
      "step": 28435
    },
    {
      "entropy": 5.689638996124268,
      "epoch": 2.2127212604551643,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004505198747761771,
      "loss": 4.9624,
      "mean_token_accuracy": 0.20511289238929747,
      "num_tokens": 49341715.0,
      "step": 28440
    },
    {
      "entropy": 5.7387665748596195,
      "epoch": 2.213110289826882,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004505025326496132,
      "loss": 5.0206,
      "mean_token_accuracy": 0.20127775222063066,
      "num_tokens": 49350505.0,
      "step": 28445
    },
    {
      "entropy": 5.726810359954834,
      "epoch": 2.2134993191985997,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045048518786007784,
      "loss": 5.0182,
      "mean_token_accuracy": 0.20388639867305755,
      "num_tokens": 49358902.0,
      "step": 28450
    },
    {
      "entropy": 5.752914619445801,
      "epoch": 2.213888348570317,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045046784040783395,
      "loss": 5.0635,
      "mean_token_accuracy": 0.19582430720329286,
      "num_tokens": 49366932.0,
      "step": 28455
    },
    {
      "entropy": 5.797455358505249,
      "epoch": 2.2142773779420346,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000450450490293145,
      "loss": 5.148,
      "mean_token_accuracy": 0.1883062332868576,
      "num_tokens": 49375907.0,
      "step": 28460
    },
    {
      "entropy": 5.788388204574585,
      "epoch": 2.2146664073137523,
      "grad_norm": 1.234375,
      "learning_rate": 0.000450433137516274,
      "loss": 5.0095,
      "mean_token_accuracy": 0.20223766267299653,
      "num_tokens": 49383871.0,
      "step": 28465
    },
    {
      "entropy": 5.705138731002807,
      "epoch": 2.2150554366854696,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004504157820774844,
      "loss": 5.0125,
      "mean_token_accuracy": 0.20448571890592576,
      "num_tokens": 49392452.0,
      "step": 28470
    },
    {
      "entropy": 5.637589311599731,
      "epoch": 2.2154444660571873,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004503984239770396,
      "loss": 4.8766,
      "mean_token_accuracy": 0.2112268775701523,
      "num_tokens": 49401698.0,
      "step": 28475
    },
    {
      "entropy": 5.753566837310791,
      "epoch": 2.215833495428905,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004503810632152028,
      "loss": 5.0406,
      "mean_token_accuracy": 0.19906357675790787,
      "num_tokens": 49409505.0,
      "step": 28480
    },
    {
      "entropy": 5.7391256332397464,
      "epoch": 2.2162225248006227,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004503636997922375,
      "loss": 5.0167,
      "mean_token_accuracy": 0.205886110663414,
      "num_tokens": 49418489.0,
      "step": 28485
    },
    {
      "entropy": 5.729737234115601,
      "epoch": 2.21661155417234,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004503463337084072,
      "loss": 4.9622,
      "mean_token_accuracy": 0.21037158817052842,
      "num_tokens": 49426989.0,
      "step": 28490
    },
    {
      "entropy": 5.719191646575927,
      "epoch": 2.2170005835440576,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004503289649639754,
      "loss": 5.0271,
      "mean_token_accuracy": 0.20467952638864517,
      "num_tokens": 49436199.0,
      "step": 28495
    },
    {
      "entropy": 5.6993835926055905,
      "epoch": 2.2173896129157753,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004503115935592056,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21046911627054216,
      "num_tokens": 49444092.0,
      "step": 28500
    },
    {
      "entropy": 5.7002115726470945,
      "epoch": 2.2177786422874926,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004502942194943614,
      "loss": 5.0302,
      "mean_token_accuracy": 0.19915594756603242,
      "num_tokens": 49452461.0,
      "step": 28505
    },
    {
      "entropy": 5.763059139251709,
      "epoch": 2.2181676716592102,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045027684276970645,
      "loss": 4.9697,
      "mean_token_accuracy": 0.20664460062980652,
      "num_tokens": 49461972.0,
      "step": 28510
    },
    {
      "entropy": 5.7821187496185305,
      "epoch": 2.218556701030928,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045025946338550434,
      "loss": 4.9718,
      "mean_token_accuracy": 0.2104617714881897,
      "num_tokens": 49470656.0,
      "step": 28515
    },
    {
      "entropy": 5.686243963241577,
      "epoch": 2.218945730402645,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004502420813420189,
      "loss": 5.0059,
      "mean_token_accuracy": 0.2045851856470108,
      "num_tokens": 49479018.0,
      "step": 28520
    },
    {
      "entropy": 5.708141899108886,
      "epoch": 2.219334759774363,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004502246966395137,
      "loss": 5.0004,
      "mean_token_accuracy": 0.20668516755104066,
      "num_tokens": 49487521.0,
      "step": 28525
    },
    {
      "entropy": 5.752575063705445,
      "epoch": 2.2197237891460806,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045020730927825274,
      "loss": 5.0786,
      "mean_token_accuracy": 0.1960342913866043,
      "num_tokens": 49495382.0,
      "step": 28530
    },
    {
      "entropy": 5.693490219116211,
      "epoch": 2.2201128185177983,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004501899192584997,
      "loss": 4.9569,
      "mean_token_accuracy": 0.20873293429613113,
      "num_tokens": 49503819.0,
      "step": 28535
    },
    {
      "entropy": 5.711993551254272,
      "epoch": 2.2205018478895155,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004501725265805185,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20590731650590896,
      "num_tokens": 49512908.0,
      "step": 28540
    },
    {
      "entropy": 5.711372327804566,
      "epoch": 2.2208908772612332,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045015513124457303,
      "loss": 5.0308,
      "mean_token_accuracy": 0.2065579041838646,
      "num_tokens": 49522462.0,
      "step": 28545
    },
    {
      "entropy": 5.742654514312744,
      "epoch": 2.221279906632951,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004501377332509272,
      "loss": 5.0322,
      "mean_token_accuracy": 0.19891941249370576,
      "num_tokens": 49531089.0,
      "step": 28550
    },
    {
      "entropy": 5.740454626083374,
      "epoch": 2.221668936004668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045012033259984494,
      "loss": 4.9535,
      "mean_token_accuracy": 0.20609834492206575,
      "num_tokens": 49539074.0,
      "step": 28555
    },
    {
      "entropy": 5.703121995925903,
      "epoch": 2.222057965376386,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004501029292915905,
      "loss": 5.0244,
      "mean_token_accuracy": 0.20597645938396453,
      "num_tokens": 49547879.0,
      "step": 28560
    },
    {
      "entropy": 5.810239505767822,
      "epoch": 2.2224469947481036,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045008552332642774,
      "loss": 5.0712,
      "mean_token_accuracy": 0.20533390939235688,
      "num_tokens": 49556220.0,
      "step": 28565
    },
    {
      "entropy": 5.759454965591431,
      "epoch": 2.222836024119821,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045006811470462083,
      "loss": 5.0022,
      "mean_token_accuracy": 0.2028620034456253,
      "num_tokens": 49564710.0,
      "step": 28570
    },
    {
      "entropy": 5.683412075042725,
      "epoch": 2.2232250534915385,
      "grad_norm": 1.28125,
      "learning_rate": 0.000450050703426434,
      "loss": 4.9053,
      "mean_token_accuracy": 0.21151494532823562,
      "num_tokens": 49572829.0,
      "step": 28575
    },
    {
      "entropy": 5.703780269622802,
      "epoch": 2.2236140828632562,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004500332894921313,
      "loss": 5.0632,
      "mean_token_accuracy": 0.20818754583597182,
      "num_tokens": 49580355.0,
      "step": 28580
    },
    {
      "entropy": 5.733652544021607,
      "epoch": 2.224003112234974,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004500158729019771,
      "loss": 4.992,
      "mean_token_accuracy": 0.20801398754119874,
      "num_tokens": 49589086.0,
      "step": 28585
    },
    {
      "entropy": 5.679855918884277,
      "epoch": 2.224392141606691,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004499984536562355,
      "loss": 4.9219,
      "mean_token_accuracy": 0.21132203042507172,
      "num_tokens": 49597434.0,
      "step": 28590
    },
    {
      "entropy": 5.752693748474121,
      "epoch": 2.224781170978409,
      "grad_norm": 1.296875,
      "learning_rate": 0.000449981031755171,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20213069021701813,
      "num_tokens": 49604981.0,
      "step": 28595
    },
    {
      "entropy": 5.7699274063110355,
      "epoch": 2.2251702003501266,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044996360719904777,
      "loss": 5.0197,
      "mean_token_accuracy": 0.20493629723787307,
      "num_tokens": 49614119.0,
      "step": 28600
    },
    {
      "entropy": 5.811347579956054,
      "epoch": 2.225559229721844,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004499461799881303,
      "loss": 5.0629,
      "mean_token_accuracy": 0.19947040528059007,
      "num_tokens": 49622705.0,
      "step": 28605
    },
    {
      "entropy": 5.765024900436401,
      "epoch": 2.2259482590935615,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004499287501226831,
      "loss": 5.0481,
      "mean_token_accuracy": 0.1937385618686676,
      "num_tokens": 49632086.0,
      "step": 28610
    },
    {
      "entropy": 5.7021410942077635,
      "epoch": 2.2263372884652792,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044991131760297045,
      "loss": 5.0169,
      "mean_token_accuracy": 0.20657098591327666,
      "num_tokens": 49640733.0,
      "step": 28615
    },
    {
      "entropy": 5.662859153747559,
      "epoch": 2.226726317836997,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044989388242925693,
      "loss": 4.9835,
      "mean_token_accuracy": 0.2068522498011589,
      "num_tokens": 49649694.0,
      "step": 28620
    },
    {
      "entropy": 5.759501314163208,
      "epoch": 2.227115347208714,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004498764446018073,
      "loss": 5.0928,
      "mean_token_accuracy": 0.1993888333439827,
      "num_tokens": 49657536.0,
      "step": 28625
    },
    {
      "entropy": 5.721528434753418,
      "epoch": 2.227504376580432,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004498590041208857,
      "loss": 4.9448,
      "mean_token_accuracy": 0.21212560981512069,
      "num_tokens": 49666332.0,
      "step": 28630
    },
    {
      "entropy": 5.6651215076446535,
      "epoch": 2.2278934059521496,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004498415609867572,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21172105073928832,
      "num_tokens": 49675364.0,
      "step": 28635
    },
    {
      "entropy": 5.74646954536438,
      "epoch": 2.228282435323867,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044982411519968625,
      "loss": 5.1076,
      "mean_token_accuracy": 0.19643208682537078,
      "num_tokens": 49684178.0,
      "step": 28640
    },
    {
      "entropy": 5.738535356521607,
      "epoch": 2.2286714646955845,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004498066667599377,
      "loss": 5.0375,
      "mean_token_accuracy": 0.20278588235378264,
      "num_tokens": 49692747.0,
      "step": 28645
    },
    {
      "entropy": 5.70613055229187,
      "epoch": 2.2290604940673022,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044978921566777604,
      "loss": 5.012,
      "mean_token_accuracy": 0.20278516560792922,
      "num_tokens": 49701534.0,
      "step": 28650
    },
    {
      "entropy": 5.766454887390137,
      "epoch": 2.2294495234390195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004497717619234664,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20247657895088195,
      "num_tokens": 49710021.0,
      "step": 28655
    },
    {
      "entropy": 5.715042400360107,
      "epoch": 2.229838552810737,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044975430552727337,
      "loss": 4.9374,
      "mean_token_accuracy": 0.2064614027738571,
      "num_tokens": 49718786.0,
      "step": 28660
    },
    {
      "entropy": 5.722701358795166,
      "epoch": 2.230227582182455,
      "grad_norm": 1.25,
      "learning_rate": 0.0004497368464794619,
      "loss": 4.9885,
      "mean_token_accuracy": 0.20379282683134078,
      "num_tokens": 49726840.0,
      "step": 28665
    },
    {
      "entropy": 5.764046335220337,
      "epoch": 2.230616611554172,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044971938478029693,
      "loss": 5.0809,
      "mean_token_accuracy": 0.19563679099082948,
      "num_tokens": 49736130.0,
      "step": 28670
    },
    {
      "entropy": 5.7257531642913815,
      "epoch": 2.23100564092589,
      "grad_norm": 1.125,
      "learning_rate": 0.00044970192043004343,
      "loss": 4.9814,
      "mean_token_accuracy": 0.2002241626381874,
      "num_tokens": 49744810.0,
      "step": 28675
    },
    {
      "entropy": 5.703462314605713,
      "epoch": 2.2313946702976075,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004496844534289663,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20451734960079193,
      "num_tokens": 49754053.0,
      "step": 28680
    },
    {
      "entropy": 5.729618787765503,
      "epoch": 2.231783699669325,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044966698377733064,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20244460552930832,
      "num_tokens": 49763735.0,
      "step": 28685
    },
    {
      "entropy": 5.752537679672241,
      "epoch": 2.2321727290410425,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004496495114754015,
      "loss": 5.0605,
      "mean_token_accuracy": 0.1975702166557312,
      "num_tokens": 49772867.0,
      "step": 28690
    },
    {
      "entropy": 5.664232444763184,
      "epoch": 2.23256175841276,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044963203652344404,
      "loss": 4.8923,
      "mean_token_accuracy": 0.2080998107790947,
      "num_tokens": 49781460.0,
      "step": 28695
    },
    {
      "entropy": 5.690932321548462,
      "epoch": 2.232950787784478,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004496145589217234,
      "loss": 4.9251,
      "mean_token_accuracy": 0.2038187339901924,
      "num_tokens": 49790249.0,
      "step": 28700
    },
    {
      "entropy": 5.754150819778443,
      "epoch": 2.233339817156195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044959707867050467,
      "loss": 4.9633,
      "mean_token_accuracy": 0.2096121594309807,
      "num_tokens": 49797959.0,
      "step": 28705
    },
    {
      "entropy": 5.665742111206055,
      "epoch": 2.233728846527913,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004495795957700532,
      "loss": 4.9149,
      "mean_token_accuracy": 0.21175573170185089,
      "num_tokens": 49806669.0,
      "step": 28710
    },
    {
      "entropy": 5.700140428543091,
      "epoch": 2.2341178758996305,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044956211022063424,
      "loss": 5.1118,
      "mean_token_accuracy": 0.19611927717924119,
      "num_tokens": 49816821.0,
      "step": 28715
    },
    {
      "entropy": 5.793018198013305,
      "epoch": 2.234506905271348,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044954462202251316,
      "loss": 5.0268,
      "mean_token_accuracy": 0.20222463011741637,
      "num_tokens": 49825411.0,
      "step": 28720
    },
    {
      "entropy": 5.777142667770386,
      "epoch": 2.2348959346430655,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044952713117595516,
      "loss": 5.067,
      "mean_token_accuracy": 0.1973838210105896,
      "num_tokens": 49833496.0,
      "step": 28725
    },
    {
      "entropy": 5.677167558670044,
      "epoch": 2.235284964014783,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044950963768122576,
      "loss": 4.9816,
      "mean_token_accuracy": 0.1960351884365082,
      "num_tokens": 49842935.0,
      "step": 28730
    },
    {
      "entropy": 5.7087047576904295,
      "epoch": 2.235673993386501,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004494921415385905,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19957512021064758,
      "num_tokens": 49851197.0,
      "step": 28735
    },
    {
      "entropy": 5.728284597396851,
      "epoch": 2.236063022758218,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044947464274831464,
      "loss": 4.9295,
      "mean_token_accuracy": 0.2125383883714676,
      "num_tokens": 49859778.0,
      "step": 28740
    },
    {
      "entropy": 5.746982288360596,
      "epoch": 2.236452052129936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004494571413106637,
      "loss": 5.011,
      "mean_token_accuracy": 0.19913018345832825,
      "num_tokens": 49868548.0,
      "step": 28745
    },
    {
      "entropy": 5.686481094360351,
      "epoch": 2.2368410815016535,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044943963722590344,
      "loss": 4.9866,
      "mean_token_accuracy": 0.20673135221004485,
      "num_tokens": 49877070.0,
      "step": 28750
    },
    {
      "entropy": 5.729356861114502,
      "epoch": 2.2372301108733708,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044942213049429917,
      "loss": 5.0348,
      "mean_token_accuracy": 0.2036292001605034,
      "num_tokens": 49885225.0,
      "step": 28755
    },
    {
      "entropy": 5.716859245300293,
      "epoch": 2.2376191402450885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044940462111611685,
      "loss": 5.1345,
      "mean_token_accuracy": 0.195256832242012,
      "num_tokens": 49894305.0,
      "step": 28760
    },
    {
      "entropy": 5.699057769775391,
      "epoch": 2.238008169616806,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044938710909162195,
      "loss": 4.9476,
      "mean_token_accuracy": 0.20902130603790284,
      "num_tokens": 49902587.0,
      "step": 28765
    },
    {
      "entropy": 5.736685562133789,
      "epoch": 2.2383971989885234,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004493695944210802,
      "loss": 4.9132,
      "mean_token_accuracy": 0.20681415498256683,
      "num_tokens": 49911037.0,
      "step": 28770
    },
    {
      "entropy": 5.706594657897949,
      "epoch": 2.238786228360241,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004493520771047574,
      "loss": 5.0336,
      "mean_token_accuracy": 0.20331882685422897,
      "num_tokens": 49920724.0,
      "step": 28775
    },
    {
      "entropy": 5.674307584762573,
      "epoch": 2.239175257731959,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044933455714291944,
      "loss": 5.0107,
      "mean_token_accuracy": 0.20807262808084487,
      "num_tokens": 49929173.0,
      "step": 28780
    },
    {
      "entropy": 5.7317914962768555,
      "epoch": 2.2395642871036765,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044931703453583195,
      "loss": 4.989,
      "mean_token_accuracy": 0.20164856910705567,
      "num_tokens": 49938213.0,
      "step": 28785
    },
    {
      "entropy": 5.684723901748657,
      "epoch": 2.2399533164753938,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044929950928376095,
      "loss": 4.8903,
      "mean_token_accuracy": 0.20966372787952423,
      "num_tokens": 49946493.0,
      "step": 28790
    },
    {
      "entropy": 5.659772300720215,
      "epoch": 2.2403423458471114,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004492819813869723,
      "loss": 4.9704,
      "mean_token_accuracy": 0.2069219946861267,
      "num_tokens": 49954730.0,
      "step": 28795
    },
    {
      "entropy": 5.654062509536743,
      "epoch": 2.240731375218829,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044926445084573206,
      "loss": 4.9227,
      "mean_token_accuracy": 0.21151854395866393,
      "num_tokens": 49962651.0,
      "step": 28800
    },
    {
      "entropy": 5.694990158081055,
      "epoch": 2.2411204045905464,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000449246917660306,
      "loss": 4.9979,
      "mean_token_accuracy": 0.19839437454938888,
      "num_tokens": 49972560.0,
      "step": 28805
    },
    {
      "entropy": 5.7244185447692875,
      "epoch": 2.241509433962264,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004492293818309604,
      "loss": 5.0354,
      "mean_token_accuracy": 0.20150577127933503,
      "num_tokens": 49982292.0,
      "step": 28810
    },
    {
      "entropy": 5.734969711303711,
      "epoch": 2.241898463333982,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004492118433579613,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19014789760112763,
      "num_tokens": 49990395.0,
      "step": 28815
    },
    {
      "entropy": 5.743406915664673,
      "epoch": 2.2422874927056995,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044919430224157463,
      "loss": 5.0429,
      "mean_token_accuracy": 0.20124853402376175,
      "num_tokens": 49998944.0,
      "step": 28820
    },
    {
      "entropy": 5.710884761810303,
      "epoch": 2.2426765220774167,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044917675848206684,
      "loss": 4.9355,
      "mean_token_accuracy": 0.2041032686829567,
      "num_tokens": 50007583.0,
      "step": 28825
    },
    {
      "entropy": 5.623904132843018,
      "epoch": 2.2430655514491344,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044915921207970387,
      "loss": 4.8607,
      "mean_token_accuracy": 0.215935055911541,
      "num_tokens": 50015854.0,
      "step": 28830
    },
    {
      "entropy": 5.624081134796143,
      "epoch": 2.243454580820852,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044914166303475206,
      "loss": 4.8808,
      "mean_token_accuracy": 0.22171128988265992,
      "num_tokens": 50024757.0,
      "step": 28835
    },
    {
      "entropy": 5.716112041473389,
      "epoch": 2.2438436101925694,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044912411134747764,
      "loss": 4.9676,
      "mean_token_accuracy": 0.20674169957637786,
      "num_tokens": 50034592.0,
      "step": 28840
    },
    {
      "entropy": 5.677056169509887,
      "epoch": 2.244232639564287,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044910655701814705,
      "loss": 4.976,
      "mean_token_accuracy": 0.2038874328136444,
      "num_tokens": 50043034.0,
      "step": 28845
    },
    {
      "entropy": 5.656762266159058,
      "epoch": 2.244621668936005,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004490890000470266,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20454513728618623,
      "num_tokens": 50051683.0,
      "step": 28850
    },
    {
      "entropy": 5.692839813232422,
      "epoch": 2.245010698307722,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044907144043438265,
      "loss": 4.997,
      "mean_token_accuracy": 0.19817980825901033,
      "num_tokens": 50060257.0,
      "step": 28855
    },
    {
      "entropy": 5.6413181781768795,
      "epoch": 2.2453997276794397,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004490538781804816,
      "loss": 4.9086,
      "mean_token_accuracy": 0.21037443578243256,
      "num_tokens": 50068188.0,
      "step": 28860
    },
    {
      "entropy": 5.750301170349121,
      "epoch": 2.2457887570511574,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044903631328559,
      "loss": 4.9734,
      "mean_token_accuracy": 0.1981503650546074,
      "num_tokens": 50076188.0,
      "step": 28865
    },
    {
      "entropy": 5.736948871612549,
      "epoch": 2.246177786422875,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044901874574997444,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21094398498535155,
      "num_tokens": 50085139.0,
      "step": 28870
    },
    {
      "entropy": 5.777136611938476,
      "epoch": 2.2465668157945924,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044900117557390137,
      "loss": 5.0655,
      "mean_token_accuracy": 0.19944897145032883,
      "num_tokens": 50093050.0,
      "step": 28875
    },
    {
      "entropy": 5.697117137908935,
      "epoch": 2.24695584516631,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044898360275763733,
      "loss": 5.0247,
      "mean_token_accuracy": 0.2058744251728058,
      "num_tokens": 50101854.0,
      "step": 28880
    },
    {
      "entropy": 5.731565999984741,
      "epoch": 2.247344874538028,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044896602730144905,
      "loss": 5.0567,
      "mean_token_accuracy": 0.19855786710977555,
      "num_tokens": 50110430.0,
      "step": 28885
    },
    {
      "entropy": 5.7700201034545895,
      "epoch": 2.247733903909745,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044894844920560323,
      "loss": 4.9866,
      "mean_token_accuracy": 0.2036704733967781,
      "num_tokens": 50118633.0,
      "step": 28890
    },
    {
      "entropy": 5.7869304656982425,
      "epoch": 2.2481229332814627,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004489308684703666,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19673241227865218,
      "num_tokens": 50129120.0,
      "step": 28895
    },
    {
      "entropy": 5.794103670120239,
      "epoch": 2.2485119626531804,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004489132850960059,
      "loss": 5.059,
      "mean_token_accuracy": 0.1987464502453804,
      "num_tokens": 50138404.0,
      "step": 28900
    },
    {
      "entropy": 5.70320692062378,
      "epoch": 2.2489009920248977,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004488956990827878,
      "loss": 4.9843,
      "mean_token_accuracy": 0.19734846651554108,
      "num_tokens": 50146908.0,
      "step": 28905
    },
    {
      "entropy": 5.654918479919433,
      "epoch": 2.2492900213966154,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004488781104309793,
      "loss": 4.9197,
      "mean_token_accuracy": 0.20398542881011963,
      "num_tokens": 50155922.0,
      "step": 28910
    },
    {
      "entropy": 5.695118093490601,
      "epoch": 2.249679050768333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044886051914084726,
      "loss": 5.0247,
      "mean_token_accuracy": 0.20175406187772751,
      "num_tokens": 50164023.0,
      "step": 28915
    },
    {
      "entropy": 5.685731744766235,
      "epoch": 2.2500680801400508,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044884292521265846,
      "loss": 4.9534,
      "mean_token_accuracy": 0.2051523059606552,
      "num_tokens": 50172147.0,
      "step": 28920
    },
    {
      "entropy": 5.724393224716186,
      "epoch": 2.250457109511768,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004488253286466801,
      "loss": 5.0773,
      "mean_token_accuracy": 0.1939166396856308,
      "num_tokens": 50180484.0,
      "step": 28925
    },
    {
      "entropy": 5.6866292476654055,
      "epoch": 2.2508461388834857,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044880772944317905,
      "loss": 4.9135,
      "mean_token_accuracy": 0.20873091518878936,
      "num_tokens": 50189176.0,
      "step": 28930
    },
    {
      "entropy": 5.7000938892364506,
      "epoch": 2.2512351682552034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044879012760242224,
      "loss": 4.9686,
      "mean_token_accuracy": 0.2151930496096611,
      "num_tokens": 50197674.0,
      "step": 28935
    },
    {
      "entropy": 5.733453559875488,
      "epoch": 2.2516241976269207,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044877252312467694,
      "loss": 5.0394,
      "mean_token_accuracy": 0.19731226116418837,
      "num_tokens": 50206681.0,
      "step": 28940
    },
    {
      "entropy": 5.79929404258728,
      "epoch": 2.2520132269986384,
      "grad_norm": 1.125,
      "learning_rate": 0.0004487549160102101,
      "loss": 5.0872,
      "mean_token_accuracy": 0.19281459152698516,
      "num_tokens": 50215457.0,
      "step": 28945
    },
    {
      "entropy": 5.739561223983765,
      "epoch": 2.252402256370356,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044873730625928914,
      "loss": 4.9618,
      "mean_token_accuracy": 0.20617651641368867,
      "num_tokens": 50224757.0,
      "step": 28950
    },
    {
      "entropy": 5.742098951339722,
      "epoch": 2.2527912857420738,
      "grad_norm": 1.125,
      "learning_rate": 0.00044871969387218094,
      "loss": 5.0264,
      "mean_token_accuracy": 0.1989448457956314,
      "num_tokens": 50233570.0,
      "step": 28955
    },
    {
      "entropy": 5.762804079055786,
      "epoch": 2.253180315113791,
      "grad_norm": 1.28125,
      "learning_rate": 0.000448702078849153,
      "loss": 4.9902,
      "mean_token_accuracy": 0.208285653591156,
      "num_tokens": 50241767.0,
      "step": 28960
    },
    {
      "entropy": 5.6715192794799805,
      "epoch": 2.2535693444855087,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044868446119047234,
      "loss": 4.9356,
      "mean_token_accuracy": 0.21125833839178085,
      "num_tokens": 50250121.0,
      "step": 28965
    },
    {
      "entropy": 5.789412879943848,
      "epoch": 2.253958373857226,
      "grad_norm": 1.125,
      "learning_rate": 0.0004486668408964065,
      "loss": 5.0867,
      "mean_token_accuracy": 0.20002398937940596,
      "num_tokens": 50258717.0,
      "step": 28970
    },
    {
      "entropy": 5.73927149772644,
      "epoch": 2.2543474032289437,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004486492179672228,
      "loss": 5.028,
      "mean_token_accuracy": 0.20437908470630645,
      "num_tokens": 50267528.0,
      "step": 28975
    },
    {
      "entropy": 5.7655116558074955,
      "epoch": 2.2547364326006614,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044863159240318863,
      "loss": 5.0201,
      "mean_token_accuracy": 0.20582814067602156,
      "num_tokens": 50276197.0,
      "step": 28980
    },
    {
      "entropy": 5.646577930450439,
      "epoch": 2.255125461972379,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004486139642045714,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21113918125629424,
      "num_tokens": 50284222.0,
      "step": 28985
    },
    {
      "entropy": 5.6721209526062015,
      "epoch": 2.2555144913440963,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004485963333716385,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20272922068834304,
      "num_tokens": 50293061.0,
      "step": 28990
    },
    {
      "entropy": 5.722727870941162,
      "epoch": 2.255903520715814,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004485786999046576,
      "loss": 4.9773,
      "mean_token_accuracy": 0.21097562462091446,
      "num_tokens": 50301601.0,
      "step": 28995
    },
    {
      "entropy": 5.71388521194458,
      "epoch": 2.2562925500875317,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044856106380389624,
      "loss": 5.02,
      "mean_token_accuracy": 0.20650473535060881,
      "num_tokens": 50311595.0,
      "step": 29000
    },
    {
      "entropy": 5.672596120834351,
      "epoch": 2.256681579459249,
      "grad_norm": 1.15625,
      "learning_rate": 0.000448543425069622,
      "loss": 4.9655,
      "mean_token_accuracy": 0.2122400313615799,
      "num_tokens": 50320205.0,
      "step": 29005
    },
    {
      "entropy": 5.711330938339233,
      "epoch": 2.2570706088309667,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004485257837021025,
      "loss": 4.9736,
      "mean_token_accuracy": 0.20538078099489213,
      "num_tokens": 50328414.0,
      "step": 29010
    },
    {
      "entropy": 5.698928356170654,
      "epoch": 2.2574596382026844,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044850813970160534,
      "loss": 4.9802,
      "mean_token_accuracy": 0.2082694411277771,
      "num_tokens": 50337206.0,
      "step": 29015
    },
    {
      "entropy": 5.74770040512085,
      "epoch": 2.257848667574402,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004484904930683984,
      "loss": 5.0223,
      "mean_token_accuracy": 0.2034577801823616,
      "num_tokens": 50346320.0,
      "step": 29020
    },
    {
      "entropy": 5.725022172927856,
      "epoch": 2.2582376969461193,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004484728438027494,
      "loss": 5.0068,
      "mean_token_accuracy": 0.20166103094816207,
      "num_tokens": 50354199.0,
      "step": 29025
    },
    {
      "entropy": 5.7522049903869625,
      "epoch": 2.258626726317837,
      "grad_norm": 1.25,
      "learning_rate": 0.0004484551919049261,
      "loss": 5.0081,
      "mean_token_accuracy": 0.20749075412750245,
      "num_tokens": 50363085.0,
      "step": 29030
    },
    {
      "entropy": 5.709029006958008,
      "epoch": 2.2590157556895547,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044843753737519625,
      "loss": 4.9745,
      "mean_token_accuracy": 0.20265224426984788,
      "num_tokens": 50371211.0,
      "step": 29035
    },
    {
      "entropy": 5.758967494964599,
      "epoch": 2.259404785061272,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004484198802138279,
      "loss": 4.9647,
      "mean_token_accuracy": 0.21568350046873092,
      "num_tokens": 50379426.0,
      "step": 29040
    },
    {
      "entropy": 5.730614089965821,
      "epoch": 2.2597938144329897,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004484022204210889,
      "loss": 5.05,
      "mean_token_accuracy": 0.20107064843177797,
      "num_tokens": 50388073.0,
      "step": 29045
    },
    {
      "entropy": 5.6381267547607425,
      "epoch": 2.2601828438047074,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044838455799724717,
      "loss": 4.9449,
      "mean_token_accuracy": 0.20804987400770186,
      "num_tokens": 50395943.0,
      "step": 29050
    },
    {
      "entropy": 5.697162771224976,
      "epoch": 2.260571873176425,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044836689294257075,
      "loss": 4.9746,
      "mean_token_accuracy": 0.20222398042678832,
      "num_tokens": 50405218.0,
      "step": 29055
    },
    {
      "entropy": 5.690928840637207,
      "epoch": 2.2609609025481423,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004483492252573276,
      "loss": 4.959,
      "mean_token_accuracy": 0.20453712046146394,
      "num_tokens": 50414169.0,
      "step": 29060
    },
    {
      "entropy": 5.705320739746094,
      "epoch": 2.26134993191986,
      "grad_norm": 1.34375,
      "learning_rate": 0.000448331554941786,
      "loss": 5.0489,
      "mean_token_accuracy": 0.20291474759578704,
      "num_tokens": 50422541.0,
      "step": 29065
    },
    {
      "entropy": 5.6856060981750485,
      "epoch": 2.2617389612915777,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044831388199621385,
      "loss": 4.9293,
      "mean_token_accuracy": 0.20846335738897323,
      "num_tokens": 50431081.0,
      "step": 29070
    },
    {
      "entropy": 5.749835681915283,
      "epoch": 2.262127990663295,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044829620642087946,
      "loss": 5.0113,
      "mean_token_accuracy": 0.20569914877414702,
      "num_tokens": 50439279.0,
      "step": 29075
    },
    {
      "entropy": 5.710979175567627,
      "epoch": 2.2625170200350126,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004482785282160509,
      "loss": 5.0075,
      "mean_token_accuracy": 0.20251786410808564,
      "num_tokens": 50447628.0,
      "step": 29080
    },
    {
      "entropy": 5.758240270614624,
      "epoch": 2.2629060494067303,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044826084738199657,
      "loss": 5.0312,
      "mean_token_accuracy": 0.20596242249011992,
      "num_tokens": 50457109.0,
      "step": 29085
    },
    {
      "entropy": 5.747400236129761,
      "epoch": 2.2632950787784476,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044824316391898464,
      "loss": 4.9288,
      "mean_token_accuracy": 0.20979313552379608,
      "num_tokens": 50466222.0,
      "step": 29090
    },
    {
      "entropy": 5.699603080749512,
      "epoch": 2.2636841081501653,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044822547782728334,
      "loss": 4.9709,
      "mean_token_accuracy": 0.20732593685388565,
      "num_tokens": 50475382.0,
      "step": 29095
    },
    {
      "entropy": 5.738552188873291,
      "epoch": 2.264073137521883,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004482077891071612,
      "loss": 5.0106,
      "mean_token_accuracy": 0.2064552828669548,
      "num_tokens": 50483501.0,
      "step": 29100
    },
    {
      "entropy": 5.724460458755493,
      "epoch": 2.2644621668936002,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044819009775888655,
      "loss": 4.9765,
      "mean_token_accuracy": 0.2058560535311699,
      "num_tokens": 50492005.0,
      "step": 29105
    },
    {
      "entropy": 5.671643400192261,
      "epoch": 2.264851196265318,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044817240378272784,
      "loss": 4.8923,
      "mean_token_accuracy": 0.2137363910675049,
      "num_tokens": 50500578.0,
      "step": 29110
    },
    {
      "entropy": 5.656417560577393,
      "epoch": 2.2652402256370356,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044815470717895345,
      "loss": 4.9223,
      "mean_token_accuracy": 0.2054235652089119,
      "num_tokens": 50509285.0,
      "step": 29115
    },
    {
      "entropy": 5.627827262878418,
      "epoch": 2.2656292550087533,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004481370079478321,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20197763293981552,
      "num_tokens": 50517519.0,
      "step": 29120
    },
    {
      "entropy": 5.690816831588745,
      "epoch": 2.2660182843804706,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004481193060896322,
      "loss": 4.9736,
      "mean_token_accuracy": 0.20609022229909896,
      "num_tokens": 50525843.0,
      "step": 29125
    },
    {
      "entropy": 5.729027223587036,
      "epoch": 2.2664073137521883,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004481016016046223,
      "loss": 5.0126,
      "mean_token_accuracy": 0.20402844548225402,
      "num_tokens": 50535638.0,
      "step": 29130
    },
    {
      "entropy": 5.757827663421631,
      "epoch": 2.266796343123906,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004480838944930712,
      "loss": 5.052,
      "mean_token_accuracy": 0.19842476397752762,
      "num_tokens": 50544327.0,
      "step": 29135
    },
    {
      "entropy": 5.729465484619141,
      "epoch": 2.2671853724956232,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004480661847552474,
      "loss": 4.9881,
      "mean_token_accuracy": 0.20816373229026794,
      "num_tokens": 50552276.0,
      "step": 29140
    },
    {
      "entropy": 5.687768363952637,
      "epoch": 2.267574401867341,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004480484723914198,
      "loss": 4.9229,
      "mean_token_accuracy": 0.21032319217920303,
      "num_tokens": 50561247.0,
      "step": 29145
    },
    {
      "entropy": 5.768047332763672,
      "epoch": 2.2679634312390586,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004480307574018571,
      "loss": 4.9809,
      "mean_token_accuracy": 0.20695863366127015,
      "num_tokens": 50569401.0,
      "step": 29150
    },
    {
      "entropy": 5.74258828163147,
      "epoch": 2.2683524606107763,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000448013039786828,
      "loss": 4.9518,
      "mean_token_accuracy": 0.20893010795116423,
      "num_tokens": 50577346.0,
      "step": 29155
    },
    {
      "entropy": 5.666473817825318,
      "epoch": 2.2687414899824936,
      "grad_norm": 1.25,
      "learning_rate": 0.00044799531954660133,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20209113508462906,
      "num_tokens": 50586370.0,
      "step": 29160
    },
    {
      "entropy": 5.696482849121094,
      "epoch": 2.2691305193542113,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004479775966814461,
      "loss": 4.9023,
      "mean_token_accuracy": 0.20927826762199403,
      "num_tokens": 50595326.0,
      "step": 29165
    },
    {
      "entropy": 5.789342641830444,
      "epoch": 2.269519548725929,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044795987119163115,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20709120631217956,
      "num_tokens": 50603513.0,
      "step": 29170
    },
    {
      "entropy": 5.691622352600097,
      "epoch": 2.2699085780976462,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004479421430774255,
      "loss": 4.9629,
      "mean_token_accuracy": 0.20148243606090546,
      "num_tokens": 50612416.0,
      "step": 29175
    },
    {
      "entropy": 5.700114631652832,
      "epoch": 2.270297607469364,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004479244123390981,
      "loss": 4.9796,
      "mean_token_accuracy": 0.21193940341472625,
      "num_tokens": 50620480.0,
      "step": 29180
    },
    {
      "entropy": 5.6114232540130615,
      "epoch": 2.2706866368410816,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044790667897691796,
      "loss": 4.8798,
      "mean_token_accuracy": 0.21626315414905548,
      "num_tokens": 50629044.0,
      "step": 29185
    },
    {
      "entropy": 5.7674500942230225,
      "epoch": 2.271075666212799,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044788894299115415,
      "loss": 5.0905,
      "mean_token_accuracy": 0.1974001407623291,
      "num_tokens": 50638610.0,
      "step": 29190
    },
    {
      "entropy": 5.6896758556365965,
      "epoch": 2.2714646955845166,
      "grad_norm": 1.125,
      "learning_rate": 0.0004478712043820758,
      "loss": 4.9791,
      "mean_token_accuracy": 0.20331458747386932,
      "num_tokens": 50647337.0,
      "step": 29195
    },
    {
      "entropy": 5.715017509460449,
      "epoch": 2.2718537249562343,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004478534631499521,
      "loss": 4.9901,
      "mean_token_accuracy": 0.2104761630296707,
      "num_tokens": 50654641.0,
      "step": 29200
    },
    {
      "entropy": 5.693295240402222,
      "epoch": 2.2722427543279515,
      "grad_norm": 1.125,
      "learning_rate": 0.00044783571929505223,
      "loss": 5.0299,
      "mean_token_accuracy": 0.19273147732019424,
      "num_tokens": 50663726.0,
      "step": 29205
    },
    {
      "entropy": 5.7791399478912355,
      "epoch": 2.2726317836996692,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004478179728176454,
      "loss": 5.0626,
      "mean_token_accuracy": 0.20494945645332335,
      "num_tokens": 50672376.0,
      "step": 29210
    },
    {
      "entropy": 5.755568408966065,
      "epoch": 2.273020813071387,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004478002237180009,
      "loss": 5.0366,
      "mean_token_accuracy": 0.20128738582134248,
      "num_tokens": 50680990.0,
      "step": 29215
    },
    {
      "entropy": 5.7578778743743895,
      "epoch": 2.2734098424431046,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000447782471996388,
      "loss": 5.0916,
      "mean_token_accuracy": 0.19580189883708954,
      "num_tokens": 50690180.0,
      "step": 29220
    },
    {
      "entropy": 5.746267080307007,
      "epoch": 2.273798871814822,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004477647176530762,
      "loss": 4.9737,
      "mean_token_accuracy": 0.2056620389223099,
      "num_tokens": 50699094.0,
      "step": 29225
    },
    {
      "entropy": 5.749994802474975,
      "epoch": 2.2741879011865396,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004477469606883347,
      "loss": 4.9964,
      "mean_token_accuracy": 0.20125244557857513,
      "num_tokens": 50708312.0,
      "step": 29230
    },
    {
      "entropy": 5.652478361129761,
      "epoch": 2.2745769305582573,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004477292011024331,
      "loss": 4.9742,
      "mean_token_accuracy": 0.20667684972286224,
      "num_tokens": 50717092.0,
      "step": 29235
    },
    {
      "entropy": 5.643951654434204,
      "epoch": 2.2749659599299745,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044771143889564075,
      "loss": 4.9548,
      "mean_token_accuracy": 0.2088184505701065,
      "num_tokens": 50725641.0,
      "step": 29240
    },
    {
      "entropy": 5.667259597778321,
      "epoch": 2.275354989301692,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044769367406822727,
      "loss": 4.9773,
      "mean_token_accuracy": 0.2048446223139763,
      "num_tokens": 50734082.0,
      "step": 29245
    },
    {
      "entropy": 5.738952589035034,
      "epoch": 2.27574401867341,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004476759066204621,
      "loss": 5.0687,
      "mean_token_accuracy": 0.19482304155826569,
      "num_tokens": 50743242.0,
      "step": 29250
    },
    {
      "entropy": 5.707887601852417,
      "epoch": 2.2761330480451276,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004476581365526149,
      "loss": 4.94,
      "mean_token_accuracy": 0.20732298046350478,
      "num_tokens": 50751693.0,
      "step": 29255
    },
    {
      "entropy": 5.712618207931518,
      "epoch": 2.276522077416845,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004476403638649553,
      "loss": 5.0045,
      "mean_token_accuracy": 0.1977794125676155,
      "num_tokens": 50760031.0,
      "step": 29260
    },
    {
      "entropy": 5.691950702667237,
      "epoch": 2.2769111067885626,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044762258855775294,
      "loss": 5.0149,
      "mean_token_accuracy": 0.20737670809030534,
      "num_tokens": 50768236.0,
      "step": 29265
    },
    {
      "entropy": 5.716222953796387,
      "epoch": 2.2773001361602803,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044760481063127754,
      "loss": 4.977,
      "mean_token_accuracy": 0.20951962918043138,
      "num_tokens": 50777055.0,
      "step": 29270
    },
    {
      "entropy": 5.738576745986938,
      "epoch": 2.2776891655319975,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000447587030085799,
      "loss": 5.0198,
      "mean_token_accuracy": 0.202374866604805,
      "num_tokens": 50785574.0,
      "step": 29275
    },
    {
      "entropy": 5.674203252792358,
      "epoch": 2.278078194903715,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044756924692158684,
      "loss": 5.0211,
      "mean_token_accuracy": 0.20544813424348832,
      "num_tokens": 50794889.0,
      "step": 29280
    },
    {
      "entropy": 5.6562903881072994,
      "epoch": 2.278467224275433,
      "grad_norm": 1.25,
      "learning_rate": 0.0004475514611389111,
      "loss": 5.0356,
      "mean_token_accuracy": 0.20114805102348327,
      "num_tokens": 50804823.0,
      "step": 29285
    },
    {
      "entropy": 5.746180677413941,
      "epoch": 2.2788562536471506,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004475336727380415,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21025109589099883,
      "num_tokens": 50813233.0,
      "step": 29290
    },
    {
      "entropy": 5.707292222976685,
      "epoch": 2.279245283018868,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004475158817192481,
      "loss": 4.9559,
      "mean_token_accuracy": 0.20495827347040177,
      "num_tokens": 50821925.0,
      "step": 29295
    },
    {
      "entropy": 5.714855813980103,
      "epoch": 2.2796343123905856,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044749808808280077,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20511671155691147,
      "num_tokens": 50831337.0,
      "step": 29300
    },
    {
      "entropy": 5.716977310180664,
      "epoch": 2.280023341762303,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044748029182896956,
      "loss": 4.9751,
      "mean_token_accuracy": 0.20896781831979752,
      "num_tokens": 50840493.0,
      "step": 29305
    },
    {
      "entropy": 5.790254545211792,
      "epoch": 2.2804123711340205,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004474624929580243,
      "loss": 5.0739,
      "mean_token_accuracy": 0.2021462142467499,
      "num_tokens": 50848843.0,
      "step": 29310
    },
    {
      "entropy": 5.74621353149414,
      "epoch": 2.280801400505738,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004474446914702354,
      "loss": 4.9961,
      "mean_token_accuracy": 0.20213095992803573,
      "num_tokens": 50857882.0,
      "step": 29315
    },
    {
      "entropy": 5.764393854141235,
      "epoch": 2.281190429877456,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004474268873658727,
      "loss": 5.1285,
      "mean_token_accuracy": 0.19607437998056412,
      "num_tokens": 50866058.0,
      "step": 29320
    },
    {
      "entropy": 5.726093912124634,
      "epoch": 2.281579459249173,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004474090806452064,
      "loss": 5.0071,
      "mean_token_accuracy": 0.20631835460662842,
      "num_tokens": 50874175.0,
      "step": 29325
    },
    {
      "entropy": 5.6904966831207275,
      "epoch": 2.281968488620891,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044739127130850675,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20511212348937988,
      "num_tokens": 50883753.0,
      "step": 29330
    },
    {
      "entropy": 5.736419534683227,
      "epoch": 2.2823575179926086,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044737345935604397,
      "loss": 5.0493,
      "mean_token_accuracy": 0.20174746066331864,
      "num_tokens": 50891962.0,
      "step": 29335
    },
    {
      "entropy": 5.7512908458709715,
      "epoch": 2.282746547364326,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004473556447880883,
      "loss": 4.9539,
      "mean_token_accuracy": 0.2080543413758278,
      "num_tokens": 50900405.0,
      "step": 29340
    },
    {
      "entropy": 5.734158134460449,
      "epoch": 2.2831355767360435,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044733782760490997,
      "loss": 5.0294,
      "mean_token_accuracy": 0.20564614236354828,
      "num_tokens": 50908850.0,
      "step": 29345
    },
    {
      "entropy": 5.7007527351379395,
      "epoch": 2.283524606107761,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004473200078067795,
      "loss": 5.0793,
      "mean_token_accuracy": 0.1987391158938408,
      "num_tokens": 50918545.0,
      "step": 29350
    },
    {
      "entropy": 5.69084529876709,
      "epoch": 2.283913635479479,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044730218539396717,
      "loss": 4.9923,
      "mean_token_accuracy": 0.20665283203125,
      "num_tokens": 50926664.0,
      "step": 29355
    },
    {
      "entropy": 5.733583688735962,
      "epoch": 2.284302664851196,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004472843603667434,
      "loss": 5.0702,
      "mean_token_accuracy": 0.19152991324663163,
      "num_tokens": 50935846.0,
      "step": 29360
    },
    {
      "entropy": 5.710142326354981,
      "epoch": 2.284691694222914,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044726653272537866,
      "loss": 4.9944,
      "mean_token_accuracy": 0.20574162304401397,
      "num_tokens": 50944289.0,
      "step": 29365
    },
    {
      "entropy": 5.723946666717529,
      "epoch": 2.2850807235946315,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004472487024701435,
      "loss": 4.9985,
      "mean_token_accuracy": 0.20516370832920075,
      "num_tokens": 50952949.0,
      "step": 29370
    },
    {
      "entropy": 5.789242267608643,
      "epoch": 2.285469752966349,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004472308696013085,
      "loss": 5.0842,
      "mean_token_accuracy": 0.1952628791332245,
      "num_tokens": 50961957.0,
      "step": 29375
    },
    {
      "entropy": 5.758067226409912,
      "epoch": 2.2858587823380665,
      "grad_norm": 1.25,
      "learning_rate": 0.0004472130341191441,
      "loss": 5.0002,
      "mean_token_accuracy": 0.20671627819538116,
      "num_tokens": 50970463.0,
      "step": 29380
    },
    {
      "entropy": 5.763094472885132,
      "epoch": 2.286247811709784,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004471951960239211,
      "loss": 5.0331,
      "mean_token_accuracy": 0.19847462326288223,
      "num_tokens": 50979350.0,
      "step": 29385
    },
    {
      "entropy": 5.75760850906372,
      "epoch": 2.286636841081502,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044717735531591,
      "loss": 5.0557,
      "mean_token_accuracy": 0.20189856141805648,
      "num_tokens": 50988048.0,
      "step": 29390
    },
    {
      "entropy": 5.735132884979248,
      "epoch": 2.287025870453219,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044715951199538156,
      "loss": 5.018,
      "mean_token_accuracy": 0.20573174804449082,
      "num_tokens": 50996714.0,
      "step": 29395
    },
    {
      "entropy": 5.7759825706481935,
      "epoch": 2.287414899824937,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044714166606260657,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20537349432706833,
      "num_tokens": 51005557.0,
      "step": 29400
    },
    {
      "entropy": 5.794282674789429,
      "epoch": 2.2878039291966545,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004471238175178559,
      "loss": 4.9453,
      "mean_token_accuracy": 0.20468135178089142,
      "num_tokens": 51014024.0,
      "step": 29405
    },
    {
      "entropy": 5.6963907241821286,
      "epoch": 2.288192958568372,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004471059663614002,
      "loss": 4.916,
      "mean_token_accuracy": 0.200111585855484,
      "num_tokens": 51022211.0,
      "step": 29410
    },
    {
      "entropy": 5.724894428253174,
      "epoch": 2.2885819879400895,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004470881125935103,
      "loss": 5.0737,
      "mean_token_accuracy": 0.20791001617908478,
      "num_tokens": 51030459.0,
      "step": 29415
    },
    {
      "entropy": 5.766465663909912,
      "epoch": 2.288971017311807,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044707025621445735,
      "loss": 5.0548,
      "mean_token_accuracy": 0.20130422413349153,
      "num_tokens": 51038439.0,
      "step": 29420
    },
    {
      "entropy": 5.723401594161987,
      "epoch": 2.2893600466835244,
      "grad_norm": 1.25,
      "learning_rate": 0.00044705239722451194,
      "loss": 4.9479,
      "mean_token_accuracy": 0.20673240274190902,
      "num_tokens": 51047698.0,
      "step": 29425
    },
    {
      "entropy": 5.738009643554688,
      "epoch": 2.289749076055242,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044703453562394544,
      "loss": 4.972,
      "mean_token_accuracy": 0.20906113535165788,
      "num_tokens": 51056211.0,
      "step": 29430
    },
    {
      "entropy": 5.753858947753907,
      "epoch": 2.29013810542696,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004470166714130286,
      "loss": 4.9946,
      "mean_token_accuracy": 0.2026839718222618,
      "num_tokens": 51064321.0,
      "step": 29435
    },
    {
      "entropy": 5.757525491714477,
      "epoch": 2.290527134798677,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044699880459203264,
      "loss": 4.9711,
      "mean_token_accuracy": 0.20630315840244293,
      "num_tokens": 51072537.0,
      "step": 29440
    },
    {
      "entropy": 5.704566812515258,
      "epoch": 2.290916164170395,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044698093516122844,
      "loss": 4.9679,
      "mean_token_accuracy": 0.2087300643324852,
      "num_tokens": 51080873.0,
      "step": 29445
    },
    {
      "entropy": 5.706537199020386,
      "epoch": 2.2913051935421125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004469630631208874,
      "loss": 5.0504,
      "mean_token_accuracy": 0.20100590884685515,
      "num_tokens": 51089462.0,
      "step": 29450
    },
    {
      "entropy": 5.712296056747436,
      "epoch": 2.29169422291383,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004469451884712805,
      "loss": 4.9977,
      "mean_token_accuracy": 0.20375248193740844,
      "num_tokens": 51098918.0,
      "step": 29455
    },
    {
      "entropy": 5.76979284286499,
      "epoch": 2.2920832522855474,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004469273112126791,
      "loss": 5.0828,
      "mean_token_accuracy": 0.20307301580905915,
      "num_tokens": 51107483.0,
      "step": 29460
    },
    {
      "entropy": 5.774715995788574,
      "epoch": 2.292472281657265,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044690943134535445,
      "loss": 5.0629,
      "mean_token_accuracy": 0.19831831604242325,
      "num_tokens": 51116590.0,
      "step": 29465
    },
    {
      "entropy": 5.657496929168701,
      "epoch": 2.292861311028983,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004468915488695777,
      "loss": 4.8832,
      "mean_token_accuracy": 0.20863723158836364,
      "num_tokens": 51125296.0,
      "step": 29470
    },
    {
      "entropy": 5.682510423660278,
      "epoch": 2.2932503404007,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044687366378562036,
      "loss": 4.9793,
      "mean_token_accuracy": 0.20140694677829743,
      "num_tokens": 51133369.0,
      "step": 29475
    },
    {
      "entropy": 5.796029996871948,
      "epoch": 2.293639369772418,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044685577609375373,
      "loss": 5.1368,
      "mean_token_accuracy": 0.20644612163305281,
      "num_tokens": 51142789.0,
      "step": 29480
    },
    {
      "entropy": 5.781443548202515,
      "epoch": 2.2940283991441355,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004468378857942492,
      "loss": 4.99,
      "mean_token_accuracy": 0.1978029727935791,
      "num_tokens": 51151706.0,
      "step": 29485
    },
    {
      "entropy": 5.7891288757324215,
      "epoch": 2.294417428515853,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044681999288737826,
      "loss": 5.0232,
      "mean_token_accuracy": 0.2058563157916069,
      "num_tokens": 51160924.0,
      "step": 29490
    },
    {
      "entropy": 5.719172668457031,
      "epoch": 2.2948064578875704,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044680209737341244,
      "loss": 5.0363,
      "mean_token_accuracy": 0.19839604198932648,
      "num_tokens": 51168871.0,
      "step": 29495
    },
    {
      "entropy": 5.715654468536377,
      "epoch": 2.295195487259288,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004467841992526233,
      "loss": 4.9532,
      "mean_token_accuracy": 0.2061698168516159,
      "num_tokens": 51177621.0,
      "step": 29500
    },
    {
      "entropy": 5.746858406066894,
      "epoch": 2.295584516631006,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044676629852528226,
      "loss": 4.9223,
      "mean_token_accuracy": 0.21218833029270173,
      "num_tokens": 51186004.0,
      "step": 29505
    },
    {
      "entropy": 5.710502052307129,
      "epoch": 2.295973546002723,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044674839519166104,
      "loss": 4.9372,
      "mean_token_accuracy": 0.20821304768323898,
      "num_tokens": 51194609.0,
      "step": 29510
    },
    {
      "entropy": 5.724316596984863,
      "epoch": 2.2963625753744408,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044673048925203136,
      "loss": 5.0593,
      "mean_token_accuracy": 0.2141966089606285,
      "num_tokens": 51203928.0,
      "step": 29515
    },
    {
      "entropy": 5.714412975311279,
      "epoch": 2.2967516047461585,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044671258070666476,
      "loss": 4.9826,
      "mean_token_accuracy": 0.20867157578468323,
      "num_tokens": 51212762.0,
      "step": 29520
    },
    {
      "entropy": 5.761064720153809,
      "epoch": 2.2971406341178757,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004466946695558331,
      "loss": 5.1065,
      "mean_token_accuracy": 0.1934242218732834,
      "num_tokens": 51221064.0,
      "step": 29525
    },
    {
      "entropy": 5.714259958267212,
      "epoch": 2.2975296634895934,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044667675579980817,
      "loss": 4.9821,
      "mean_token_accuracy": 0.20306518971920012,
      "num_tokens": 51229633.0,
      "step": 29530
    },
    {
      "entropy": 5.759859609603882,
      "epoch": 2.297918692861311,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004466588394388616,
      "loss": 5.0328,
      "mean_token_accuracy": 0.20471571981906891,
      "num_tokens": 51239051.0,
      "step": 29535
    },
    {
      "entropy": 5.753395986557007,
      "epoch": 2.2983077222330284,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004466409204732654,
      "loss": 4.9743,
      "mean_token_accuracy": 0.20595187246799468,
      "num_tokens": 51247703.0,
      "step": 29540
    },
    {
      "entropy": 5.714046430587769,
      "epoch": 2.298696751604746,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044662299890329144,
      "loss": 4.9263,
      "mean_token_accuracy": 0.21250727623701096,
      "num_tokens": 51256394.0,
      "step": 29545
    },
    {
      "entropy": 5.7732462882995605,
      "epoch": 2.2990857809764638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004466050747292116,
      "loss": 5.0106,
      "mean_token_accuracy": 0.20575936436653136,
      "num_tokens": 51264181.0,
      "step": 29550
    },
    {
      "entropy": 5.6658360958099365,
      "epoch": 2.2994748103481815,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000446587147951298,
      "loss": 4.9091,
      "mean_token_accuracy": 0.2106197327375412,
      "num_tokens": 51272758.0,
      "step": 29555
    },
    {
      "entropy": 5.739393997192383,
      "epoch": 2.2998638397198987,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004465692185698224,
      "loss": 5.0291,
      "mean_token_accuracy": 0.2024053528904915,
      "num_tokens": 51281681.0,
      "step": 29560
    },
    {
      "entropy": 5.710256624221802,
      "epoch": 2.3002528690916164,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044655128658505717,
      "loss": 5.0158,
      "mean_token_accuracy": 0.2016099736094475,
      "num_tokens": 51290060.0,
      "step": 29565
    },
    {
      "entropy": 5.759611320495606,
      "epoch": 2.300641898463334,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004465333519972741,
      "loss": 4.9513,
      "mean_token_accuracy": 0.20766474753618241,
      "num_tokens": 51298262.0,
      "step": 29570
    },
    {
      "entropy": 5.658017778396607,
      "epoch": 2.3010309278350514,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004465154148067454,
      "loss": 4.8698,
      "mean_token_accuracy": 0.20922183692455293,
      "num_tokens": 51306436.0,
      "step": 29575
    },
    {
      "entropy": 5.686435699462891,
      "epoch": 2.301419957206769,
      "grad_norm": 1.375,
      "learning_rate": 0.00044649747501374336,
      "loss": 4.9623,
      "mean_token_accuracy": 0.20191514045000075,
      "num_tokens": 51315132.0,
      "step": 29580
    },
    {
      "entropy": 5.668139028549194,
      "epoch": 2.3018089865784868,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004464795326185401,
      "loss": 4.9199,
      "mean_token_accuracy": 0.21099537760019302,
      "num_tokens": 51324189.0,
      "step": 29585
    },
    {
      "entropy": 5.6724100589752195,
      "epoch": 2.3021980159502045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004464615876214078,
      "loss": 4.9718,
      "mean_token_accuracy": 0.2052620381116867,
      "num_tokens": 51332680.0,
      "step": 29590
    },
    {
      "entropy": 5.746179246902466,
      "epoch": 2.3025870453219217,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004464436400226188,
      "loss": 5.0253,
      "mean_token_accuracy": 0.2039091095328331,
      "num_tokens": 51342036.0,
      "step": 29595
    },
    {
      "entropy": 5.731257915496826,
      "epoch": 2.3029760746936394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004464256898224455,
      "loss": 4.9661,
      "mean_token_accuracy": 0.19533417373895645,
      "num_tokens": 51350987.0,
      "step": 29600
    },
    {
      "entropy": 5.720395755767822,
      "epoch": 2.303365104065357,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004464077370211602,
      "loss": 4.9621,
      "mean_token_accuracy": 0.2061502829194069,
      "num_tokens": 51359787.0,
      "step": 29605
    },
    {
      "entropy": 5.728523445129395,
      "epoch": 2.3037541334370744,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044638978161903533,
      "loss": 5.0561,
      "mean_token_accuracy": 0.20532746165990828,
      "num_tokens": 51368329.0,
      "step": 29610
    },
    {
      "entropy": 5.719857168197632,
      "epoch": 2.304143162808792,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004463718236163433,
      "loss": 5.0346,
      "mean_token_accuracy": 0.2014698639512062,
      "num_tokens": 51377104.0,
      "step": 29615
    },
    {
      "entropy": 5.751528024673462,
      "epoch": 2.3045321921805098,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044635386301335666,
      "loss": 5.069,
      "mean_token_accuracy": 0.19991710931062698,
      "num_tokens": 51385210.0,
      "step": 29620
    },
    {
      "entropy": 5.761771154403687,
      "epoch": 2.304921221552227,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004463358998103478,
      "loss": 5.0129,
      "mean_token_accuracy": 0.19605372846126556,
      "num_tokens": 51393492.0,
      "step": 29625
    },
    {
      "entropy": 5.663180255889893,
      "epoch": 2.3053102509239447,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004463179340075894,
      "loss": 4.8613,
      "mean_token_accuracy": 0.2150268316268921,
      "num_tokens": 51401953.0,
      "step": 29630
    },
    {
      "entropy": 5.704414415359497,
      "epoch": 2.3056992802956624,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004462999656053541,
      "loss": 5.0549,
      "mean_token_accuracy": 0.19720006138086318,
      "num_tokens": 51410874.0,
      "step": 29635
    },
    {
      "entropy": 5.695713138580322,
      "epoch": 2.3060883096673797,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044628199460391445,
      "loss": 4.9221,
      "mean_token_accuracy": 0.21578446477651597,
      "num_tokens": 51419189.0,
      "step": 29640
    },
    {
      "entropy": 5.766446352005005,
      "epoch": 2.3064773390390974,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044626402100354307,
      "loss": 4.9982,
      "mean_token_accuracy": 0.20555685162544252,
      "num_tokens": 51427970.0,
      "step": 29645
    },
    {
      "entropy": 5.790774965286255,
      "epoch": 2.306866368410815,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004462460448045129,
      "loss": 5.0076,
      "mean_token_accuracy": 0.20453614741563797,
      "num_tokens": 51436069.0,
      "step": 29650
    },
    {
      "entropy": 5.679125356674194,
      "epoch": 2.3072553977825327,
      "grad_norm": 1.25,
      "learning_rate": 0.00044622806600709645,
      "loss": 5.0259,
      "mean_token_accuracy": 0.2024261549115181,
      "num_tokens": 51444424.0,
      "step": 29655
    },
    {
      "entropy": 5.703893947601318,
      "epoch": 2.30764442715425,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004462100846115667,
      "loss": 4.9952,
      "mean_token_accuracy": 0.2071802109479904,
      "num_tokens": 51453095.0,
      "step": 29660
    },
    {
      "entropy": 5.773790693283081,
      "epoch": 2.3080334565259677,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004461921006181964,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21115196496248245,
      "num_tokens": 51461525.0,
      "step": 29665
    },
    {
      "entropy": 5.7106016159057615,
      "epoch": 2.3084224858976854,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004461741140272584,
      "loss": 4.9558,
      "mean_token_accuracy": 0.20782775431871414,
      "num_tokens": 51469918.0,
      "step": 29670
    },
    {
      "entropy": 5.728863000869751,
      "epoch": 2.3088115152694026,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004461561248390257,
      "loss": 5.0548,
      "mean_token_accuracy": 0.20880459249019623,
      "num_tokens": 51478747.0,
      "step": 29675
    },
    {
      "entropy": 5.713484382629394,
      "epoch": 2.3092005446411203,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004461381330537713,
      "loss": 4.9987,
      "mean_token_accuracy": 0.19966049641370773,
      "num_tokens": 51487730.0,
      "step": 29680
    },
    {
      "entropy": 5.789279365539551,
      "epoch": 2.309589574012838,
      "grad_norm": 1.25,
      "learning_rate": 0.000446120138671768,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20064719319343566,
      "num_tokens": 51495672.0,
      "step": 29685
    },
    {
      "entropy": 5.774441957473755,
      "epoch": 2.3099786033845557,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044610214169328913,
      "loss": 4.9599,
      "mean_token_accuracy": 0.2028159260749817,
      "num_tokens": 51503377.0,
      "step": 29690
    },
    {
      "entropy": 5.639253854751587,
      "epoch": 2.310367632756273,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004460841421186075,
      "loss": 4.9143,
      "mean_token_accuracy": 0.21235134452581406,
      "num_tokens": 51511932.0,
      "step": 29695
    },
    {
      "entropy": 5.745395231246948,
      "epoch": 2.3107566621279907,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004460661399479963,
      "loss": 4.9948,
      "mean_token_accuracy": 0.19869115501642226,
      "num_tokens": 51520565.0,
      "step": 29700
    },
    {
      "entropy": 5.683946990966797,
      "epoch": 2.3111456914997084,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044604813518172876,
      "loss": 4.8983,
      "mean_token_accuracy": 0.2108920156955719,
      "num_tokens": 51529778.0,
      "step": 29705
    },
    {
      "entropy": 5.750917053222656,
      "epoch": 2.3115347208714256,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044603012782007796,
      "loss": 5.0363,
      "mean_token_accuracy": 0.20505820959806442,
      "num_tokens": 51539394.0,
      "step": 29710
    },
    {
      "entropy": 5.77166051864624,
      "epoch": 2.3119237502431433,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004460121178633172,
      "loss": 4.9563,
      "mean_token_accuracy": 0.2054141029715538,
      "num_tokens": 51548314.0,
      "step": 29715
    },
    {
      "entropy": 5.654262113571167,
      "epoch": 2.312312779614861,
      "grad_norm": 1.25,
      "learning_rate": 0.00044599410531171986,
      "loss": 4.8967,
      "mean_token_accuracy": 0.21590431034564972,
      "num_tokens": 51556828.0,
      "step": 29720
    },
    {
      "entropy": 5.687180852890014,
      "epoch": 2.3127018089865787,
      "grad_norm": 1.203125,
      "learning_rate": 0.000445976090165559,
      "loss": 5.014,
      "mean_token_accuracy": 0.19985330998897552,
      "num_tokens": 51565031.0,
      "step": 29725
    },
    {
      "entropy": 5.798959827423095,
      "epoch": 2.313090838358296,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004459580724251082,
      "loss": 5.0909,
      "mean_token_accuracy": 0.19339765310287477,
      "num_tokens": 51574218.0,
      "step": 29730
    },
    {
      "entropy": 5.683251619338989,
      "epoch": 2.3134798677300137,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044594005209064064,
      "loss": 4.9725,
      "mean_token_accuracy": 0.21131201684474946,
      "num_tokens": 51583229.0,
      "step": 29735
    },
    {
      "entropy": 5.681477355957031,
      "epoch": 2.313868897101731,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044592202916243003,
      "loss": 4.9586,
      "mean_token_accuracy": 0.2036895826458931,
      "num_tokens": 51591969.0,
      "step": 29740
    },
    {
      "entropy": 5.758718776702881,
      "epoch": 2.3142579264734486,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004459040036407495,
      "loss": 4.9736,
      "mean_token_accuracy": 0.2052181839942932,
      "num_tokens": 51600840.0,
      "step": 29745
    },
    {
      "entropy": 5.653624629974365,
      "epoch": 2.3146469558451663,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004458859755258729,
      "loss": 4.8336,
      "mean_token_accuracy": 0.2199634999036789,
      "num_tokens": 51608750.0,
      "step": 29750
    },
    {
      "entropy": 5.722110891342163,
      "epoch": 2.315035985216884,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044586794481807357,
      "loss": 5.0017,
      "mean_token_accuracy": 0.20519475340843202,
      "num_tokens": 51617079.0,
      "step": 29755
    },
    {
      "entropy": 5.661220693588257,
      "epoch": 2.3154250145886013,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044584991151762506,
      "loss": 5.0089,
      "mean_token_accuracy": 0.2047076478600502,
      "num_tokens": 51625310.0,
      "step": 29760
    },
    {
      "entropy": 5.729980897903443,
      "epoch": 2.315814043960319,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004458318756248011,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20155461728572846,
      "num_tokens": 51634071.0,
      "step": 29765
    },
    {
      "entropy": 5.718720388412476,
      "epoch": 2.3162030733320367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044581383713987547,
      "loss": 4.968,
      "mean_token_accuracy": 0.20319255888462068,
      "num_tokens": 51642761.0,
      "step": 29770
    },
    {
      "entropy": 5.716540384292602,
      "epoch": 2.316592102703754,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004457957960631216,
      "loss": 4.9713,
      "mean_token_accuracy": 0.20801773071289062,
      "num_tokens": 51652160.0,
      "step": 29775
    },
    {
      "entropy": 5.643708753585815,
      "epoch": 2.3169811320754716,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004457777523948134,
      "loss": 4.9263,
      "mean_token_accuracy": 0.20331234484910965,
      "num_tokens": 51661345.0,
      "step": 29780
    },
    {
      "entropy": 5.728304195404053,
      "epoch": 2.3173701614471893,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004457597061352247,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20740967392921447,
      "num_tokens": 51669855.0,
      "step": 29785
    },
    {
      "entropy": 5.782248210906983,
      "epoch": 2.317759190818907,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004457416572846291,
      "loss": 5.0805,
      "mean_token_accuracy": 0.20205772519111634,
      "num_tokens": 51679272.0,
      "step": 29790
    },
    {
      "entropy": 5.712372303009033,
      "epoch": 2.3181482201906243,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004457236058433008,
      "loss": 4.9708,
      "mean_token_accuracy": 0.19845490753650666,
      "num_tokens": 51688383.0,
      "step": 29795
    },
    {
      "entropy": 5.75201096534729,
      "epoch": 2.318537249562342,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044570555181151333,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19853007644414902,
      "num_tokens": 51696996.0,
      "step": 29800
    },
    {
      "entropy": 5.694565296173096,
      "epoch": 2.3189262789340597,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000445687495189541,
      "loss": 5.0474,
      "mean_token_accuracy": 0.19959439486265182,
      "num_tokens": 51705830.0,
      "step": 29805
    },
    {
      "entropy": 5.697658824920654,
      "epoch": 2.319315308305777,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044566943597765745,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20759567767381668,
      "num_tokens": 51715357.0,
      "step": 29810
    },
    {
      "entropy": 5.758824968338013,
      "epoch": 2.3197043376774946,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044565137417613694,
      "loss": 5.0424,
      "mean_token_accuracy": 0.20284349471330643,
      "num_tokens": 51725342.0,
      "step": 29815
    },
    {
      "entropy": 5.808252143859863,
      "epoch": 2.3200933670492123,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004456333097852534,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20676441192626954,
      "num_tokens": 51733788.0,
      "step": 29820
    },
    {
      "entropy": 5.752743291854858,
      "epoch": 2.32048239642093,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044561524280528104,
      "loss": 5.023,
      "mean_token_accuracy": 0.20815668255090714,
      "num_tokens": 51742713.0,
      "step": 29825
    },
    {
      "entropy": 5.795190382003784,
      "epoch": 2.3208714257926473,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044559717323649383,
      "loss": 5.1125,
      "mean_token_accuracy": 0.19606551826000213,
      "num_tokens": 51751482.0,
      "step": 29830
    },
    {
      "entropy": 5.67575306892395,
      "epoch": 2.321260455164365,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004455791010791662,
      "loss": 4.8639,
      "mean_token_accuracy": 0.21459861993789672,
      "num_tokens": 51760217.0,
      "step": 29835
    },
    {
      "entropy": 5.744908666610717,
      "epoch": 2.3216494845360827,
      "grad_norm": 1.25,
      "learning_rate": 0.0004455610263335721,
      "loss": 5.0148,
      "mean_token_accuracy": 0.20066949129104614,
      "num_tokens": 51767945.0,
      "step": 29840
    },
    {
      "entropy": 5.7202118873596195,
      "epoch": 2.3220385139078,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004455429489999859,
      "loss": 5.0376,
      "mean_token_accuracy": 0.19415888488292693,
      "num_tokens": 51777318.0,
      "step": 29845
    },
    {
      "entropy": 5.720464563369751,
      "epoch": 2.3224275432795176,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044552486907868194,
      "loss": 4.9895,
      "mean_token_accuracy": 0.1989004895091057,
      "num_tokens": 51786571.0,
      "step": 29850
    },
    {
      "entropy": 5.81879563331604,
      "epoch": 2.3228165726512353,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044550678656993454,
      "loss": 5.0793,
      "mean_token_accuracy": 0.1992500200867653,
      "num_tokens": 51794872.0,
      "step": 29855
    },
    {
      "entropy": 5.801921749114991,
      "epoch": 2.3232056020229526,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000445488701474018,
      "loss": 5.0064,
      "mean_token_accuracy": 0.20138911306858062,
      "num_tokens": 51803111.0,
      "step": 29860
    },
    {
      "entropy": 5.709191560745239,
      "epoch": 2.3235946313946703,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004454706137912067,
      "loss": 4.9104,
      "mean_token_accuracy": 0.20999972075223922,
      "num_tokens": 51811574.0,
      "step": 29865
    },
    {
      "entropy": 5.692693710327148,
      "epoch": 2.323983660766388,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004454525235217753,
      "loss": 4.917,
      "mean_token_accuracy": 0.20232735127210616,
      "num_tokens": 51820120.0,
      "step": 29870
    },
    {
      "entropy": 5.753451633453369,
      "epoch": 2.324372690138105,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044543443066599807,
      "loss": 5.0701,
      "mean_token_accuracy": 0.20338790565729142,
      "num_tokens": 51828293.0,
      "step": 29875
    },
    {
      "entropy": 5.780474519729614,
      "epoch": 2.324761719509823,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004454163352241498,
      "loss": 4.9598,
      "mean_token_accuracy": 0.2057456523180008,
      "num_tokens": 51836595.0,
      "step": 29880
    },
    {
      "entropy": 5.782578039169311,
      "epoch": 2.3251507488815406,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044539823719650463,
      "loss": 5.024,
      "mean_token_accuracy": 0.19959788620471955,
      "num_tokens": 51845415.0,
      "step": 29885
    },
    {
      "entropy": 5.74521689414978,
      "epoch": 2.3255397782532583,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004453801365833376,
      "loss": 4.9883,
      "mean_token_accuracy": 0.206538125872612,
      "num_tokens": 51854091.0,
      "step": 29890
    },
    {
      "entropy": 5.698307847976684,
      "epoch": 2.3259288076249756,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044536203338492317,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20884074717760087,
      "num_tokens": 51862767.0,
      "step": 29895
    },
    {
      "entropy": 5.678468990325928,
      "epoch": 2.3263178369966933,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044534392760153596,
      "loss": 4.8944,
      "mean_token_accuracy": 0.21100373417139054,
      "num_tokens": 51871593.0,
      "step": 29900
    },
    {
      "entropy": 5.770551633834839,
      "epoch": 2.326706866368411,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004453258192334508,
      "loss": 4.9891,
      "mean_token_accuracy": 0.20914898365736007,
      "num_tokens": 51880775.0,
      "step": 29905
    },
    {
      "entropy": 5.70711236000061,
      "epoch": 2.327095895740128,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004453077082809425,
      "loss": 4.997,
      "mean_token_accuracy": 0.20321535021066667,
      "num_tokens": 51889663.0,
      "step": 29910
    },
    {
      "entropy": 5.745099020004273,
      "epoch": 2.327484925111846,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044528959474428575,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20376275330781937,
      "num_tokens": 51898565.0,
      "step": 29915
    },
    {
      "entropy": 5.707531881332398,
      "epoch": 2.3278739544835636,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004452714786237555,
      "loss": 4.9847,
      "mean_token_accuracy": 0.20590702146291734,
      "num_tokens": 51906828.0,
      "step": 29920
    },
    {
      "entropy": 5.7044360637664795,
      "epoch": 2.3282629838552813,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004452533599196265,
      "loss": 4.943,
      "mean_token_accuracy": 0.20438348799943923,
      "num_tokens": 51914305.0,
      "step": 29925
    },
    {
      "entropy": 5.762250661849976,
      "epoch": 2.3286520132269986,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044523523863217374,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20471137166023254,
      "num_tokens": 51923544.0,
      "step": 29930
    },
    {
      "entropy": 5.7381110191345215,
      "epoch": 2.3290410425987162,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004452171147616723,
      "loss": 4.9658,
      "mean_token_accuracy": 0.20559497326612472,
      "num_tokens": 51932198.0,
      "step": 29935
    },
    {
      "entropy": 5.6668556213378904,
      "epoch": 2.329430071970434,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000445198988308397,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21267364919185638,
      "num_tokens": 51940291.0,
      "step": 29940
    },
    {
      "entropy": 5.664438152313233,
      "epoch": 2.329819101342151,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044518085927262293,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2096809044480324,
      "num_tokens": 51948899.0,
      "step": 29945
    },
    {
      "entropy": 5.714698743820191,
      "epoch": 2.330208130713869,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004451627276546252,
      "loss": 4.9387,
      "mean_token_accuracy": 0.20590226948261262,
      "num_tokens": 51957726.0,
      "step": 29950
    },
    {
      "entropy": 5.767014837265014,
      "epoch": 2.3305971600855866,
      "grad_norm": 1.28125,
      "learning_rate": 0.000445144593454679,
      "loss": 4.9825,
      "mean_token_accuracy": 0.20608766078948976,
      "num_tokens": 51965922.0,
      "step": 29955
    },
    {
      "entropy": 5.720744800567627,
      "epoch": 2.330986189457304,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004451264566730593,
      "loss": 4.919,
      "mean_token_accuracy": 0.20810355246067047,
      "num_tokens": 51974096.0,
      "step": 29960
    },
    {
      "entropy": 5.7252710342407225,
      "epoch": 2.3313752188290215,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044510831731004146,
      "loss": 4.9941,
      "mean_token_accuracy": 0.19764285981655122,
      "num_tokens": 51982954.0,
      "step": 29965
    },
    {
      "entropy": 5.684654808044433,
      "epoch": 2.3317642482007392,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004450901753659007,
      "loss": 4.948,
      "mean_token_accuracy": 0.21018617004156112,
      "num_tokens": 51991738.0,
      "step": 29970
    },
    {
      "entropy": 5.741859579086304,
      "epoch": 2.3321532775724565,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044507203084091215,
      "loss": 4.9554,
      "mean_token_accuracy": 0.20965650528669358,
      "num_tokens": 52000227.0,
      "step": 29975
    },
    {
      "entropy": 5.8212151527404785,
      "epoch": 2.332542306944174,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004450538837353513,
      "loss": 5.0956,
      "mean_token_accuracy": 0.19478950798511505,
      "num_tokens": 52009336.0,
      "step": 29980
    },
    {
      "entropy": 5.724214744567871,
      "epoch": 2.332931336315892,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044503573404949343,
      "loss": 5.0147,
      "mean_token_accuracy": 0.2076914578676224,
      "num_tokens": 52017850.0,
      "step": 29985
    },
    {
      "entropy": 5.754072856903076,
      "epoch": 2.3333203656876096,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004450175817836139,
      "loss": 5.0008,
      "mean_token_accuracy": 0.2020827680826187,
      "num_tokens": 52027672.0,
      "step": 29990
    },
    {
      "entropy": 5.74042387008667,
      "epoch": 2.333709395059327,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044499942693798823,
      "loss": 5.0081,
      "mean_token_accuracy": 0.19772714972496033,
      "num_tokens": 52036134.0,
      "step": 29995
    },
    {
      "entropy": 5.726049900054932,
      "epoch": 2.3340984244310445,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004449812695128918,
      "loss": 4.9945,
      "mean_token_accuracy": 0.20394964069128035,
      "num_tokens": 52044719.0,
      "step": 30000
    },
    {
      "epoch": 2.3340984244310445,
      "eval_entropy": 5.485113825444652,
      "eval_loss": 5.081615447998047,
      "eval_mean_token_accuracy": 0.21106990087020844,
      "eval_num_tokens": 52044719.0,
      "eval_runtime": 21.6427,
      "eval_samples_per_second": 1920.185,
      "eval_steps_per_second": 240.035,
      "step": 30000
    },
    {
      "entropy": 5.730231857299804,
      "epoch": 2.3344874538027622,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044496310950860015,
      "loss": 5.0483,
      "mean_token_accuracy": 0.20220550894737244,
      "num_tokens": 52053238.0,
      "step": 30005
    },
    {
      "entropy": 5.755889797210694,
      "epoch": 2.3348764831744795,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044494494692538886,
      "loss": 5.0434,
      "mean_token_accuracy": 0.20430008471012115,
      "num_tokens": 52062258.0,
      "step": 30010
    },
    {
      "entropy": 5.764511919021606,
      "epoch": 2.335265512546197,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044492678176353347,
      "loss": 5.0289,
      "mean_token_accuracy": 0.20457670837640762,
      "num_tokens": 52070527.0,
      "step": 30015
    },
    {
      "entropy": 5.70976676940918,
      "epoch": 2.335654541917915,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044490861402330967,
      "loss": 5.0517,
      "mean_token_accuracy": 0.20189506113529204,
      "num_tokens": 52079827.0,
      "step": 30020
    },
    {
      "entropy": 5.7340700149536135,
      "epoch": 2.3360435712896326,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004448904437049931,
      "loss": 5.0021,
      "mean_token_accuracy": 0.20535786151885987,
      "num_tokens": 52088963.0,
      "step": 30025
    },
    {
      "entropy": 5.741357040405274,
      "epoch": 2.33643260066135,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004448722708088594,
      "loss": 4.9832,
      "mean_token_accuracy": 0.20150717347860336,
      "num_tokens": 52096885.0,
      "step": 30030
    },
    {
      "entropy": 5.805893468856811,
      "epoch": 2.3368216300330675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004448540953351844,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20869221687316894,
      "num_tokens": 52105530.0,
      "step": 30035
    },
    {
      "entropy": 5.693532180786133,
      "epoch": 2.3372106594047852,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004448359172842439,
      "loss": 4.9286,
      "mean_token_accuracy": 0.2123520165681839,
      "num_tokens": 52114233.0,
      "step": 30040
    },
    {
      "entropy": 5.7057770729064945,
      "epoch": 2.3375996887765025,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044481773665631355,
      "loss": 5.0108,
      "mean_token_accuracy": 0.20500598400831221,
      "num_tokens": 52123561.0,
      "step": 30045
    },
    {
      "entropy": 5.681247329711914,
      "epoch": 2.33798871814822,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004447995534516695,
      "loss": 4.9355,
      "mean_token_accuracy": 0.2107487216591835,
      "num_tokens": 52132329.0,
      "step": 30050
    },
    {
      "entropy": 5.747380113601684,
      "epoch": 2.338377747519938,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044478136767058733,
      "loss": 4.9879,
      "mean_token_accuracy": 0.2071058362722397,
      "num_tokens": 52141019.0,
      "step": 30055
    },
    {
      "entropy": 5.704432106018066,
      "epoch": 2.338766776891655,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004447631793133432,
      "loss": 5.0404,
      "mean_token_accuracy": 0.20373893678188323,
      "num_tokens": 52149865.0,
      "step": 30060
    },
    {
      "entropy": 5.744853925704956,
      "epoch": 2.339155806263373,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004447449883802131,
      "loss": 5.04,
      "mean_token_accuracy": 0.19684580117464065,
      "num_tokens": 52158991.0,
      "step": 30065
    },
    {
      "entropy": 5.671135187149048,
      "epoch": 2.3395448356350905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044472679487147295,
      "loss": 4.9365,
      "mean_token_accuracy": 0.2057758867740631,
      "num_tokens": 52167141.0,
      "step": 30070
    },
    {
      "entropy": 5.724252605438233,
      "epoch": 2.339933865006808,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044470859878739877,
      "loss": 4.9659,
      "mean_token_accuracy": 0.20847930908203124,
      "num_tokens": 52175483.0,
      "step": 30075
    },
    {
      "entropy": 5.707490301132202,
      "epoch": 2.3403228943785255,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044469040012826676,
      "loss": 4.9409,
      "mean_token_accuracy": 0.20540456622838973,
      "num_tokens": 52184290.0,
      "step": 30080
    },
    {
      "entropy": 5.689335870742798,
      "epoch": 2.340711923750243,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044467219889435304,
      "loss": 4.9341,
      "mean_token_accuracy": 0.2038714110851288,
      "num_tokens": 52193747.0,
      "step": 30085
    },
    {
      "entropy": 5.703516912460327,
      "epoch": 2.341100953121961,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004446539950859337,
      "loss": 5.0163,
      "mean_token_accuracy": 0.20252951830625535,
      "num_tokens": 52202744.0,
      "step": 30090
    },
    {
      "entropy": 5.707350063323974,
      "epoch": 2.341489982493678,
      "grad_norm": 1.25,
      "learning_rate": 0.00044463578870328506,
      "loss": 4.9669,
      "mean_token_accuracy": 0.20770437866449357,
      "num_tokens": 52211257.0,
      "step": 30095
    },
    {
      "entropy": 5.782588291168213,
      "epoch": 2.341879011865396,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004446175797466834,
      "loss": 4.9919,
      "mean_token_accuracy": 0.2027645528316498,
      "num_tokens": 52219797.0,
      "step": 30100
    },
    {
      "entropy": 5.767563152313232,
      "epoch": 2.3422680412371135,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044459936821640485,
      "loss": 5.0886,
      "mean_token_accuracy": 0.19037699252367019,
      "num_tokens": 52228157.0,
      "step": 30105
    },
    {
      "entropy": 5.7159055233001705,
      "epoch": 2.3426570706088308,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004445811541127258,
      "loss": 5.0369,
      "mean_token_accuracy": 0.20837035179138183,
      "num_tokens": 52237136.0,
      "step": 30110
    },
    {
      "entropy": 5.740798759460449,
      "epoch": 2.3430460999805485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044456293743592274,
      "loss": 5.0059,
      "mean_token_accuracy": 0.20688659846782684,
      "num_tokens": 52245966.0,
      "step": 30115
    },
    {
      "entropy": 5.770652437210083,
      "epoch": 2.343435129352266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044454471818627196,
      "loss": 5.0116,
      "mean_token_accuracy": 0.19893572628498077,
      "num_tokens": 52254782.0,
      "step": 30120
    },
    {
      "entropy": 5.763385438919068,
      "epoch": 2.343824158723984,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044452649636404995,
      "loss": 5.07,
      "mean_token_accuracy": 0.19890822172164918,
      "num_tokens": 52264132.0,
      "step": 30125
    },
    {
      "entropy": 5.683999347686767,
      "epoch": 2.344213188095701,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004445082719695332,
      "loss": 4.8966,
      "mean_token_accuracy": 0.20838618725538255,
      "num_tokens": 52272763.0,
      "step": 30130
    },
    {
      "entropy": 5.741971492767334,
      "epoch": 2.344602217467419,
      "grad_norm": 1.25,
      "learning_rate": 0.00044449004500299815,
      "loss": 5.0382,
      "mean_token_accuracy": 0.20356519222259523,
      "num_tokens": 52281505.0,
      "step": 30135
    },
    {
      "entropy": 5.719330930709839,
      "epoch": 2.3449912468391365,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044447181546472146,
      "loss": 4.8855,
      "mean_token_accuracy": 0.20684665590524673,
      "num_tokens": 52289559.0,
      "step": 30140
    },
    {
      "entropy": 5.685330152511597,
      "epoch": 2.3453802762108538,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044445358335497975,
      "loss": 4.9759,
      "mean_token_accuracy": 0.20889951139688492,
      "num_tokens": 52298318.0,
      "step": 30145
    },
    {
      "entropy": 5.6751243591308596,
      "epoch": 2.3457693055825715,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004444353486740496,
      "loss": 4.9676,
      "mean_token_accuracy": 0.20808221846818925,
      "num_tokens": 52307908.0,
      "step": 30150
    },
    {
      "entropy": 5.774218940734864,
      "epoch": 2.346158334954289,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004444171114222078,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20700594633817673,
      "num_tokens": 52315879.0,
      "step": 30155
    },
    {
      "entropy": 5.724532747268677,
      "epoch": 2.346547364326007,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044439887159973096,
      "loss": 5.0082,
      "mean_token_accuracy": 0.1991189658641815,
      "num_tokens": 52324747.0,
      "step": 30160
    },
    {
      "entropy": 5.733247995376587,
      "epoch": 2.346936393697724,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004443806292068958,
      "loss": 5.0066,
      "mean_token_accuracy": 0.20071657747030258,
      "num_tokens": 52333655.0,
      "step": 30165
    },
    {
      "entropy": 5.686466407775879,
      "epoch": 2.347325423069442,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004443623842439792,
      "loss": 5.0348,
      "mean_token_accuracy": 0.198613478243351,
      "num_tokens": 52342111.0,
      "step": 30170
    },
    {
      "entropy": 5.711019039154053,
      "epoch": 2.347714452441159,
      "grad_norm": 1.21875,
      "learning_rate": 0.000444344136711258,
      "loss": 4.9721,
      "mean_token_accuracy": 0.20846291333436967,
      "num_tokens": 52351390.0,
      "step": 30175
    },
    {
      "entropy": 5.742067861557007,
      "epoch": 2.3481034818128768,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044432588660900905,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20401663482189178,
      "num_tokens": 52361602.0,
      "step": 30180
    },
    {
      "entropy": 5.761131620407104,
      "epoch": 2.3484925111845945,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004443076339375093,
      "loss": 5.0093,
      "mean_token_accuracy": 0.21114796847105027,
      "num_tokens": 52370304.0,
      "step": 30185
    },
    {
      "entropy": 5.688096714019776,
      "epoch": 2.348881540556312,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044428937869703575,
      "loss": 4.8696,
      "mean_token_accuracy": 0.2130025252699852,
      "num_tokens": 52378437.0,
      "step": 30190
    },
    {
      "entropy": 5.752493619918823,
      "epoch": 2.3492705699280294,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044427112088786525,
      "loss": 4.9833,
      "mean_token_accuracy": 0.20742824971675872,
      "num_tokens": 52388413.0,
      "step": 30195
    },
    {
      "entropy": 5.690942811965942,
      "epoch": 2.349659599299747,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004442528605102749,
      "loss": 4.9544,
      "mean_token_accuracy": 0.21171332895755768,
      "num_tokens": 52396296.0,
      "step": 30200
    },
    {
      "entropy": 5.686701345443725,
      "epoch": 2.350048628671465,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004442345975645418,
      "loss": 4.9396,
      "mean_token_accuracy": 0.20832957476377487,
      "num_tokens": 52403967.0,
      "step": 30205
    },
    {
      "entropy": 5.685170221328735,
      "epoch": 2.350437658043182,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004442163320509431,
      "loss": 4.9943,
      "mean_token_accuracy": 0.21321189999580384,
      "num_tokens": 52413107.0,
      "step": 30210
    },
    {
      "entropy": 5.7878213882446286,
      "epoch": 2.3508266874148998,
      "grad_norm": 1.25,
      "learning_rate": 0.0004441980639697558,
      "loss": 4.9667,
      "mean_token_accuracy": 0.20879411101341247,
      "num_tokens": 52421029.0,
      "step": 30215
    },
    {
      "entropy": 5.765743064880371,
      "epoch": 2.3512157167866174,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004441797933212573,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20232466757297515,
      "num_tokens": 52430538.0,
      "step": 30220
    },
    {
      "entropy": 5.776279592514038,
      "epoch": 2.351604746158335,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004441615201057247,
      "loss": 5.0413,
      "mean_token_accuracy": 0.19553168267011642,
      "num_tokens": 52440013.0,
      "step": 30225
    },
    {
      "entropy": 5.788431072235108,
      "epoch": 2.3519937755300524,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004441432443234352,
      "loss": 5.1365,
      "mean_token_accuracy": 0.1974329873919487,
      "num_tokens": 52449034.0,
      "step": 30230
    },
    {
      "entropy": 5.771905851364136,
      "epoch": 2.35238280490177,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044412496597466626,
      "loss": 5.0526,
      "mean_token_accuracy": 0.20038086026906968,
      "num_tokens": 52457649.0,
      "step": 30235
    },
    {
      "entropy": 5.7094920635223385,
      "epoch": 2.352771834273488,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004441066850596951,
      "loss": 4.9171,
      "mean_token_accuracy": 0.20861826986074447,
      "num_tokens": 52465861.0,
      "step": 30240
    },
    {
      "entropy": 5.795872783660888,
      "epoch": 2.353160863645205,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004440884015787992,
      "loss": 5.0447,
      "mean_token_accuracy": 0.2072172924876213,
      "num_tokens": 52474630.0,
      "step": 30245
    },
    {
      "entropy": 5.772836112976075,
      "epoch": 2.3535498930169227,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004440701155322558,
      "loss": 5.084,
      "mean_token_accuracy": 0.20028264224529266,
      "num_tokens": 52482759.0,
      "step": 30250
    },
    {
      "entropy": 5.779893827438355,
      "epoch": 2.3539389223886404,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004440518269203427,
      "loss": 5.0992,
      "mean_token_accuracy": 0.19213925004005433,
      "num_tokens": 52491839.0,
      "step": 30255
    },
    {
      "entropy": 5.741687774658203,
      "epoch": 2.354327951760358,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044403353574333715,
      "loss": 4.9324,
      "mean_token_accuracy": 0.20814654529094695,
      "num_tokens": 52500688.0,
      "step": 30260
    },
    {
      "entropy": 5.732140731811524,
      "epoch": 2.3547169811320754,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044401524200151667,
      "loss": 5.0315,
      "mean_token_accuracy": 0.20167298913002013,
      "num_tokens": 52509031.0,
      "step": 30265
    },
    {
      "entropy": 5.716607999801636,
      "epoch": 2.355106010503793,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044399694569515895,
      "loss": 4.9718,
      "mean_token_accuracy": 0.19978254586458205,
      "num_tokens": 52518337.0,
      "step": 30270
    },
    {
      "entropy": 5.755332660675049,
      "epoch": 2.355495039875511,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044397864682454156,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20657032877206802,
      "num_tokens": 52527257.0,
      "step": 30275
    },
    {
      "entropy": 5.763963270187378,
      "epoch": 2.355884069247228,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004439603453899421,
      "loss": 5.0615,
      "mean_token_accuracy": 0.20134871900081636,
      "num_tokens": 52536999.0,
      "step": 30280
    },
    {
      "entropy": 5.774192905426025,
      "epoch": 2.3562730986189457,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004439420413916384,
      "loss": 5.0662,
      "mean_token_accuracy": 0.19252280294895172,
      "num_tokens": 52545975.0,
      "step": 30285
    },
    {
      "entropy": 5.805777502059937,
      "epoch": 2.3566621279906634,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044392373482990817,
      "loss": 5.0306,
      "mean_token_accuracy": 0.20294691175222396,
      "num_tokens": 52554667.0,
      "step": 30290
    },
    {
      "entropy": 5.755733966827393,
      "epoch": 2.3570511573623807,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004439054257050291,
      "loss": 4.9999,
      "mean_token_accuracy": 0.2121336743235588,
      "num_tokens": 52563469.0,
      "step": 30295
    },
    {
      "entropy": 5.664082145690918,
      "epoch": 2.3574401867340984,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004438871140172789,
      "loss": 4.9468,
      "mean_token_accuracy": 0.21007690578699112,
      "num_tokens": 52572038.0,
      "step": 30300
    },
    {
      "entropy": 5.7204447269439695,
      "epoch": 2.357829216105816,
      "grad_norm": 1.375,
      "learning_rate": 0.0004438687997669357,
      "loss": 5.014,
      "mean_token_accuracy": 0.20014785677194596,
      "num_tokens": 52579642.0,
      "step": 30305
    },
    {
      "entropy": 5.7093226432800295,
      "epoch": 2.3582182454775333,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044385048295427724,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20871184766292572,
      "num_tokens": 52587425.0,
      "step": 30310
    },
    {
      "entropy": 5.707075023651123,
      "epoch": 2.358607274849251,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004438321635795814,
      "loss": 4.951,
      "mean_token_accuracy": 0.2113201156258583,
      "num_tokens": 52596560.0,
      "step": 30315
    },
    {
      "entropy": 5.799524021148682,
      "epoch": 2.3589963042209687,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004438138416431262,
      "loss": 5.0872,
      "mean_token_accuracy": 0.20104867070913315,
      "num_tokens": 52604977.0,
      "step": 30320
    },
    {
      "entropy": 5.776003074645996,
      "epoch": 2.3593853335926864,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044379551714518966,
      "loss": 5.019,
      "mean_token_accuracy": 0.20619112849235535,
      "num_tokens": 52614069.0,
      "step": 30325
    },
    {
      "entropy": 5.750258731842041,
      "epoch": 2.3597743629644037,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044377719008604984,
      "loss": 5.0262,
      "mean_token_accuracy": 0.2068758338689804,
      "num_tokens": 52623065.0,
      "step": 30330
    },
    {
      "entropy": 5.769026756286621,
      "epoch": 2.3601633923361214,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004437588604659848,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20404935777187347,
      "num_tokens": 52632422.0,
      "step": 30335
    },
    {
      "entropy": 5.777815341949463,
      "epoch": 2.360552421707839,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004437405282852726,
      "loss": 5.0954,
      "mean_token_accuracy": 0.19611981213092805,
      "num_tokens": 52641358.0,
      "step": 30340
    },
    {
      "entropy": 5.752602815628052,
      "epoch": 2.3609414510795563,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044372219354419154,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20513907372951506,
      "num_tokens": 52650011.0,
      "step": 30345
    },
    {
      "entropy": 5.804011869430542,
      "epoch": 2.361330480451274,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004437038562430197,
      "loss": 5.0525,
      "mean_token_accuracy": 0.19754240959882735,
      "num_tokens": 52658111.0,
      "step": 30350
    },
    {
      "entropy": 5.746442890167236,
      "epoch": 2.3617195098229917,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044368551638203536,
      "loss": 4.907,
      "mean_token_accuracy": 0.21603894531726836,
      "num_tokens": 52666181.0,
      "step": 30355
    },
    {
      "entropy": 5.70458459854126,
      "epoch": 2.3621085391947094,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004436671739615168,
      "loss": 5.0278,
      "mean_token_accuracy": 0.20600993633270265,
      "num_tokens": 52675218.0,
      "step": 30360
    },
    {
      "entropy": 5.722906923294067,
      "epoch": 2.3624975685664267,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044364882898174235,
      "loss": 5.0673,
      "mean_token_accuracy": 0.2028121054172516,
      "num_tokens": 52684254.0,
      "step": 30365
    },
    {
      "entropy": 5.76861801147461,
      "epoch": 2.3628865979381444,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004436304814429903,
      "loss": 5.0208,
      "mean_token_accuracy": 0.19871625304222107,
      "num_tokens": 52693106.0,
      "step": 30370
    },
    {
      "entropy": 5.794282293319702,
      "epoch": 2.363275627309862,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004436121313455391,
      "loss": 5.0428,
      "mean_token_accuracy": 0.19950013011693954,
      "num_tokens": 52702260.0,
      "step": 30375
    },
    {
      "entropy": 5.759212684631348,
      "epoch": 2.3636646566815793,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004435937786896673,
      "loss": 5.0115,
      "mean_token_accuracy": 0.20813346803188323,
      "num_tokens": 52710546.0,
      "step": 30380
    },
    {
      "entropy": 5.664081382751465,
      "epoch": 2.364053686053297,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044357542347565323,
      "loss": 5.0164,
      "mean_token_accuracy": 0.20363387316465378,
      "num_tokens": 52719197.0,
      "step": 30385
    },
    {
      "entropy": 5.837940692901611,
      "epoch": 2.3644427154250147,
      "grad_norm": 1.25,
      "learning_rate": 0.0004435570657037753,
      "loss": 5.1217,
      "mean_token_accuracy": 0.1986718848347664,
      "num_tokens": 52728242.0,
      "step": 30390
    },
    {
      "entropy": 5.725397253036499,
      "epoch": 2.364831744796732,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004435387053743123,
      "loss": 4.9105,
      "mean_token_accuracy": 0.2097751960158348,
      "num_tokens": 52736943.0,
      "step": 30395
    },
    {
      "entropy": 5.669610118865966,
      "epoch": 2.3652207741684497,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044352034248754265,
      "loss": 4.956,
      "mean_token_accuracy": 0.21170894652605057,
      "num_tokens": 52744893.0,
      "step": 30400
    },
    {
      "entropy": 5.635525035858154,
      "epoch": 2.3656098035401674,
      "grad_norm": 1.171875,
      "learning_rate": 0.000443501977043745,
      "loss": 4.8753,
      "mean_token_accuracy": 0.21222026497125626,
      "num_tokens": 52753492.0,
      "step": 30405
    },
    {
      "entropy": 5.760714197158814,
      "epoch": 2.3659988329118846,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004434836090431981,
      "loss": 5.0052,
      "mean_token_accuracy": 0.2014874905347824,
      "num_tokens": 52762598.0,
      "step": 30410
    },
    {
      "entropy": 5.70653600692749,
      "epoch": 2.3663878622836023,
      "grad_norm": 1.25,
      "learning_rate": 0.0004434652384861806,
      "loss": 5.0386,
      "mean_token_accuracy": 0.20415837168693543,
      "num_tokens": 52770942.0,
      "step": 30415
    },
    {
      "entropy": 5.747300529479981,
      "epoch": 2.36677689165532,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004434468653729712,
      "loss": 4.9821,
      "mean_token_accuracy": 0.2032642111182213,
      "num_tokens": 52779237.0,
      "step": 30420
    },
    {
      "entropy": 5.774164152145386,
      "epoch": 2.3671659210270377,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044342848970384876,
      "loss": 5.0511,
      "mean_token_accuracy": 0.19744522571563722,
      "num_tokens": 52787721.0,
      "step": 30425
    },
    {
      "entropy": 5.781344842910767,
      "epoch": 2.367554950398755,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044341011147909206,
      "loss": 5.007,
      "mean_token_accuracy": 0.20270038545131683,
      "num_tokens": 52795765.0,
      "step": 30430
    },
    {
      "entropy": 5.683582496643067,
      "epoch": 2.3679439797704727,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044339173069897996,
      "loss": 4.9193,
      "mean_token_accuracy": 0.2117093250155449,
      "num_tokens": 52804104.0,
      "step": 30435
    },
    {
      "entropy": 5.68350772857666,
      "epoch": 2.3683330091421904,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044337334736379143,
      "loss": 4.9472,
      "mean_token_accuracy": 0.20307990908622742,
      "num_tokens": 52812817.0,
      "step": 30440
    },
    {
      "entropy": 5.68878402709961,
      "epoch": 2.3687220385139076,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004433549614738053,
      "loss": 4.9583,
      "mean_token_accuracy": 0.20704864859580993,
      "num_tokens": 52821924.0,
      "step": 30445
    },
    {
      "entropy": 5.7439206600189205,
      "epoch": 2.3691110678856253,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044333657302930056,
      "loss": 4.9407,
      "mean_token_accuracy": 0.21496936231851577,
      "num_tokens": 52830943.0,
      "step": 30450
    },
    {
      "entropy": 5.7390539169311525,
      "epoch": 2.369500097257343,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004433181820305564,
      "loss": 5.063,
      "mean_token_accuracy": 0.2065854251384735,
      "num_tokens": 52839697.0,
      "step": 30455
    },
    {
      "entropy": 5.73288049697876,
      "epoch": 2.3698891266290607,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044329978847785156,
      "loss": 5.0313,
      "mean_token_accuracy": 0.19832108914852142,
      "num_tokens": 52848713.0,
      "step": 30460
    },
    {
      "entropy": 5.844886350631714,
      "epoch": 2.370278156000778,
      "grad_norm": 1.25,
      "learning_rate": 0.0004432813923714654,
      "loss": 5.0961,
      "mean_token_accuracy": 0.1919212132692337,
      "num_tokens": 52856852.0,
      "step": 30465
    },
    {
      "entropy": 5.715063190460205,
      "epoch": 2.3706671853724957,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044326299371167695,
      "loss": 4.9321,
      "mean_token_accuracy": 0.2084295392036438,
      "num_tokens": 52865235.0,
      "step": 30470
    },
    {
      "entropy": 5.748228549957275,
      "epoch": 2.3710562147442134,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044324459249876536,
      "loss": 4.9341,
      "mean_token_accuracy": 0.20734256356954575,
      "num_tokens": 52873454.0,
      "step": 30475
    },
    {
      "entropy": 5.716871500015259,
      "epoch": 2.3714452441159306,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004432261887330099,
      "loss": 4.9796,
      "mean_token_accuracy": 0.210710708796978,
      "num_tokens": 52882551.0,
      "step": 30480
    },
    {
      "entropy": 5.7935709953308105,
      "epoch": 2.3718342734876483,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004432077824146898,
      "loss": 5.045,
      "mean_token_accuracy": 0.19619593322277068,
      "num_tokens": 52891145.0,
      "step": 30485
    },
    {
      "entropy": 5.770518255233765,
      "epoch": 2.372223302859366,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004431893735440843,
      "loss": 5.0,
      "mean_token_accuracy": 0.196412356197834,
      "num_tokens": 52899486.0,
      "step": 30490
    },
    {
      "entropy": 5.695523595809936,
      "epoch": 2.3726123322310833,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004431709621214727,
      "loss": 4.986,
      "mean_token_accuracy": 0.20175232738256454,
      "num_tokens": 52907896.0,
      "step": 30495
    },
    {
      "entropy": 5.704916524887085,
      "epoch": 2.373001361602801,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044315254814713455,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20173268169164657,
      "num_tokens": 52916489.0,
      "step": 30500
    },
    {
      "entropy": 5.742334461212158,
      "epoch": 2.3733903909745186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044313413162134894,
      "loss": 5.0152,
      "mean_token_accuracy": 0.2031744346022606,
      "num_tokens": 52924559.0,
      "step": 30505
    },
    {
      "entropy": 5.740375137329101,
      "epoch": 2.373779420346236,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004431157125443958,
      "loss": 5.0519,
      "mean_token_accuracy": 0.20284222513437272,
      "num_tokens": 52934064.0,
      "step": 30510
    },
    {
      "entropy": 5.757227277755737,
      "epoch": 2.3741684497179536,
      "grad_norm": 1.25,
      "learning_rate": 0.000443097290916554,
      "loss": 5.0104,
      "mean_token_accuracy": 0.20499978959560394,
      "num_tokens": 52943283.0,
      "step": 30515
    },
    {
      "entropy": 5.709242486953736,
      "epoch": 2.3745574790896713,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004430788667381035,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21229366213083267,
      "num_tokens": 52951694.0,
      "step": 30520
    },
    {
      "entropy": 5.739170932769776,
      "epoch": 2.374946508461389,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044306044000932374,
      "loss": 5.0198,
      "mean_token_accuracy": 0.19908483624458312,
      "num_tokens": 52959850.0,
      "step": 30525
    },
    {
      "entropy": 5.803647422790528,
      "epoch": 2.3753355378331062,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004430420107304943,
      "loss": 5.1635,
      "mean_token_accuracy": 0.19567000716924668,
      "num_tokens": 52969378.0,
      "step": 30530
    },
    {
      "entropy": 5.772091484069824,
      "epoch": 2.375724567204824,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00044302357890189475,
      "loss": 5.0154,
      "mean_token_accuracy": 0.21185450851917267,
      "num_tokens": 52977178.0,
      "step": 30535
    },
    {
      "entropy": 5.813728094100952,
      "epoch": 2.3761135965765416,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004430051445238049,
      "loss": 5.0389,
      "mean_token_accuracy": 0.20136389434337615,
      "num_tokens": 52986035.0,
      "step": 30540
    },
    {
      "entropy": 5.709472703933716,
      "epoch": 2.376502625948259,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044298670759650426,
      "loss": 5.0506,
      "mean_token_accuracy": 0.20140443742275238,
      "num_tokens": 52995399.0,
      "step": 30545
    },
    {
      "entropy": 5.755706453323365,
      "epoch": 2.3768916553199766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004429682681202728,
      "loss": 4.9912,
      "mean_token_accuracy": 0.20847638547420502,
      "num_tokens": 53004500.0,
      "step": 30550
    },
    {
      "entropy": 5.822487306594849,
      "epoch": 2.3772806846916943,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044294982609539027,
      "loss": 5.1016,
      "mean_token_accuracy": 0.20143768489360808,
      "num_tokens": 53014161.0,
      "step": 30555
    },
    {
      "entropy": 5.693950366973877,
      "epoch": 2.377669714063412,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004429313815221363,
      "loss": 4.8739,
      "mean_token_accuracy": 0.21416620016098023,
      "num_tokens": 53022560.0,
      "step": 30560
    },
    {
      "entropy": 5.737769937515258,
      "epoch": 2.3780587434351292,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044291293440079107,
      "loss": 5.0289,
      "mean_token_accuracy": 0.20435957461595536,
      "num_tokens": 53031194.0,
      "step": 30565
    },
    {
      "entropy": 5.735292530059814,
      "epoch": 2.378447772806847,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004428944847316342,
      "loss": 5.0421,
      "mean_token_accuracy": 0.20446851551532746,
      "num_tokens": 53039512.0,
      "step": 30570
    },
    {
      "entropy": 5.790896081924439,
      "epoch": 2.3788368021785646,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004428760325149458,
      "loss": 5.0248,
      "mean_token_accuracy": 0.2068217396736145,
      "num_tokens": 53048579.0,
      "step": 30575
    },
    {
      "entropy": 5.820733880996704,
      "epoch": 2.379225831550282,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044285757775100584,
      "loss": 5.14,
      "mean_token_accuracy": 0.1972361534833908,
      "num_tokens": 53057770.0,
      "step": 30580
    },
    {
      "entropy": 5.7189453601837155,
      "epoch": 2.3796148609219996,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044283912044009436,
      "loss": 5.0124,
      "mean_token_accuracy": 0.2032826766371727,
      "num_tokens": 53067127.0,
      "step": 30585
    },
    {
      "entropy": 5.792666721343994,
      "epoch": 2.3800038902937173,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044282066058249123,
      "loss": 4.9823,
      "mean_token_accuracy": 0.20830806344747543,
      "num_tokens": 53076351.0,
      "step": 30590
    },
    {
      "entropy": 5.776123332977295,
      "epoch": 2.380392919665435,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004428021981784768,
      "loss": 5.034,
      "mean_token_accuracy": 0.20373939722776413,
      "num_tokens": 53085472.0,
      "step": 30595
    },
    {
      "entropy": 5.73580641746521,
      "epoch": 2.3807819490371522,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044278373322833106,
      "loss": 4.959,
      "mean_token_accuracy": 0.2072167068719864,
      "num_tokens": 53094181.0,
      "step": 30600
    },
    {
      "entropy": 5.705240201950073,
      "epoch": 2.38117097840887,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044276526573233416,
      "loss": 5.0479,
      "mean_token_accuracy": 0.19591701924800872,
      "num_tokens": 53103421.0,
      "step": 30605
    },
    {
      "entropy": 5.766218090057373,
      "epoch": 2.381560007780587,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004427467956907664,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20815358310937881,
      "num_tokens": 53112058.0,
      "step": 30610
    },
    {
      "entropy": 5.792999029159546,
      "epoch": 2.381949037152305,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044272832310390814,
      "loss": 5.0503,
      "mean_token_accuracy": 0.20328713655471803,
      "num_tokens": 53119755.0,
      "step": 30615
    },
    {
      "entropy": 5.642931079864502,
      "epoch": 2.3823380665240226,
      "grad_norm": 1.25,
      "learning_rate": 0.0004427098479720394,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21384665220975876,
      "num_tokens": 53128529.0,
      "step": 30620
    },
    {
      "entropy": 5.7326866626739506,
      "epoch": 2.3827270958957403,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044269137029544073,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21244428008794786,
      "num_tokens": 53137594.0,
      "step": 30625
    },
    {
      "entropy": 5.82869176864624,
      "epoch": 2.3831161252674575,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004426728900743924,
      "loss": 5.0179,
      "mean_token_accuracy": 0.19769106209278106,
      "num_tokens": 53146521.0,
      "step": 30630
    },
    {
      "entropy": 5.752753734588623,
      "epoch": 2.3835051546391752,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044265440730917484,
      "loss": 4.9845,
      "mean_token_accuracy": 0.21093598306179046,
      "num_tokens": 53154832.0,
      "step": 30635
    },
    {
      "entropy": 5.728213930130005,
      "epoch": 2.383894184010893,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044263592200006845,
      "loss": 5.0589,
      "mean_token_accuracy": 0.2018907830119133,
      "num_tokens": 53163889.0,
      "step": 30640
    },
    {
      "entropy": 5.729156875610352,
      "epoch": 2.38428321338261,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044261743414735383,
      "loss": 5.0119,
      "mean_token_accuracy": 0.2039561614394188,
      "num_tokens": 53172688.0,
      "step": 30645
    },
    {
      "entropy": 5.796695184707642,
      "epoch": 2.384672242754328,
      "grad_norm": 1.25,
      "learning_rate": 0.0004425989437513114,
      "loss": 4.9963,
      "mean_token_accuracy": 0.19968696385622026,
      "num_tokens": 53180742.0,
      "step": 30650
    },
    {
      "entropy": 5.702871942520142,
      "epoch": 2.3850612721260456,
      "grad_norm": 1.25,
      "learning_rate": 0.0004425804508122218,
      "loss": 4.9739,
      "mean_token_accuracy": 0.21033537536859512,
      "num_tokens": 53188716.0,
      "step": 30655
    },
    {
      "entropy": 5.811306953430176,
      "epoch": 2.3854503014977633,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044256195533036566,
      "loss": 5.1091,
      "mean_token_accuracy": 0.19327108711004257,
      "num_tokens": 53197420.0,
      "step": 30660
    },
    {
      "entropy": 5.762665414810181,
      "epoch": 2.3858393308694805,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044254345730602345,
      "loss": 5.021,
      "mean_token_accuracy": 0.2021334320306778,
      "num_tokens": 53205927.0,
      "step": 30665
    },
    {
      "entropy": 5.772464609146118,
      "epoch": 2.386228360241198,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004425249567394759,
      "loss": 4.9792,
      "mean_token_accuracy": 0.20600952059030533,
      "num_tokens": 53214793.0,
      "step": 30670
    },
    {
      "entropy": 5.752489042282105,
      "epoch": 2.386617389612916,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044250645363100386,
      "loss": 4.9737,
      "mean_token_accuracy": 0.20647225081920623,
      "num_tokens": 53223429.0,
      "step": 30675
    },
    {
      "entropy": 5.741060638427735,
      "epoch": 2.387006418984633,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044248794798088805,
      "loss": 4.8766,
      "mean_token_accuracy": 0.21601116806268691,
      "num_tokens": 53231697.0,
      "step": 30680
    },
    {
      "entropy": 5.787207794189453,
      "epoch": 2.387395448356351,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004424694397894091,
      "loss": 5.0133,
      "mean_token_accuracy": 0.2037672594189644,
      "num_tokens": 53239318.0,
      "step": 30685
    },
    {
      "entropy": 5.757215976715088,
      "epoch": 2.3877844777280686,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044245092905684794,
      "loss": 5.0923,
      "mean_token_accuracy": 0.19857320189476013,
      "num_tokens": 53248079.0,
      "step": 30690
    },
    {
      "entropy": 5.762946271896363,
      "epoch": 2.3881735070997863,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044243241578348553,
      "loss": 5.0488,
      "mean_token_accuracy": 0.1989555060863495,
      "num_tokens": 53256856.0,
      "step": 30695
    },
    {
      "entropy": 5.792494916915894,
      "epoch": 2.3885625364715035,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004424138999696027,
      "loss": 5.0671,
      "mean_token_accuracy": 0.19989698827266694,
      "num_tokens": 53265810.0,
      "step": 30700
    },
    {
      "entropy": 5.827932357788086,
      "epoch": 2.388951565843221,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004423953816154804,
      "loss": 5.0587,
      "mean_token_accuracy": 0.20383265018463134,
      "num_tokens": 53274485.0,
      "step": 30705
    },
    {
      "entropy": 5.771138143539429,
      "epoch": 2.389340595214939,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044237686072139967,
      "loss": 4.9374,
      "mean_token_accuracy": 0.20443988889455794,
      "num_tokens": 53282322.0,
      "step": 30710
    },
    {
      "entropy": 5.79857816696167,
      "epoch": 2.389729624586656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004423583372876414,
      "loss": 5.0335,
      "mean_token_accuracy": 0.2043583109974861,
      "num_tokens": 53291868.0,
      "step": 30715
    },
    {
      "entropy": 5.733408451080322,
      "epoch": 2.390118653958374,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044233981131448677,
      "loss": 4.9341,
      "mean_token_accuracy": 0.21419957280158997,
      "num_tokens": 53300654.0,
      "step": 30720
    },
    {
      "entropy": 5.70671124458313,
      "epoch": 2.3905076833300916,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044232128280221683,
      "loss": 4.8393,
      "mean_token_accuracy": 0.21562927663326265,
      "num_tokens": 53309178.0,
      "step": 30725
    },
    {
      "entropy": 5.737383270263672,
      "epoch": 2.390896712701809,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004423027517511128,
      "loss": 5.0262,
      "mean_token_accuracy": 0.1998502254486084,
      "num_tokens": 53317536.0,
      "step": 30730
    },
    {
      "entropy": 5.730509662628174,
      "epoch": 2.3912857420735265,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044228421816145573,
      "loss": 5.0006,
      "mean_token_accuracy": 0.2007122352719307,
      "num_tokens": 53325942.0,
      "step": 30735
    },
    {
      "entropy": 5.732168245315552,
      "epoch": 2.391674771445244,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044226568203352694,
      "loss": 4.9936,
      "mean_token_accuracy": 0.205831778049469,
      "num_tokens": 53335124.0,
      "step": 30740
    },
    {
      "entropy": 5.749680995941162,
      "epoch": 2.3920638008169615,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044224714336760766,
      "loss": 5.0793,
      "mean_token_accuracy": 0.20469998717308044,
      "num_tokens": 53344222.0,
      "step": 30745
    },
    {
      "entropy": 5.737440967559815,
      "epoch": 2.392452830188679,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004422286021639792,
      "loss": 4.9954,
      "mean_token_accuracy": 0.20277657210826874,
      "num_tokens": 53352639.0,
      "step": 30750
    },
    {
      "entropy": 5.756830215454102,
      "epoch": 2.392841859560397,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004422100584229228,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20646135061979293,
      "num_tokens": 53361357.0,
      "step": 30755
    },
    {
      "entropy": 5.794086933135986,
      "epoch": 2.3932308889321146,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044219151214472,
      "loss": 5.0185,
      "mean_token_accuracy": 0.2117227792739868,
      "num_tokens": 53369680.0,
      "step": 30760
    },
    {
      "entropy": 5.74252347946167,
      "epoch": 2.393619918303832,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004421729633296521,
      "loss": 4.9986,
      "mean_token_accuracy": 0.2038286119699478,
      "num_tokens": 53378619.0,
      "step": 30765
    },
    {
      "entropy": 5.723381185531617,
      "epoch": 2.3940089476755495,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044215441197800054,
      "loss": 4.9878,
      "mean_token_accuracy": 0.2036074861884117,
      "num_tokens": 53387630.0,
      "step": 30770
    },
    {
      "entropy": 5.762703514099121,
      "epoch": 2.394397977047267,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044213585809004685,
      "loss": 4.9987,
      "mean_token_accuracy": 0.19673932939767838,
      "num_tokens": 53396696.0,
      "step": 30775
    },
    {
      "entropy": 5.799515295028686,
      "epoch": 2.3947870064189845,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004421173016660725,
      "loss": 5.0138,
      "mean_token_accuracy": 0.20214379578828812,
      "num_tokens": 53405097.0,
      "step": 30780
    },
    {
      "entropy": 5.6947455406188965,
      "epoch": 2.395176035790702,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004420987427063592,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2013086885213852,
      "num_tokens": 53413670.0,
      "step": 30785
    },
    {
      "entropy": 5.7160943984985355,
      "epoch": 2.39556506516242,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044208018121118833,
      "loss": 5.0827,
      "mean_token_accuracy": 0.19391552805900575,
      "num_tokens": 53422370.0,
      "step": 30790
    },
    {
      "entropy": 5.792573642730713,
      "epoch": 2.3959540945341375,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044206161718084164,
      "loss": 5.086,
      "mean_token_accuracy": 0.1956275537610054,
      "num_tokens": 53431607.0,
      "step": 30795
    },
    {
      "entropy": 5.752659749984741,
      "epoch": 2.396343123905855,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004420430506156009,
      "loss": 4.8988,
      "mean_token_accuracy": 0.21708032637834548,
      "num_tokens": 53440042.0,
      "step": 30800
    },
    {
      "entropy": 5.694740295410156,
      "epoch": 2.3967321532775725,
      "grad_norm": 1.359375,
      "learning_rate": 0.00044202448151574764,
      "loss": 4.9978,
      "mean_token_accuracy": 0.20830352753400802,
      "num_tokens": 53448780.0,
      "step": 30805
    },
    {
      "entropy": 5.774925899505615,
      "epoch": 2.39712118264929,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004420059098815638,
      "loss": 5.099,
      "mean_token_accuracy": 0.19672624468803407,
      "num_tokens": 53457598.0,
      "step": 30810
    },
    {
      "entropy": 5.804679822921753,
      "epoch": 2.3975102120210074,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004419873357133311,
      "loss": 5.0927,
      "mean_token_accuracy": 0.2033015638589859,
      "num_tokens": 53466387.0,
      "step": 30815
    },
    {
      "entropy": 5.769283628463745,
      "epoch": 2.397899241392725,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004419687590113313,
      "loss": 5.0465,
      "mean_token_accuracy": 0.20034196376800537,
      "num_tokens": 53475032.0,
      "step": 30820
    },
    {
      "entropy": 5.775168466567993,
      "epoch": 2.398288270764443,
      "grad_norm": 1.25,
      "learning_rate": 0.00044195017977584637,
      "loss": 5.0477,
      "mean_token_accuracy": 0.20598347634077072,
      "num_tokens": 53483142.0,
      "step": 30825
    },
    {
      "entropy": 5.803859758377075,
      "epoch": 2.39867730013616,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044193159800715823,
      "loss": 5.1145,
      "mean_token_accuracy": 0.20330944061279296,
      "num_tokens": 53492820.0,
      "step": 30830
    },
    {
      "entropy": 5.777610111236572,
      "epoch": 2.399066329507878,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044191301370554874,
      "loss": 5.0296,
      "mean_token_accuracy": 0.20308308452367782,
      "num_tokens": 53502018.0,
      "step": 30835
    },
    {
      "entropy": 5.759009170532226,
      "epoch": 2.3994553588795955,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044189442687129994,
      "loss": 4.9658,
      "mean_token_accuracy": 0.20476564019918442,
      "num_tokens": 53510287.0,
      "step": 30840
    },
    {
      "entropy": 5.732327318191528,
      "epoch": 2.3998443882513127,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004418758375046939,
      "loss": 4.9997,
      "mean_token_accuracy": 0.19870276004076004,
      "num_tokens": 53519042.0,
      "step": 30845
    },
    {
      "entropy": 5.818260145187378,
      "epoch": 2.4002334176230304,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044185724560601267,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20580644756555558,
      "num_tokens": 53527678.0,
      "step": 30850
    },
    {
      "entropy": 5.813451051712036,
      "epoch": 2.400622446994748,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004418386511755382,
      "loss": 5.0498,
      "mean_token_accuracy": 0.19405657649040223,
      "num_tokens": 53536371.0,
      "step": 30855
    },
    {
      "entropy": 5.733941507339478,
      "epoch": 2.401011476366466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004418200542135528,
      "loss": 5.0607,
      "mean_token_accuracy": 0.20138001292943955,
      "num_tokens": 53545534.0,
      "step": 30860
    },
    {
      "entropy": 5.735194826126099,
      "epoch": 2.401400505738183,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004418014547203386,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20785641372203828,
      "num_tokens": 53554904.0,
      "step": 30865
    },
    {
      "entropy": 5.807973337173462,
      "epoch": 2.401789535109901,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004417828526961778,
      "loss": 5.008,
      "mean_token_accuracy": 0.2054468497633934,
      "num_tokens": 53563359.0,
      "step": 30870
    },
    {
      "entropy": 5.743387937545776,
      "epoch": 2.4021785644816185,
      "grad_norm": 1.5,
      "learning_rate": 0.00044176424814135266,
      "loss": 5.0915,
      "mean_token_accuracy": 0.20372617542743682,
      "num_tokens": 53573428.0,
      "step": 30875
    },
    {
      "entropy": 5.784433507919312,
      "epoch": 2.4025675938533357,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004417456410561455,
      "loss": 5.0854,
      "mean_token_accuracy": 0.1971151813864708,
      "num_tokens": 53581719.0,
      "step": 30880
    },
    {
      "entropy": 5.809368228912353,
      "epoch": 2.4029566232250534,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004417270314408387,
      "loss": 5.129,
      "mean_token_accuracy": 0.19603049904108047,
      "num_tokens": 53589715.0,
      "step": 30885
    },
    {
      "entropy": 5.7506828784942625,
      "epoch": 2.403345652596771,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044170841929571454,
      "loss": 5.0026,
      "mean_token_accuracy": 0.20411199778318406,
      "num_tokens": 53598273.0,
      "step": 30890
    },
    {
      "entropy": 5.766194534301758,
      "epoch": 2.403734681968489,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044168980462105543,
      "loss": 5.0286,
      "mean_token_accuracy": 0.20072245746850967,
      "num_tokens": 53607824.0,
      "step": 30895
    },
    {
      "entropy": 5.767139101028443,
      "epoch": 2.404123711340206,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004416711874171439,
      "loss": 4.9619,
      "mean_token_accuracy": 0.2064263105392456,
      "num_tokens": 53616734.0,
      "step": 30900
    },
    {
      "entropy": 5.741784954071045,
      "epoch": 2.404512740711924,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004416525676842624,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20714612156152726,
      "num_tokens": 53625877.0,
      "step": 30905
    },
    {
      "entropy": 5.770332956314087,
      "epoch": 2.4049017700836415,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004416339454226933,
      "loss": 5.054,
      "mean_token_accuracy": 0.20587658435106276,
      "num_tokens": 53634872.0,
      "step": 30910
    },
    {
      "entropy": 5.725870943069458,
      "epoch": 2.4052907994553587,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004416153206327194,
      "loss": 4.9877,
      "mean_token_accuracy": 0.20971613377332687,
      "num_tokens": 53643916.0,
      "step": 30915
    },
    {
      "entropy": 5.722827577590943,
      "epoch": 2.4056798288270764,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044159669331462326,
      "loss": 5.0212,
      "mean_token_accuracy": 0.2008420631289482,
      "num_tokens": 53652496.0,
      "step": 30920
    },
    {
      "entropy": 5.71380033493042,
      "epoch": 2.406068858198794,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044157806346868737,
      "loss": 4.9827,
      "mean_token_accuracy": 0.20694566667079925,
      "num_tokens": 53661191.0,
      "step": 30925
    },
    {
      "entropy": 5.754685735702514,
      "epoch": 2.406457887570512,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044155943109519454,
      "loss": 5.0168,
      "mean_token_accuracy": 0.20084999948740007,
      "num_tokens": 53669520.0,
      "step": 30930
    },
    {
      "entropy": 5.764369535446167,
      "epoch": 2.406846916942229,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004415407961944274,
      "loss": 5.0288,
      "mean_token_accuracy": 0.20243485122919083,
      "num_tokens": 53678398.0,
      "step": 30935
    },
    {
      "entropy": 5.753860664367676,
      "epoch": 2.4072359463139468,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044152215876666883,
      "loss": 5.0459,
      "mean_token_accuracy": 0.2029173880815506,
      "num_tokens": 53687870.0,
      "step": 30940
    },
    {
      "entropy": 5.827104091644287,
      "epoch": 2.407624975685664,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004415035188122016,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20591508001089096,
      "num_tokens": 53697335.0,
      "step": 30945
    },
    {
      "entropy": 5.684218549728394,
      "epoch": 2.4080140050573817,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044148487633130837,
      "loss": 4.9132,
      "mean_token_accuracy": 0.20648579597473143,
      "num_tokens": 53705457.0,
      "step": 30950
    },
    {
      "entropy": 5.68250527381897,
      "epoch": 2.4084030344290994,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044146623132427213,
      "loss": 4.8394,
      "mean_token_accuracy": 0.2154872477054596,
      "num_tokens": 53714092.0,
      "step": 30955
    },
    {
      "entropy": 5.748840093612671,
      "epoch": 2.408792063800817,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044144758379137584,
      "loss": 5.0216,
      "mean_token_accuracy": 0.20288541316986083,
      "num_tokens": 53722972.0,
      "step": 30960
    },
    {
      "entropy": 5.762690925598145,
      "epoch": 2.4091810931725344,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044142893373290236,
      "loss": 4.9568,
      "mean_token_accuracy": 0.20952513664960862,
      "num_tokens": 53730842.0,
      "step": 30965
    },
    {
      "entropy": 5.778192853927612,
      "epoch": 2.409570122544252,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004414102811491348,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20045557171106337,
      "num_tokens": 53740241.0,
      "step": 30970
    },
    {
      "entropy": 5.693689203262329,
      "epoch": 2.4099591519159698,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044139162604035597,
      "loss": 4.9602,
      "mean_token_accuracy": 0.20526500344276427,
      "num_tokens": 53749653.0,
      "step": 30975
    },
    {
      "entropy": 5.749517774581909,
      "epoch": 2.410348181287687,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044137296840684917,
      "loss": 4.9706,
      "mean_token_accuracy": 0.2130556657910347,
      "num_tokens": 53758161.0,
      "step": 30980
    },
    {
      "entropy": 5.740653800964355,
      "epoch": 2.4107372106594047,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004413543082488973,
      "loss": 4.8745,
      "mean_token_accuracy": 0.2261263757944107,
      "num_tokens": 53766345.0,
      "step": 30985
    },
    {
      "entropy": 5.707566356658935,
      "epoch": 2.4111262400311224,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004413356455667836,
      "loss": 4.9267,
      "mean_token_accuracy": 0.21193260997533797,
      "num_tokens": 53774725.0,
      "step": 30990
    },
    {
      "entropy": 5.6797408103942875,
      "epoch": 2.41151526940284,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004413169803607913,
      "loss": 4.9629,
      "mean_token_accuracy": 0.20364972203969955,
      "num_tokens": 53783386.0,
      "step": 30995
    },
    {
      "entropy": 5.7507082462310795,
      "epoch": 2.4119042987745574,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044129831263120344,
      "loss": 4.9594,
      "mean_token_accuracy": 0.20621503591537477,
      "num_tokens": 53791167.0,
      "step": 31000
    },
    {
      "entropy": 5.761775636672974,
      "epoch": 2.412293328146275,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004412796423783034,
      "loss": 5.0043,
      "mean_token_accuracy": 0.1969989135861397,
      "num_tokens": 53800227.0,
      "step": 31005
    },
    {
      "entropy": 5.720632743835449,
      "epoch": 2.4126823575179928,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004412609696023745,
      "loss": 5.0076,
      "mean_token_accuracy": 0.205685855448246,
      "num_tokens": 53808591.0,
      "step": 31010
    },
    {
      "entropy": 5.796106052398682,
      "epoch": 2.41307138688971,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004412422943037,
      "loss": 5.1009,
      "mean_token_accuracy": 0.1994265303015709,
      "num_tokens": 53817738.0,
      "step": 31015
    },
    {
      "entropy": 5.794653224945068,
      "epoch": 2.4134604162614277,
      "grad_norm": 1.125,
      "learning_rate": 0.00044122361648256327,
      "loss": 5.0238,
      "mean_token_accuracy": 0.20598164796829224,
      "num_tokens": 53826919.0,
      "step": 31020
    },
    {
      "entropy": 5.758966541290283,
      "epoch": 2.4138494456331454,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044120493613924766,
      "loss": 4.948,
      "mean_token_accuracy": 0.20949797630310057,
      "num_tokens": 53835415.0,
      "step": 31025
    },
    {
      "entropy": 5.7432861328125,
      "epoch": 2.414238475004863,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044118625327403677,
      "loss": 5.044,
      "mean_token_accuracy": 0.20488122552633287,
      "num_tokens": 53844992.0,
      "step": 31030
    },
    {
      "entropy": 5.751257276535034,
      "epoch": 2.4146275043765804,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000441167567887214,
      "loss": 5.0056,
      "mean_token_accuracy": 0.2039441704750061,
      "num_tokens": 53853864.0,
      "step": 31035
    },
    {
      "entropy": 5.773746061325073,
      "epoch": 2.415016533748298,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004411488799790629,
      "loss": 4.9927,
      "mean_token_accuracy": 0.20326935350894929,
      "num_tokens": 53863021.0,
      "step": 31040
    },
    {
      "entropy": 5.780874156951905,
      "epoch": 2.4154055631200158,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044113018954986693,
      "loss": 5.0471,
      "mean_token_accuracy": 0.19775691926479338,
      "num_tokens": 53873259.0,
      "step": 31045
    },
    {
      "entropy": 5.69344129562378,
      "epoch": 2.415794592491733,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004411114965999098,
      "loss": 4.9648,
      "mean_token_accuracy": 0.2138729512691498,
      "num_tokens": 53881861.0,
      "step": 31050
    },
    {
      "entropy": 5.742616128921509,
      "epoch": 2.4161836218634507,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000441092801129475,
      "loss": 5.0412,
      "mean_token_accuracy": 0.20045126080513,
      "num_tokens": 53890852.0,
      "step": 31055
    },
    {
      "entropy": 5.7446253299713135,
      "epoch": 2.4165726512351684,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044107410313884643,
      "loss": 4.9213,
      "mean_token_accuracy": 0.20891886502504348,
      "num_tokens": 53899452.0,
      "step": 31060
    },
    {
      "entropy": 5.778521108627319,
      "epoch": 2.4169616806068857,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044105540262830763,
      "loss": 5.1194,
      "mean_token_accuracy": 0.19871475398540497,
      "num_tokens": 53908458.0,
      "step": 31065
    },
    {
      "entropy": 5.750793981552124,
      "epoch": 2.4173507099786034,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004410366995981424,
      "loss": 5.0326,
      "mean_token_accuracy": 0.1982795551419258,
      "num_tokens": 53917998.0,
      "step": 31070
    },
    {
      "entropy": 5.804146003723145,
      "epoch": 2.417739739350321,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044101799404863457,
      "loss": 5.116,
      "mean_token_accuracy": 0.19242469370365142,
      "num_tokens": 53926320.0,
      "step": 31075
    },
    {
      "entropy": 5.756047534942627,
      "epoch": 2.4181287687220383,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004409992859800679,
      "loss": 5.0208,
      "mean_token_accuracy": 0.2014957696199417,
      "num_tokens": 53934723.0,
      "step": 31080
    },
    {
      "entropy": 5.825441360473633,
      "epoch": 2.418517798093756,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004409805753927263,
      "loss": 5.0987,
      "mean_token_accuracy": 0.19815285503864288,
      "num_tokens": 53943125.0,
      "step": 31085
    },
    {
      "entropy": 5.77634825706482,
      "epoch": 2.4189068274654737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004409618622868936,
      "loss": 5.0286,
      "mean_token_accuracy": 0.19822373241186142,
      "num_tokens": 53951781.0,
      "step": 31090
    },
    {
      "entropy": 5.7268750190734865,
      "epoch": 2.4192958568371914,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044094314666285385,
      "loss": 5.0073,
      "mean_token_accuracy": 0.19966767877340316,
      "num_tokens": 53960556.0,
      "step": 31095
    },
    {
      "entropy": 5.710502624511719,
      "epoch": 2.4196848862089086,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044092442852089095,
      "loss": 4.9146,
      "mean_token_accuracy": 0.21497627794742585,
      "num_tokens": 53969075.0,
      "step": 31100
    },
    {
      "entropy": 5.847242403030395,
      "epoch": 2.4200739155806263,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004409057078612889,
      "loss": 5.1673,
      "mean_token_accuracy": 0.1934692770242691,
      "num_tokens": 53978203.0,
      "step": 31105
    },
    {
      "entropy": 5.764504766464233,
      "epoch": 2.420462944952344,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044088698468433194,
      "loss": 4.9957,
      "mean_token_accuracy": 0.2009505346417427,
      "num_tokens": 53986935.0,
      "step": 31110
    },
    {
      "entropy": 5.7430871486663815,
      "epoch": 2.4208519743240613,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044086825899030383,
      "loss": 5.0115,
      "mean_token_accuracy": 0.19897337406873702,
      "num_tokens": 53996167.0,
      "step": 31115
    },
    {
      "entropy": 5.733886003494263,
      "epoch": 2.421241003695779,
      "grad_norm": 1.1875,
      "learning_rate": 0.000440849530779489,
      "loss": 5.0268,
      "mean_token_accuracy": 0.20216165482997894,
      "num_tokens": 54004722.0,
      "step": 31120
    },
    {
      "entropy": 5.722410106658936,
      "epoch": 2.4216300330674967,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004408308000521715,
      "loss": 4.9736,
      "mean_token_accuracy": 0.21243120580911637,
      "num_tokens": 54012660.0,
      "step": 31125
    },
    {
      "entropy": 5.777955722808838,
      "epoch": 2.4220190624392144,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044081206680863556,
      "loss": 5.0389,
      "mean_token_accuracy": 0.2026614874601364,
      "num_tokens": 54021517.0,
      "step": 31130
    },
    {
      "entropy": 5.784278154373169,
      "epoch": 2.4224080918109316,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044079333104916554,
      "loss": 5.0208,
      "mean_token_accuracy": 0.20796803534030914,
      "num_tokens": 54030081.0,
      "step": 31135
    },
    {
      "entropy": 5.7475159645080565,
      "epoch": 2.4227971211826493,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004407745927740454,
      "loss": 4.9613,
      "mean_token_accuracy": 0.20729743093252181,
      "num_tokens": 54038214.0,
      "step": 31140
    },
    {
      "entropy": 5.664539432525634,
      "epoch": 2.423186150554367,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004407558519835598,
      "loss": 4.9098,
      "mean_token_accuracy": 0.20092959851026534,
      "num_tokens": 54046813.0,
      "step": 31145
    },
    {
      "entropy": 5.710810232162475,
      "epoch": 2.4235751799260843,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044073710867799295,
      "loss": 5.0038,
      "mean_token_accuracy": 0.20505196750164031,
      "num_tokens": 54056095.0,
      "step": 31150
    },
    {
      "entropy": 5.719667387008667,
      "epoch": 2.423964209297802,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004407183628576293,
      "loss": 4.9183,
      "mean_token_accuracy": 0.21075420677661896,
      "num_tokens": 54065173.0,
      "step": 31155
    },
    {
      "entropy": 5.811587333679199,
      "epoch": 2.4243532386695197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004406996145227532,
      "loss": 5.1532,
      "mean_token_accuracy": 0.19166182577610016,
      "num_tokens": 54073858.0,
      "step": 31160
    },
    {
      "entropy": 5.815307760238648,
      "epoch": 2.424742268041237,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004406808636736492,
      "loss": 5.0374,
      "mean_token_accuracy": 0.20236337631940843,
      "num_tokens": 54082129.0,
      "step": 31165
    },
    {
      "entropy": 5.761281251907349,
      "epoch": 2.4251312974129546,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044066211031060183,
      "loss": 4.976,
      "mean_token_accuracy": 0.20168417692184448,
      "num_tokens": 54090360.0,
      "step": 31170
    },
    {
      "entropy": 5.763425207138061,
      "epoch": 2.4255203267846723,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004406433544338956,
      "loss": 5.026,
      "mean_token_accuracy": 0.20317935943603516,
      "num_tokens": 54098309.0,
      "step": 31175
    },
    {
      "entropy": 5.802192735671997,
      "epoch": 2.4259093561563896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004406245960438151,
      "loss": 5.0041,
      "mean_token_accuracy": 0.2124850258231163,
      "num_tokens": 54106505.0,
      "step": 31180
    },
    {
      "entropy": 5.830397033691407,
      "epoch": 2.4262983855281073,
      "grad_norm": 1.21875,
      "learning_rate": 0.000440605835140645,
      "loss": 5.1705,
      "mean_token_accuracy": 0.1923434004187584,
      "num_tokens": 54115791.0,
      "step": 31185
    },
    {
      "entropy": 5.745622873306274,
      "epoch": 2.426687414899825,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004405870717246699,
      "loss": 4.9376,
      "mean_token_accuracy": 0.207210411131382,
      "num_tokens": 54124187.0,
      "step": 31190
    },
    {
      "entropy": 5.776695775985718,
      "epoch": 2.4270764442715427,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044056830579617464,
      "loss": 5.0581,
      "mean_token_accuracy": 0.19857363998889924,
      "num_tokens": 54133566.0,
      "step": 31195
    },
    {
      "entropy": 5.767778778076172,
      "epoch": 2.42746547364326,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044054953735544376,
      "loss": 4.9528,
      "mean_token_accuracy": 0.20657921433448792,
      "num_tokens": 54141698.0,
      "step": 31200
    },
    {
      "entropy": 5.763343000411988,
      "epoch": 2.4278545030149776,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004405307664027622,
      "loss": 5.0863,
      "mean_token_accuracy": 0.19878880977630614,
      "num_tokens": 54150094.0,
      "step": 31205
    },
    {
      "entropy": 5.715404081344604,
      "epoch": 2.4282435323866953,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004405119929384147,
      "loss": 4.9571,
      "mean_token_accuracy": 0.21758323162794113,
      "num_tokens": 54158770.0,
      "step": 31210
    },
    {
      "entropy": 5.7260936260223385,
      "epoch": 2.4286325617584126,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004404932169626862,
      "loss": 4.948,
      "mean_token_accuracy": 0.20435978472232819,
      "num_tokens": 54166912.0,
      "step": 31215
    },
    {
      "entropy": 5.73943977355957,
      "epoch": 2.4290215911301303,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004404744384758615,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20478032082319259,
      "num_tokens": 54176027.0,
      "step": 31220
    },
    {
      "entropy": 5.736687421798706,
      "epoch": 2.429410620501848,
      "grad_norm": 1.25,
      "learning_rate": 0.0004404556574782256,
      "loss": 4.9712,
      "mean_token_accuracy": 0.20925674438476563,
      "num_tokens": 54184116.0,
      "step": 31225
    },
    {
      "entropy": 5.69991250038147,
      "epoch": 2.4297996498735657,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044043687397006336,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21037691831588745,
      "num_tokens": 54193043.0,
      "step": 31230
    },
    {
      "entropy": 5.765369939804077,
      "epoch": 2.430188679245283,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044041808795166,
      "loss": 5.039,
      "mean_token_accuracy": 0.19867083430290222,
      "num_tokens": 54201953.0,
      "step": 31235
    },
    {
      "entropy": 5.759497499465942,
      "epoch": 2.4305777086170006,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004403992994233004,
      "loss": 5.0197,
      "mean_token_accuracy": 0.20836419314146043,
      "num_tokens": 54210316.0,
      "step": 31240
    },
    {
      "entropy": 5.783961439132691,
      "epoch": 2.4309667379887183,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044038050838526974,
      "loss": 4.9457,
      "mean_token_accuracy": 0.21202143728733064,
      "num_tokens": 54218795.0,
      "step": 31245
    },
    {
      "entropy": 5.740437269210815,
      "epoch": 2.4313557673604356,
      "grad_norm": 1.265625,
      "learning_rate": 0.000440361714837853,
      "loss": 4.994,
      "mean_token_accuracy": 0.19851709604263307,
      "num_tokens": 54227455.0,
      "step": 31250
    },
    {
      "entropy": 5.700189018249512,
      "epoch": 2.4317447967321533,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004403429187813355,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20650274753570558,
      "num_tokens": 54236124.0,
      "step": 31255
    },
    {
      "entropy": 5.716236925125122,
      "epoch": 2.432133826103871,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004403241202160024,
      "loss": 4.9581,
      "mean_token_accuracy": 0.20620622783899306,
      "num_tokens": 54244638.0,
      "step": 31260
    },
    {
      "entropy": 5.745661401748658,
      "epoch": 2.432522855475588,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004403053191421388,
      "loss": 4.9906,
      "mean_token_accuracy": 0.20385528057813646,
      "num_tokens": 54253678.0,
      "step": 31265
    },
    {
      "entropy": 5.827294254302979,
      "epoch": 2.432911884847306,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004402865155600302,
      "loss": 5.1361,
      "mean_token_accuracy": 0.19948826283216475,
      "num_tokens": 54262414.0,
      "step": 31270
    },
    {
      "entropy": 5.657776594161987,
      "epoch": 2.4333009142190236,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044026770946996183,
      "loss": 4.9066,
      "mean_token_accuracy": 0.21363504678010942,
      "num_tokens": 54270597.0,
      "step": 31275
    },
    {
      "entropy": 5.71487398147583,
      "epoch": 2.433689943590741,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044024890087221896,
      "loss": 4.9917,
      "mean_token_accuracy": 0.209790201485157,
      "num_tokens": 54279432.0,
      "step": 31280
    },
    {
      "entropy": 5.732794570922851,
      "epoch": 2.4340789729624586,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044023008976708705,
      "loss": 4.9351,
      "mean_token_accuracy": 0.2075113534927368,
      "num_tokens": 54287578.0,
      "step": 31285
    },
    {
      "entropy": 5.787867498397827,
      "epoch": 2.4344680023341763,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004402112761548515,
      "loss": 5.103,
      "mean_token_accuracy": 0.1984708860516548,
      "num_tokens": 54296119.0,
      "step": 31290
    },
    {
      "entropy": 5.698449754714966,
      "epoch": 2.434857031705894,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044019246003579776,
      "loss": 4.9191,
      "mean_token_accuracy": 0.20722371190786362,
      "num_tokens": 54304363.0,
      "step": 31295
    },
    {
      "entropy": 5.807777452468872,
      "epoch": 2.435246061077611,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004401736414102115,
      "loss": 5.0327,
      "mean_token_accuracy": 0.20351162552833557,
      "num_tokens": 54313672.0,
      "step": 31300
    },
    {
      "entropy": 5.7652710437774655,
      "epoch": 2.435635090449329,
      "grad_norm": 1.3125,
      "learning_rate": 0.000440154820278378,
      "loss": 5.0086,
      "mean_token_accuracy": 0.21334877461194993,
      "num_tokens": 54322163.0,
      "step": 31305
    },
    {
      "entropy": 5.746228218078613,
      "epoch": 2.4360241198210466,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004401359966405831,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2121948480606079,
      "num_tokens": 54331176.0,
      "step": 31310
    },
    {
      "entropy": 5.748073244094849,
      "epoch": 2.436413149192764,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044011717049711216,
      "loss": 5.0174,
      "mean_token_accuracy": 0.20464986860752105,
      "num_tokens": 54340404.0,
      "step": 31315
    },
    {
      "entropy": 5.774827384948731,
      "epoch": 2.4368021785644816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044009834184825103,
      "loss": 4.9563,
      "mean_token_accuracy": 0.20782427191734315,
      "num_tokens": 54349704.0,
      "step": 31320
    },
    {
      "entropy": 5.788943433761597,
      "epoch": 2.4371912079361993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004400795106942853,
      "loss": 5.0364,
      "mean_token_accuracy": 0.20307955741882325,
      "num_tokens": 54358869.0,
      "step": 31325
    },
    {
      "entropy": 5.714692640304565,
      "epoch": 2.437580237307917,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044006067703550083,
      "loss": 4.9285,
      "mean_token_accuracy": 0.21125146746635437,
      "num_tokens": 54367918.0,
      "step": 31330
    },
    {
      "entropy": 5.7495436668396,
      "epoch": 2.437969266679634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004400418408721833,
      "loss": 5.0077,
      "mean_token_accuracy": 0.1992639869451523,
      "num_tokens": 54375798.0,
      "step": 31335
    },
    {
      "entropy": 5.79714298248291,
      "epoch": 2.438358296051352,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044002300220461843,
      "loss": 4.9637,
      "mean_token_accuracy": 0.20415841341018676,
      "num_tokens": 54384230.0,
      "step": 31340
    },
    {
      "entropy": 5.728126430511475,
      "epoch": 2.4387473254230696,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004400041610330922,
      "loss": 4.9974,
      "mean_token_accuracy": 0.20550402104854584,
      "num_tokens": 54392444.0,
      "step": 31345
    },
    {
      "entropy": 5.741849517822265,
      "epoch": 2.439136354794787,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004399853173578905,
      "loss": 5.0064,
      "mean_token_accuracy": 0.20437606126070024,
      "num_tokens": 54400788.0,
      "step": 31350
    },
    {
      "entropy": 5.734001970291137,
      "epoch": 2.4395253841665046,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043996647117929916,
      "loss": 5.0268,
      "mean_token_accuracy": 0.20591295659542083,
      "num_tokens": 54409749.0,
      "step": 31355
    },
    {
      "entropy": 5.87733736038208,
      "epoch": 2.4399144135382222,
      "grad_norm": 1.125,
      "learning_rate": 0.0004399476224976043,
      "loss": 5.1653,
      "mean_token_accuracy": 0.19837927669286728,
      "num_tokens": 54419677.0,
      "step": 31360
    },
    {
      "entropy": 5.86934084892273,
      "epoch": 2.44030344290994,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043992877131309165,
      "loss": 5.1215,
      "mean_token_accuracy": 0.2048010468482971,
      "num_tokens": 54428821.0,
      "step": 31365
    },
    {
      "entropy": 5.795098209381104,
      "epoch": 2.440692472281657,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004399099176260475,
      "loss": 5.0641,
      "mean_token_accuracy": 0.20285094231367112,
      "num_tokens": 54437549.0,
      "step": 31370
    },
    {
      "entropy": 5.723764514923095,
      "epoch": 2.441081501653375,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004398910614367579,
      "loss": 4.9006,
      "mean_token_accuracy": 0.21222235709428788,
      "num_tokens": 54446247.0,
      "step": 31375
    },
    {
      "entropy": 5.813949203491211,
      "epoch": 2.441470531025092,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043987220274550876,
      "loss": 5.0433,
      "mean_token_accuracy": 0.19619151651859285,
      "num_tokens": 54455191.0,
      "step": 31380
    },
    {
      "entropy": 5.7631762504577635,
      "epoch": 2.44185956039681,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043985334155258643,
      "loss": 5.0256,
      "mean_token_accuracy": 0.2027174562215805,
      "num_tokens": 54463094.0,
      "step": 31385
    },
    {
      "entropy": 5.746235322952271,
      "epoch": 2.4422485897685275,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000439834477858277,
      "loss": 4.9753,
      "mean_token_accuracy": 0.19904089272022246,
      "num_tokens": 54471314.0,
      "step": 31390
    },
    {
      "entropy": 5.783353662490844,
      "epoch": 2.4426376191402452,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043981561166286676,
      "loss": 5.1012,
      "mean_token_accuracy": 0.19774192422628403,
      "num_tokens": 54480283.0,
      "step": 31395
    },
    {
      "entropy": 5.751322793960571,
      "epoch": 2.4430266485119625,
      "grad_norm": 1.265625,
      "learning_rate": 0.000439796742966642,
      "loss": 4.9819,
      "mean_token_accuracy": 0.20422791540622712,
      "num_tokens": 54488588.0,
      "step": 31400
    },
    {
      "entropy": 5.767865562438965,
      "epoch": 2.44341567788368,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043977787176988884,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20709134340286256,
      "num_tokens": 54497532.0,
      "step": 31405
    },
    {
      "entropy": 5.8346107006073,
      "epoch": 2.443804707255398,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004397589980728938,
      "loss": 5.0686,
      "mean_token_accuracy": 0.20094753354787825,
      "num_tokens": 54506152.0,
      "step": 31410
    },
    {
      "entropy": 5.767714548110962,
      "epoch": 2.444193736627115,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043974012187594324,
      "loss": 5.02,
      "mean_token_accuracy": 0.19887570142745972,
      "num_tokens": 54515674.0,
      "step": 31415
    },
    {
      "entropy": 5.679424238204956,
      "epoch": 2.444582765998833,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004397212431793235,
      "loss": 4.8437,
      "mean_token_accuracy": 0.21661657243967056,
      "num_tokens": 54523399.0,
      "step": 31420
    },
    {
      "entropy": 5.644727611541748,
      "epoch": 2.4449717953705505,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043970236198332107,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20299065560102464,
      "num_tokens": 54532128.0,
      "step": 31425
    },
    {
      "entropy": 5.766310691833496,
      "epoch": 2.4453608247422682,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043968347828822246,
      "loss": 4.9621,
      "mean_token_accuracy": 0.20218459665775299,
      "num_tokens": 54540474.0,
      "step": 31430
    },
    {
      "entropy": 5.748452758789062,
      "epoch": 2.4457498541139855,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004396645920943142,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20714785307645797,
      "num_tokens": 54549534.0,
      "step": 31435
    },
    {
      "entropy": 5.770737171173096,
      "epoch": 2.446138883485703,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004396457034018828,
      "loss": 4.9903,
      "mean_token_accuracy": 0.20018388330936432,
      "num_tokens": 54558071.0,
      "step": 31440
    },
    {
      "entropy": 5.807193231582642,
      "epoch": 2.446527912857421,
      "grad_norm": 1.171875,
      "learning_rate": 0.000439626812211215,
      "loss": 5.0679,
      "mean_token_accuracy": 0.19669016301631928,
      "num_tokens": 54565814.0,
      "step": 31445
    },
    {
      "entropy": 5.743308448791504,
      "epoch": 2.446916942229138,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043960791852259735,
      "loss": 5.0117,
      "mean_token_accuracy": 0.1916562482714653,
      "num_tokens": 54574911.0,
      "step": 31450
    },
    {
      "entropy": 5.745396375656128,
      "epoch": 2.447305971600856,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004395890223363165,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20544773042201997,
      "num_tokens": 54584129.0,
      "step": 31455
    },
    {
      "entropy": 5.730145740509033,
      "epoch": 2.4476950009725735,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043957012365265915,
      "loss": 4.9484,
      "mean_token_accuracy": 0.20830039083957672,
      "num_tokens": 54593203.0,
      "step": 31460
    },
    {
      "entropy": 5.738918590545654,
      "epoch": 2.4480840303442912,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043955122247191214,
      "loss": 5.0781,
      "mean_token_accuracy": 0.2047096699476242,
      "num_tokens": 54601436.0,
      "step": 31465
    },
    {
      "entropy": 5.696729850769043,
      "epoch": 2.4484730597160085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004395323187943623,
      "loss": 5.0405,
      "mean_token_accuracy": 0.19940754771232605,
      "num_tokens": 54610550.0,
      "step": 31470
    },
    {
      "entropy": 5.716815090179443,
      "epoch": 2.448862089087726,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004395134126202964,
      "loss": 4.9526,
      "mean_token_accuracy": 0.20434477180242538,
      "num_tokens": 54619176.0,
      "step": 31475
    },
    {
      "entropy": 5.764438629150391,
      "epoch": 2.449251118459444,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004394945039500012,
      "loss": 5.0045,
      "mean_token_accuracy": 0.2024429038167,
      "num_tokens": 54627259.0,
      "step": 31480
    },
    {
      "entropy": 5.714690780639648,
      "epoch": 2.449640147831161,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043947559278376385,
      "loss": 4.9963,
      "mean_token_accuracy": 0.1958848774433136,
      "num_tokens": 54635892.0,
      "step": 31485
    },
    {
      "entropy": 5.812391662597657,
      "epoch": 2.450029177202879,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043945667912187117,
      "loss": 5.0658,
      "mean_token_accuracy": 0.19724818766117097,
      "num_tokens": 54645360.0,
      "step": 31490
    },
    {
      "entropy": 5.762310361862182,
      "epoch": 2.4504182065745965,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004394377629646101,
      "loss": 4.9851,
      "mean_token_accuracy": 0.2070889264345169,
      "num_tokens": 54654968.0,
      "step": 31495
    },
    {
      "entropy": 5.73056263923645,
      "epoch": 2.4508072359463138,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043941884431226763,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21319037228822707,
      "num_tokens": 54663347.0,
      "step": 31500
    },
    {
      "entropy": 5.784131193161011,
      "epoch": 2.4511962653180315,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000439399923165131,
      "loss": 5.0546,
      "mean_token_accuracy": 0.19909590035676955,
      "num_tokens": 54671744.0,
      "step": 31505
    },
    {
      "entropy": 5.764223575592041,
      "epoch": 2.451585294689749,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004393809995234872,
      "loss": 4.9851,
      "mean_token_accuracy": 0.19961896985769273,
      "num_tokens": 54680098.0,
      "step": 31510
    },
    {
      "entropy": 5.762333345413208,
      "epoch": 2.4519743240614664,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004393620733876233,
      "loss": 5.065,
      "mean_token_accuracy": 0.2048175096511841,
      "num_tokens": 54689513.0,
      "step": 31515
    },
    {
      "entropy": 5.775946044921875,
      "epoch": 2.452363353433184,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004393431447578267,
      "loss": 5.0044,
      "mean_token_accuracy": 0.20746286809444428,
      "num_tokens": 54698079.0,
      "step": 31520
    },
    {
      "entropy": 5.756911849975586,
      "epoch": 2.452752382804902,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043932421363438427,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21012334376573563,
      "num_tokens": 54706956.0,
      "step": 31525
    },
    {
      "entropy": 5.687871599197388,
      "epoch": 2.4531414121766195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004393052800175835,
      "loss": 5.0253,
      "mean_token_accuracy": 0.2010633334517479,
      "num_tokens": 54716131.0,
      "step": 31530
    },
    {
      "entropy": 5.75602240562439,
      "epoch": 2.4535304415483368,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043928634390771166,
      "loss": 4.9597,
      "mean_token_accuracy": 0.20553407669067383,
      "num_tokens": 54725080.0,
      "step": 31535
    },
    {
      "entropy": 5.748590517044067,
      "epoch": 2.4539194709200545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043926740530505603,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20106348991394044,
      "num_tokens": 54733273.0,
      "step": 31540
    },
    {
      "entropy": 5.76924991607666,
      "epoch": 2.454308500291772,
      "grad_norm": 1.25,
      "learning_rate": 0.000439248464209904,
      "loss": 5.0344,
      "mean_token_accuracy": 0.2017551690340042,
      "num_tokens": 54742359.0,
      "step": 31545
    },
    {
      "entropy": 5.755935382843018,
      "epoch": 2.4546975296634894,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000439229520622543,
      "loss": 5.0151,
      "mean_token_accuracy": 0.2005144849419594,
      "num_tokens": 54750360.0,
      "step": 31550
    },
    {
      "entropy": 5.728951787948608,
      "epoch": 2.455086559035207,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004392105745432603,
      "loss": 5.0001,
      "mean_token_accuracy": 0.2034292086958885,
      "num_tokens": 54758663.0,
      "step": 31555
    },
    {
      "entropy": 5.783328580856323,
      "epoch": 2.455475588406925,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004391916259723436,
      "loss": 4.9575,
      "mean_token_accuracy": 0.21689614951610564,
      "num_tokens": 54767578.0,
      "step": 31560
    },
    {
      "entropy": 5.8123914241790775,
      "epoch": 2.4558646177786425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043917267491008023,
      "loss": 5.1216,
      "mean_token_accuracy": 0.19739974439144134,
      "num_tokens": 54776651.0,
      "step": 31565
    },
    {
      "entropy": 5.801870441436767,
      "epoch": 2.4562536471503598,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004391537213567579,
      "loss": 5.0087,
      "mean_token_accuracy": 0.19909669011831282,
      "num_tokens": 54784865.0,
      "step": 31570
    },
    {
      "entropy": 5.725860691070556,
      "epoch": 2.4566426765220775,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004391347653126641,
      "loss": 4.8611,
      "mean_token_accuracy": 0.21357672959566115,
      "num_tokens": 54794367.0,
      "step": 31575
    },
    {
      "entropy": 5.673507976531982,
      "epoch": 2.457031705893795,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043911580677808646,
      "loss": 4.9085,
      "mean_token_accuracy": 0.2100159779191017,
      "num_tokens": 54802606.0,
      "step": 31580
    },
    {
      "entropy": 5.680382013320923,
      "epoch": 2.4574207352655124,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043909684575331264,
      "loss": 4.933,
      "mean_token_accuracy": 0.22076333612203597,
      "num_tokens": 54812207.0,
      "step": 31585
    },
    {
      "entropy": 5.794582748413086,
      "epoch": 2.45780976463723,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004390778822386304,
      "loss": 4.9831,
      "mean_token_accuracy": 0.20951906144618987,
      "num_tokens": 54821047.0,
      "step": 31590
    },
    {
      "entropy": 5.895704650878907,
      "epoch": 2.458198794008948,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043905891623432754,
      "loss": 5.133,
      "mean_token_accuracy": 0.20003800988197326,
      "num_tokens": 54829849.0,
      "step": 31595
    },
    {
      "entropy": 5.7574687004089355,
      "epoch": 2.458587823380665,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043903994774069165,
      "loss": 5.0873,
      "mean_token_accuracy": 0.19183467775583268,
      "num_tokens": 54838690.0,
      "step": 31600
    },
    {
      "entropy": 5.741592216491699,
      "epoch": 2.4589768527523828,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004390209767580107,
      "loss": 5.0214,
      "mean_token_accuracy": 0.20689265578985214,
      "num_tokens": 54847795.0,
      "step": 31605
    },
    {
      "entropy": 5.7613077640533445,
      "epoch": 2.4593658821241005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004390020032865725,
      "loss": 4.9369,
      "mean_token_accuracy": 0.21359774172306062,
      "num_tokens": 54856676.0,
      "step": 31610
    },
    {
      "entropy": 5.739563512802124,
      "epoch": 2.4597549114958177,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004389830273266649,
      "loss": 4.9847,
      "mean_token_accuracy": 0.2077259197831154,
      "num_tokens": 54865177.0,
      "step": 31615
    },
    {
      "entropy": 5.723583555221557,
      "epoch": 2.4601439408675354,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043896404887857596,
      "loss": 4.957,
      "mean_token_accuracy": 0.20325980335474014,
      "num_tokens": 54872921.0,
      "step": 31620
    },
    {
      "entropy": 5.7308413028717045,
      "epoch": 2.460532970239253,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004389450679425935,
      "loss": 5.0765,
      "mean_token_accuracy": 0.19511204063892365,
      "num_tokens": 54882045.0,
      "step": 31625
    },
    {
      "entropy": 5.869683408737183,
      "epoch": 2.460921999610971,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004389260845190055,
      "loss": 5.0139,
      "mean_token_accuracy": 0.2037695601582527,
      "num_tokens": 54890647.0,
      "step": 31630
    },
    {
      "entropy": 5.771480894088745,
      "epoch": 2.461311028982688,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043890709860810014,
      "loss": 5.0387,
      "mean_token_accuracy": 0.20240434557199477,
      "num_tokens": 54898958.0,
      "step": 31635
    },
    {
      "entropy": 5.749298143386841,
      "epoch": 2.4617000583544058,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043888811021016545,
      "loss": 5.058,
      "mean_token_accuracy": 0.19514483362436294,
      "num_tokens": 54907329.0,
      "step": 31640
    },
    {
      "entropy": 5.7637872219085695,
      "epoch": 2.4620890877261234,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043886911932548953,
      "loss": 5.0158,
      "mean_token_accuracy": 0.20775482654571534,
      "num_tokens": 54916093.0,
      "step": 31645
    },
    {
      "entropy": 5.7986475944519045,
      "epoch": 2.4624781170978407,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004388501259543605,
      "loss": 5.0398,
      "mean_token_accuracy": 0.19661003053188325,
      "num_tokens": 54925030.0,
      "step": 31650
    },
    {
      "entropy": 5.770588684082031,
      "epoch": 2.4628671464695584,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004388311300970667,
      "loss": 4.9874,
      "mean_token_accuracy": 0.20603961050510405,
      "num_tokens": 54933719.0,
      "step": 31655
    },
    {
      "entropy": 5.780603313446045,
      "epoch": 2.463256175841276,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004388121317538962,
      "loss": 5.1059,
      "mean_token_accuracy": 0.1977620929479599,
      "num_tokens": 54941533.0,
      "step": 31660
    },
    {
      "entropy": 5.705181074142456,
      "epoch": 2.463645205212994,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004387931309251374,
      "loss": 4.945,
      "mean_token_accuracy": 0.21359920203685762,
      "num_tokens": 54949819.0,
      "step": 31665
    },
    {
      "entropy": 5.793100643157959,
      "epoch": 2.464034234584711,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004387741276110785,
      "loss": 5.0675,
      "mean_token_accuracy": 0.19822730422019957,
      "num_tokens": 54959199.0,
      "step": 31670
    },
    {
      "entropy": 5.780916166305542,
      "epoch": 2.4644232639564287,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043875512181200784,
      "loss": 5.05,
      "mean_token_accuracy": 0.20439544171094895,
      "num_tokens": 54967742.0,
      "step": 31675
    },
    {
      "entropy": 5.820680427551269,
      "epoch": 2.4648122933281464,
      "grad_norm": 1.25,
      "learning_rate": 0.0004387361135282139,
      "loss": 5.1369,
      "mean_token_accuracy": 0.1970336452126503,
      "num_tokens": 54976419.0,
      "step": 31680
    },
    {
      "entropy": 5.7901537895202635,
      "epoch": 2.4652013226998637,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043871710275998507,
      "loss": 4.9824,
      "mean_token_accuracy": 0.20564540028572081,
      "num_tokens": 54984975.0,
      "step": 31685
    },
    {
      "entropy": 5.766719961166382,
      "epoch": 2.4655903520715814,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004386980895076097,
      "loss": 5.0166,
      "mean_token_accuracy": 0.20487027019262313,
      "num_tokens": 54993746.0,
      "step": 31690
    },
    {
      "entropy": 5.74505934715271,
      "epoch": 2.465979381443299,
      "grad_norm": 1.25,
      "learning_rate": 0.00043867907377137646,
      "loss": 4.9347,
      "mean_token_accuracy": 0.20835868120193482,
      "num_tokens": 55002130.0,
      "step": 31695
    },
    {
      "entropy": 5.756981325149536,
      "epoch": 2.4663684108150163,
      "grad_norm": 1.25,
      "learning_rate": 0.00043866005555157376,
      "loss": 5.0658,
      "mean_token_accuracy": 0.19973008334636688,
      "num_tokens": 55010790.0,
      "step": 31700
    },
    {
      "entropy": 5.754735660552979,
      "epoch": 2.466757440186734,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004386410348484903,
      "loss": 5.0144,
      "mean_token_accuracy": 0.20036692768335343,
      "num_tokens": 55019542.0,
      "step": 31705
    },
    {
      "entropy": 5.7102643013000485,
      "epoch": 2.4671464695584517,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004386220116624145,
      "loss": 4.8777,
      "mean_token_accuracy": 0.21158256530761718,
      "num_tokens": 55029230.0,
      "step": 31710
    },
    {
      "entropy": 5.766079998016357,
      "epoch": 2.467535498930169,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043860298599363506,
      "loss": 4.9856,
      "mean_token_accuracy": 0.2016436815261841,
      "num_tokens": 55037371.0,
      "step": 31715
    },
    {
      "entropy": 5.720347738265991,
      "epoch": 2.4679245283018867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004385839578424408,
      "loss": 4.9826,
      "mean_token_accuracy": 0.2094579502940178,
      "num_tokens": 55046022.0,
      "step": 31720
    },
    {
      "entropy": 5.79997296333313,
      "epoch": 2.4683135576736044,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043856492720912037,
      "loss": 4.9472,
      "mean_token_accuracy": 0.216387040913105,
      "num_tokens": 55054713.0,
      "step": 31725
    },
    {
      "entropy": 5.644680595397949,
      "epoch": 2.468702587045322,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004385458940939624,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21954280883073807,
      "num_tokens": 55064317.0,
      "step": 31730
    },
    {
      "entropy": 5.7111180305480955,
      "epoch": 2.4690916164170393,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004385268584972559,
      "loss": 4.9972,
      "mean_token_accuracy": 0.21303058862686158,
      "num_tokens": 55072700.0,
      "step": 31735
    },
    {
      "entropy": 5.793711519241333,
      "epoch": 2.469480645788757,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004385078204192896,
      "loss": 5.104,
      "mean_token_accuracy": 0.19352618753910064,
      "num_tokens": 55081794.0,
      "step": 31740
    },
    {
      "entropy": 5.826669645309448,
      "epoch": 2.4698696751604747,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004384887798603523,
      "loss": 5.0928,
      "mean_token_accuracy": 0.20183662474155425,
      "num_tokens": 55090665.0,
      "step": 31745
    },
    {
      "entropy": 5.787430763244629,
      "epoch": 2.470258704532192,
      "grad_norm": 1.234375,
      "learning_rate": 0.000438469736820733,
      "loss": 4.9988,
      "mean_token_accuracy": 0.20787119418382644,
      "num_tokens": 55099306.0,
      "step": 31750
    },
    {
      "entropy": 5.761940956115723,
      "epoch": 2.4706477339039097,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043845069130072067,
      "loss": 5.0642,
      "mean_token_accuracy": 0.20064076781272888,
      "num_tokens": 55108420.0,
      "step": 31755
    },
    {
      "entropy": 5.822754383087158,
      "epoch": 2.4710367632756274,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004384316433006043,
      "loss": 4.992,
      "mean_token_accuracy": 0.2033417195081711,
      "num_tokens": 55116629.0,
      "step": 31760
    },
    {
      "entropy": 5.791800785064697,
      "epoch": 2.471425792647345,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043841259282067276,
      "loss": 5.0402,
      "mean_token_accuracy": 0.1981917977333069,
      "num_tokens": 55125469.0,
      "step": 31765
    },
    {
      "entropy": 5.746028041839599,
      "epoch": 2.4718148220190623,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043839353986121533,
      "loss": 5.0205,
      "mean_token_accuracy": 0.2037227600812912,
      "num_tokens": 55134194.0,
      "step": 31770
    },
    {
      "entropy": 5.796371269226074,
      "epoch": 2.47220385139078,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004383744844225209,
      "loss": 5.0446,
      "mean_token_accuracy": 0.20169512778520585,
      "num_tokens": 55142931.0,
      "step": 31775
    },
    {
      "entropy": 5.800432300567627,
      "epoch": 2.4725928807624977,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043835542650487875,
      "loss": 5.0394,
      "mean_token_accuracy": 0.20607608705759048,
      "num_tokens": 55152331.0,
      "step": 31780
    },
    {
      "entropy": 5.762153816223145,
      "epoch": 2.472981910134215,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004383363661085781,
      "loss": 5.0031,
      "mean_token_accuracy": 0.20307793468236923,
      "num_tokens": 55160860.0,
      "step": 31785
    },
    {
      "entropy": 5.79943962097168,
      "epoch": 2.4733709395059327,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004383173032339079,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19712002128362655,
      "num_tokens": 55169891.0,
      "step": 31790
    },
    {
      "entropy": 5.7095544815063475,
      "epoch": 2.4737599688776504,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004382982378811577,
      "loss": 4.9902,
      "mean_token_accuracy": 0.20529333502054214,
      "num_tokens": 55178283.0,
      "step": 31795
    },
    {
      "entropy": 5.78421311378479,
      "epoch": 2.474148998249368,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004382791700506166,
      "loss": 5.0528,
      "mean_token_accuracy": 0.20129430741071702,
      "num_tokens": 55186683.0,
      "step": 31800
    },
    {
      "entropy": 5.721342658996582,
      "epoch": 2.4745380276210853,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000438260099742574,
      "loss": 4.9198,
      "mean_token_accuracy": 0.21597544401884078,
      "num_tokens": 55194758.0,
      "step": 31805
    },
    {
      "entropy": 5.739154481887818,
      "epoch": 2.474927056992803,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004382410269573192,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20473675578832626,
      "num_tokens": 55203171.0,
      "step": 31810
    },
    {
      "entropy": 5.750705623626709,
      "epoch": 2.4753160863645203,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043822195169514167,
      "loss": 5.0074,
      "mean_token_accuracy": 0.20244007855653762,
      "num_tokens": 55210942.0,
      "step": 31815
    },
    {
      "entropy": 5.767849683761597,
      "epoch": 2.475705115736238,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043820287395633086,
      "loss": 5.0962,
      "mean_token_accuracy": 0.198972587287426,
      "num_tokens": 55219937.0,
      "step": 31820
    },
    {
      "entropy": 5.774168491363525,
      "epoch": 2.4760941451079557,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004381837937411761,
      "loss": 5.047,
      "mean_token_accuracy": 0.20371242463588715,
      "num_tokens": 55229461.0,
      "step": 31825
    },
    {
      "entropy": 5.700222206115723,
      "epoch": 2.4764831744796734,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043816471104996706,
      "loss": 4.9346,
      "mean_token_accuracy": 0.20494415163993834,
      "num_tokens": 55237910.0,
      "step": 31830
    },
    {
      "entropy": 5.740320014953613,
      "epoch": 2.4768722038513906,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004381456258829933,
      "loss": 5.0065,
      "mean_token_accuracy": 0.19810711294412614,
      "num_tokens": 55246131.0,
      "step": 31835
    },
    {
      "entropy": 5.739391565322876,
      "epoch": 2.4772612332231083,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043812653824054426,
      "loss": 4.9897,
      "mean_token_accuracy": 0.2071801021695137,
      "num_tokens": 55254677.0,
      "step": 31840
    },
    {
      "entropy": 5.794559383392334,
      "epoch": 2.477650262594826,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004381074481229097,
      "loss": 4.9836,
      "mean_token_accuracy": 0.2015093371272087,
      "num_tokens": 55263496.0,
      "step": 31845
    },
    {
      "entropy": 5.723418283462524,
      "epoch": 2.4780392919665433,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004380883555303791,
      "loss": 4.9411,
      "mean_token_accuracy": 0.20745514035224916,
      "num_tokens": 55272614.0,
      "step": 31850
    },
    {
      "entropy": 5.747090530395508,
      "epoch": 2.478428321338261,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004380692604632424,
      "loss": 4.9844,
      "mean_token_accuracy": 0.2028457134962082,
      "num_tokens": 55281595.0,
      "step": 31855
    },
    {
      "entropy": 5.810975694656372,
      "epoch": 2.4788173507099787,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043805016292178923,
      "loss": 5.089,
      "mean_token_accuracy": 0.20829716324806213,
      "num_tokens": 55291089.0,
      "step": 31860
    },
    {
      "entropy": 5.725016260147095,
      "epoch": 2.4792063800816964,
      "grad_norm": 1.25,
      "learning_rate": 0.0004380310629063094,
      "loss": 4.8796,
      "mean_token_accuracy": 0.2153559848666191,
      "num_tokens": 55299505.0,
      "step": 31865
    },
    {
      "entropy": 5.76038088798523,
      "epoch": 2.4795954094534136,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004380119604170925,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19794440418481826,
      "num_tokens": 55307383.0,
      "step": 31870
    },
    {
      "entropy": 5.777197980880738,
      "epoch": 2.4799844388251313,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043799285545442876,
      "loss": 5.0706,
      "mean_token_accuracy": 0.19132609963417052,
      "num_tokens": 55316312.0,
      "step": 31875
    },
    {
      "entropy": 5.807772874832153,
      "epoch": 2.480373468196849,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004379737480186078,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19863110035657883,
      "num_tokens": 55325818.0,
      "step": 31880
    },
    {
      "entropy": 5.745872354507446,
      "epoch": 2.4807624975685663,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043795463810991956,
      "loss": 5.0455,
      "mean_token_accuracy": 0.19534438997507095,
      "num_tokens": 55334641.0,
      "step": 31885
    },
    {
      "entropy": 5.812019014358521,
      "epoch": 2.481151526940284,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004379355257286541,
      "loss": 5.0246,
      "mean_token_accuracy": 0.20190017074346542,
      "num_tokens": 55343109.0,
      "step": 31890
    },
    {
      "entropy": 5.754524040222168,
      "epoch": 2.4815405563120017,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043791641087510136,
      "loss": 4.9794,
      "mean_token_accuracy": 0.21192089468240738,
      "num_tokens": 55351832.0,
      "step": 31895
    },
    {
      "entropy": 5.780402278900146,
      "epoch": 2.4819295856837194,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004378972935495514,
      "loss": 5.0364,
      "mean_token_accuracy": 0.20486932545900344,
      "num_tokens": 55361688.0,
      "step": 31900
    },
    {
      "entropy": 5.808208751678467,
      "epoch": 2.4823186150554366,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004378781737522943,
      "loss": 5.009,
      "mean_token_accuracy": 0.19845264852046968,
      "num_tokens": 55370572.0,
      "step": 31905
    },
    {
      "entropy": 5.7486252784729,
      "epoch": 2.4827076444271543,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043785905148362007,
      "loss": 4.9951,
      "mean_token_accuracy": 0.21258895695209504,
      "num_tokens": 55379134.0,
      "step": 31910
    },
    {
      "entropy": 5.722753095626831,
      "epoch": 2.483096673798872,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043783992674381903,
      "loss": 4.9737,
      "mean_token_accuracy": 0.20613017827272415,
      "num_tokens": 55387312.0,
      "step": 31915
    },
    {
      "entropy": 5.674282550811768,
      "epoch": 2.4834857031705893,
      "grad_norm": 1.125,
      "learning_rate": 0.00043782079953318126,
      "loss": 4.8764,
      "mean_token_accuracy": 0.211236609518528,
      "num_tokens": 55396208.0,
      "step": 31920
    },
    {
      "entropy": 5.766021251678467,
      "epoch": 2.483874732542307,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004378016698519971,
      "loss": 5.0383,
      "mean_token_accuracy": 0.20523053258657456,
      "num_tokens": 55404142.0,
      "step": 31925
    },
    {
      "entropy": 5.724481678009033,
      "epoch": 2.4842637619140246,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043778253770055666,
      "loss": 4.94,
      "mean_token_accuracy": 0.2143607333302498,
      "num_tokens": 55412227.0,
      "step": 31930
    },
    {
      "entropy": 5.6903464794158936,
      "epoch": 2.484652791285742,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043776340307915033,
      "loss": 4.9472,
      "mean_token_accuracy": 0.20541326254606246,
      "num_tokens": 55420566.0,
      "step": 31935
    },
    {
      "entropy": 5.742184591293335,
      "epoch": 2.4850418206574596,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004377442659880684,
      "loss": 4.9342,
      "mean_token_accuracy": 0.2088986814022064,
      "num_tokens": 55429456.0,
      "step": 31940
    },
    {
      "entropy": 5.7033790111541744,
      "epoch": 2.4854308500291773,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004377251264276013,
      "loss": 4.9276,
      "mean_token_accuracy": 0.20734609961509703,
      "num_tokens": 55437607.0,
      "step": 31945
    },
    {
      "entropy": 5.659671401977539,
      "epoch": 2.4858198794008945,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043770598439803936,
      "loss": 4.9472,
      "mean_token_accuracy": 0.21211216896772384,
      "num_tokens": 55446654.0,
      "step": 31950
    },
    {
      "entropy": 5.717013835906982,
      "epoch": 2.4862089087726122,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043768683989967304,
      "loss": 5.0241,
      "mean_token_accuracy": 0.20815762728452683,
      "num_tokens": 55454963.0,
      "step": 31955
    },
    {
      "entropy": 5.751030969619751,
      "epoch": 2.48659793814433,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043766769293279294,
      "loss": 4.972,
      "mean_token_accuracy": 0.200008387863636,
      "num_tokens": 55463259.0,
      "step": 31960
    },
    {
      "entropy": 5.679394721984863,
      "epoch": 2.4869869675160476,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043764854349768957,
      "loss": 4.9675,
      "mean_token_accuracy": 0.20318163335323333,
      "num_tokens": 55472390.0,
      "step": 31965
    },
    {
      "entropy": 5.774547958374024,
      "epoch": 2.487375996887765,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043762939159465333,
      "loss": 5.0457,
      "mean_token_accuracy": 0.20378268659114837,
      "num_tokens": 55480821.0,
      "step": 31970
    },
    {
      "entropy": 5.823787021636963,
      "epoch": 2.4877650262594826,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043761023722397495,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20083165615797044,
      "num_tokens": 55489270.0,
      "step": 31975
    },
    {
      "entropy": 5.7002945899963375,
      "epoch": 2.4881540556312003,
      "grad_norm": 1.203125,
      "learning_rate": 0.000437591080385945,
      "loss": 4.9592,
      "mean_token_accuracy": 0.2086532697081566,
      "num_tokens": 55497811.0,
      "step": 31980
    },
    {
      "entropy": 5.730239295959473,
      "epoch": 2.4885430850029175,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004375719210808543,
      "loss": 4.934,
      "mean_token_accuracy": 0.2009611189365387,
      "num_tokens": 55507068.0,
      "step": 31985
    },
    {
      "entropy": 5.71619668006897,
      "epoch": 2.4889321143746352,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004375527593089934,
      "loss": 4.9091,
      "mean_token_accuracy": 0.20773205161094666,
      "num_tokens": 55515922.0,
      "step": 31990
    },
    {
      "entropy": 5.735994291305542,
      "epoch": 2.489321143746353,
      "grad_norm": 1.125,
      "learning_rate": 0.0004375335950706531,
      "loss": 4.9887,
      "mean_token_accuracy": 0.20560373812913896,
      "num_tokens": 55524839.0,
      "step": 31995
    },
    {
      "entropy": 5.589614200592041,
      "epoch": 2.4897101731180706,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004375144283661242,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2079756587743759,
      "num_tokens": 55534426.0,
      "step": 32000
    },
    {
      "entropy": 5.744394731521607,
      "epoch": 2.490099202489788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004374952591956975,
      "loss": 5.0093,
      "mean_token_accuracy": 0.20008148550987243,
      "num_tokens": 55542773.0,
      "step": 32005
    },
    {
      "entropy": 5.797062540054322,
      "epoch": 2.4904882318615056,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043747608755966396,
      "loss": 4.9824,
      "mean_token_accuracy": 0.208916737139225,
      "num_tokens": 55551582.0,
      "step": 32010
    },
    {
      "entropy": 5.819371604919434,
      "epoch": 2.4908772612332233,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004374569134583143,
      "loss": 5.1113,
      "mean_token_accuracy": 0.20079169869422914,
      "num_tokens": 55561067.0,
      "step": 32015
    },
    {
      "entropy": 5.711532926559448,
      "epoch": 2.4912662906049405,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004374377368919396,
      "loss": 4.9649,
      "mean_token_accuracy": 0.20974690169095994,
      "num_tokens": 55569677.0,
      "step": 32020
    },
    {
      "entropy": 5.821907472610474,
      "epoch": 2.4916553199766582,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043741855786083085,
      "loss": 5.1517,
      "mean_token_accuracy": 0.19647407680749893,
      "num_tokens": 55578391.0,
      "step": 32025
    },
    {
      "entropy": 5.825729560852051,
      "epoch": 2.492044349348376,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043739937636527894,
      "loss": 5.0448,
      "mean_token_accuracy": 0.2019263833761215,
      "num_tokens": 55587074.0,
      "step": 32030
    },
    {
      "entropy": 5.772421598434448,
      "epoch": 2.492433378720093,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043738019240557494,
      "loss": 4.9672,
      "mean_token_accuracy": 0.2033824309706688,
      "num_tokens": 55596944.0,
      "step": 32035
    },
    {
      "entropy": 5.782548332214356,
      "epoch": 2.492822408091811,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043736100598201,
      "loss": 4.9707,
      "mean_token_accuracy": 0.2072565123438835,
      "num_tokens": 55605763.0,
      "step": 32040
    },
    {
      "entropy": 5.699039554595947,
      "epoch": 2.4932114374635286,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004373418170948752,
      "loss": 5.0059,
      "mean_token_accuracy": 0.2040828734636307,
      "num_tokens": 55614693.0,
      "step": 32045
    },
    {
      "entropy": 5.818109512329102,
      "epoch": 2.493600466835246,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004373226257444617,
      "loss": 5.159,
      "mean_token_accuracy": 0.1983616456389427,
      "num_tokens": 55623520.0,
      "step": 32050
    },
    {
      "entropy": 5.775056648254394,
      "epoch": 2.4939894962069635,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004373034319310608,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21263279616832734,
      "num_tokens": 55631818.0,
      "step": 32055
    },
    {
      "entropy": 5.829336404800415,
      "epoch": 2.4943785255786812,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043728423565496367,
      "loss": 5.1061,
      "mean_token_accuracy": 0.19901492893695832,
      "num_tokens": 55641388.0,
      "step": 32060
    },
    {
      "entropy": 5.705246019363403,
      "epoch": 2.494767554950399,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004372650369164615,
      "loss": 4.9924,
      "mean_token_accuracy": 0.20553259402513505,
      "num_tokens": 55650374.0,
      "step": 32065
    },
    {
      "entropy": 5.75170407295227,
      "epoch": 2.495156584322116,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043724583571584565,
      "loss": 5.0193,
      "mean_token_accuracy": 0.20523622632026672,
      "num_tokens": 55658598.0,
      "step": 32070
    },
    {
      "entropy": 5.756912517547607,
      "epoch": 2.495545613693834,
      "grad_norm": 1.25,
      "learning_rate": 0.00043722663205340753,
      "loss": 5.0558,
      "mean_token_accuracy": 0.19172070175409317,
      "num_tokens": 55666954.0,
      "step": 32075
    },
    {
      "entropy": 5.786532831192017,
      "epoch": 2.4959346430655516,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004372074259294384,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20152796655893326,
      "num_tokens": 55675451.0,
      "step": 32080
    },
    {
      "entropy": 5.778197383880615,
      "epoch": 2.496323672437269,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004371882173442299,
      "loss": 5.1166,
      "mean_token_accuracy": 0.19818124622106553,
      "num_tokens": 55683778.0,
      "step": 32085
    },
    {
      "entropy": 5.6984151840209964,
      "epoch": 2.4967127018089865,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043716900629807323,
      "loss": 4.9556,
      "mean_token_accuracy": 0.20971695482730865,
      "num_tokens": 55692440.0,
      "step": 32090
    },
    {
      "entropy": 5.730139636993409,
      "epoch": 2.497101731180704,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043714979279126006,
      "loss": 4.9795,
      "mean_token_accuracy": 0.20638670772314072,
      "num_tokens": 55700648.0,
      "step": 32095
    },
    {
      "entropy": 5.737063407897949,
      "epoch": 2.497490760552422,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004371305768240818,
      "loss": 4.97,
      "mean_token_accuracy": 0.2069748967885971,
      "num_tokens": 55709515.0,
      "step": 32100
    },
    {
      "entropy": 5.768846130371093,
      "epoch": 2.497879789924139,
      "grad_norm": 1.25,
      "learning_rate": 0.0004371113583968302,
      "loss": 5.0599,
      "mean_token_accuracy": 0.20146727412939072,
      "num_tokens": 55717876.0,
      "step": 32105
    },
    {
      "entropy": 5.735267639160156,
      "epoch": 2.498268819295857,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004370921375097967,
      "loss": 4.9965,
      "mean_token_accuracy": 0.20283293575048447,
      "num_tokens": 55726330.0,
      "step": 32110
    },
    {
      "entropy": 5.762791013717651,
      "epoch": 2.4986578486675746,
      "grad_norm": 1.125,
      "learning_rate": 0.00043707291416327307,
      "loss": 5.0682,
      "mean_token_accuracy": 0.20141904056072235,
      "num_tokens": 55736102.0,
      "step": 32115
    },
    {
      "entropy": 5.802491569519043,
      "epoch": 2.499046878039292,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004370536883575509,
      "loss": 5.0872,
      "mean_token_accuracy": 0.20419587641954423,
      "num_tokens": 55744022.0,
      "step": 32120
    },
    {
      "entropy": 5.764334774017334,
      "epoch": 2.4994359074110095,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004370344600929219,
      "loss": 5.0217,
      "mean_token_accuracy": 0.1992722675204277,
      "num_tokens": 55752613.0,
      "step": 32125
    },
    {
      "entropy": 5.752253675460816,
      "epoch": 2.499824936782727,
      "grad_norm": 1.125,
      "learning_rate": 0.0004370152293696779,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21064266562461853,
      "num_tokens": 55761402.0,
      "step": 32130
    },
    {
      "entropy": 5.732954216003418,
      "epoch": 2.500213966154445,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004369959961881107,
      "loss": 4.9117,
      "mean_token_accuracy": 0.20638342201709747,
      "num_tokens": 55769630.0,
      "step": 32135
    },
    {
      "entropy": 5.704201078414917,
      "epoch": 2.500602995526162,
      "grad_norm": 1.3125,
      "learning_rate": 0.000436976760548512,
      "loss": 5.0143,
      "mean_token_accuracy": 0.21001150012016295,
      "num_tokens": 55777928.0,
      "step": 32140
    },
    {
      "entropy": 5.718116235733032,
      "epoch": 2.50099202489788,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043695752245117393,
      "loss": 5.0205,
      "mean_token_accuracy": 0.2009720668196678,
      "num_tokens": 55786585.0,
      "step": 32145
    },
    {
      "entropy": 5.735006332397461,
      "epoch": 2.501381054269597,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004369382818963882,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21772377341985702,
      "num_tokens": 55795565.0,
      "step": 32150
    },
    {
      "entropy": 5.7350903987884525,
      "epoch": 2.501770083641315,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004369190388844467,
      "loss": 5.002,
      "mean_token_accuracy": 0.20105064213275908,
      "num_tokens": 55804211.0,
      "step": 32155
    },
    {
      "entropy": 5.7108855724334715,
      "epoch": 2.5021591130130325,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043689979341564154,
      "loss": 5.0133,
      "mean_token_accuracy": 0.2048020452260971,
      "num_tokens": 55812604.0,
      "step": 32160
    },
    {
      "entropy": 5.752273368835449,
      "epoch": 2.50254814238475,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004368805454902648,
      "loss": 4.9705,
      "mean_token_accuracy": 0.20754743218421937,
      "num_tokens": 55821293.0,
      "step": 32165
    },
    {
      "entropy": 5.717906951904297,
      "epoch": 2.5029371717564675,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043686129510860826,
      "loss": 4.9297,
      "mean_token_accuracy": 0.2031100198626518,
      "num_tokens": 55830423.0,
      "step": 32170
    },
    {
      "entropy": 5.792725658416748,
      "epoch": 2.503326201128185,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043684204227096435,
      "loss": 5.0088,
      "mean_token_accuracy": 0.20371535867452623,
      "num_tokens": 55838889.0,
      "step": 32175
    },
    {
      "entropy": 5.776485109329224,
      "epoch": 2.503715230499903,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000436822786977625,
      "loss": 5.0367,
      "mean_token_accuracy": 0.20217114090919494,
      "num_tokens": 55846670.0,
      "step": 32180
    },
    {
      "entropy": 5.7591992855072025,
      "epoch": 2.50410425987162,
      "grad_norm": 1.25,
      "learning_rate": 0.00043680352922888244,
      "loss": 5.04,
      "mean_token_accuracy": 0.20684549808502198,
      "num_tokens": 55855341.0,
      "step": 32185
    },
    {
      "entropy": 5.779816293716431,
      "epoch": 2.504493289243338,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004367842690250288,
      "loss": 4.9991,
      "mean_token_accuracy": 0.20337263196706773,
      "num_tokens": 55863491.0,
      "step": 32190
    },
    {
      "entropy": 5.692119836807251,
      "epoch": 2.5048823186150555,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004367650063663565,
      "loss": 4.9117,
      "mean_token_accuracy": 0.21335217505693435,
      "num_tokens": 55872346.0,
      "step": 32195
    },
    {
      "entropy": 5.689235258102417,
      "epoch": 2.505271347986773,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004367457412531576,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21118418127298355,
      "num_tokens": 55880672.0,
      "step": 32200
    },
    {
      "entropy": 5.808881855010986,
      "epoch": 2.5056603773584905,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004367264736857245,
      "loss": 5.0577,
      "mean_token_accuracy": 0.20485388189554216,
      "num_tokens": 55888938.0,
      "step": 32205
    },
    {
      "entropy": 5.770830678939819,
      "epoch": 2.506049406730208,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004367072036643497,
      "loss": 4.9413,
      "mean_token_accuracy": 0.2053830936551094,
      "num_tokens": 55897522.0,
      "step": 32210
    },
    {
      "entropy": 5.779721784591675,
      "epoch": 2.506438436101926,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043668793118932536,
      "loss": 4.9787,
      "mean_token_accuracy": 0.2041122019290924,
      "num_tokens": 55905425.0,
      "step": 32215
    },
    {
      "entropy": 5.70716962814331,
      "epoch": 2.506827465473643,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043666865626094405,
      "loss": 4.9948,
      "mean_token_accuracy": 0.20949369221925734,
      "num_tokens": 55913990.0,
      "step": 32220
    },
    {
      "entropy": 5.833953523635865,
      "epoch": 2.507216494845361,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043664937887949825,
      "loss": 5.1435,
      "mean_token_accuracy": 0.19138181060552598,
      "num_tokens": 55922899.0,
      "step": 32225
    },
    {
      "entropy": 5.7639604091644285,
      "epoch": 2.5076055242170785,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004366300990452803,
      "loss": 5.0029,
      "mean_token_accuracy": 0.20684336125850677,
      "num_tokens": 55930935.0,
      "step": 32230
    },
    {
      "entropy": 5.814674329757691,
      "epoch": 2.507994553588796,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043661081675858295,
      "loss": 5.0191,
      "mean_token_accuracy": 0.20735725164413452,
      "num_tokens": 55939258.0,
      "step": 32235
    },
    {
      "entropy": 5.729979562759399,
      "epoch": 2.5083835829605134,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043659153201969865,
      "loss": 5.0309,
      "mean_token_accuracy": 0.20199084877967835,
      "num_tokens": 55947672.0,
      "step": 32240
    },
    {
      "entropy": 5.736585330963135,
      "epoch": 2.508772612332231,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004365722448289201,
      "loss": 5.0356,
      "mean_token_accuracy": 0.2096530392765999,
      "num_tokens": 55956251.0,
      "step": 32245
    },
    {
      "entropy": 5.702652072906494,
      "epoch": 2.5091616417039484,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004365529551865399,
      "loss": 4.8701,
      "mean_token_accuracy": 0.2121596157550812,
      "num_tokens": 55965038.0,
      "step": 32250
    },
    {
      "entropy": 5.822845697402954,
      "epoch": 2.509550671075666,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043653366309285066,
      "loss": 5.0801,
      "mean_token_accuracy": 0.19495744109153748,
      "num_tokens": 55973140.0,
      "step": 32255
    },
    {
      "entropy": 5.743675374984742,
      "epoch": 2.509939700447384,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004365143685481452,
      "loss": 5.0338,
      "mean_token_accuracy": 0.19905417859554292,
      "num_tokens": 55981097.0,
      "step": 32260
    },
    {
      "entropy": 5.702450895309449,
      "epoch": 2.5103287298191015,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004364950715527164,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2115337923169136,
      "num_tokens": 55990077.0,
      "step": 32265
    },
    {
      "entropy": 5.765177869796753,
      "epoch": 2.5107177591908187,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004364757721068568,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20931134521961212,
      "num_tokens": 55999289.0,
      "step": 32270
    },
    {
      "entropy": 5.740087366104126,
      "epoch": 2.5111067885625364,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043645647021085945,
      "loss": 4.9436,
      "mean_token_accuracy": 0.21452676355838776,
      "num_tokens": 56007736.0,
      "step": 32275
    },
    {
      "entropy": 5.734495210647583,
      "epoch": 2.511495817934254,
      "grad_norm": 1.25,
      "learning_rate": 0.00043643716586501706,
      "loss": 4.9392,
      "mean_token_accuracy": 0.20805250108242035,
      "num_tokens": 56016623.0,
      "step": 32280
    },
    {
      "entropy": 5.753150081634521,
      "epoch": 2.5118848473059714,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043641785906962276,
      "loss": 5.0104,
      "mean_token_accuracy": 0.20457341521978378,
      "num_tokens": 56024536.0,
      "step": 32285
    },
    {
      "entropy": 5.740625381469727,
      "epoch": 2.512273876677689,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004363985498249693,
      "loss": 5.0371,
      "mean_token_accuracy": 0.2015567034482956,
      "num_tokens": 56033618.0,
      "step": 32290
    },
    {
      "entropy": 5.788509368896484,
      "epoch": 2.512662906049407,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043637923813134974,
      "loss": 5.0615,
      "mean_token_accuracy": 0.2025325432419777,
      "num_tokens": 56042836.0,
      "step": 32295
    },
    {
      "entropy": 5.768290042877197,
      "epoch": 2.5130519354211245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043635992398905713,
      "loss": 4.9952,
      "mean_token_accuracy": 0.19941093474626542,
      "num_tokens": 56051296.0,
      "step": 32300
    },
    {
      "entropy": 5.744476270675659,
      "epoch": 2.5134409647928417,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004363406073983844,
      "loss": 4.972,
      "mean_token_accuracy": 0.21004229933023452,
      "num_tokens": 56060093.0,
      "step": 32305
    },
    {
      "entropy": 5.781641483306885,
      "epoch": 2.5138299941645594,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043632128835962485,
      "loss": 4.9675,
      "mean_token_accuracy": 0.20435622781515123,
      "num_tokens": 56068441.0,
      "step": 32310
    },
    {
      "entropy": 5.719205331802368,
      "epoch": 2.514219023536277,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043630196687307156,
      "loss": 4.9269,
      "mean_token_accuracy": 0.2032313972711563,
      "num_tokens": 56076592.0,
      "step": 32315
    },
    {
      "entropy": 5.69644079208374,
      "epoch": 2.5146080529079944,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004362826429390176,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21266957521438598,
      "num_tokens": 56085299.0,
      "step": 32320
    },
    {
      "entropy": 5.724690389633179,
      "epoch": 2.514997082279712,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043626331655775624,
      "loss": 4.9456,
      "mean_token_accuracy": 0.21317402571439742,
      "num_tokens": 56093891.0,
      "step": 32325
    },
    {
      "entropy": 5.745850419998169,
      "epoch": 2.51538611165143,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043624398772958075,
      "loss": 5.073,
      "mean_token_accuracy": 0.19919949173927307,
      "num_tokens": 56102767.0,
      "step": 32330
    },
    {
      "entropy": 5.725226211547851,
      "epoch": 2.5157751410231475,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004362246564547844,
      "loss": 4.977,
      "mean_token_accuracy": 0.20842111557722093,
      "num_tokens": 56110861.0,
      "step": 32335
    },
    {
      "entropy": 5.79917802810669,
      "epoch": 2.5161641703948647,
      "grad_norm": 1.25,
      "learning_rate": 0.00043620532273366044,
      "loss": 5.0871,
      "mean_token_accuracy": 0.196714124083519,
      "num_tokens": 56119842.0,
      "step": 32340
    },
    {
      "entropy": 5.803333473205567,
      "epoch": 2.5165531997665824,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004361859865665024,
      "loss": 5.0577,
      "mean_token_accuracy": 0.20200078040361405,
      "num_tokens": 56128246.0,
      "step": 32345
    },
    {
      "entropy": 5.775778198242188,
      "epoch": 2.5169422291382997,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043616664795360346,
      "loss": 5.0466,
      "mean_token_accuracy": 0.2056443825364113,
      "num_tokens": 56137320.0,
      "step": 32350
    },
    {
      "entropy": 5.740857982635498,
      "epoch": 2.5173312585100174,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043614730689525716,
      "loss": 4.9751,
      "mean_token_accuracy": 0.20886615216732024,
      "num_tokens": 56145739.0,
      "step": 32355
    },
    {
      "entropy": 5.741092014312744,
      "epoch": 2.517720287881735,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043612796339175707,
      "loss": 4.9752,
      "mean_token_accuracy": 0.21255193799734115,
      "num_tokens": 56154526.0,
      "step": 32360
    },
    {
      "entropy": 5.824405384063721,
      "epoch": 2.5181093172534528,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043610861744339653,
      "loss": 5.1498,
      "mean_token_accuracy": 0.19722249656915664,
      "num_tokens": 56163209.0,
      "step": 32365
    },
    {
      "entropy": 5.754294395446777,
      "epoch": 2.5184983466251705,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004360892690504692,
      "loss": 4.981,
      "mean_token_accuracy": 0.2129140540957451,
      "num_tokens": 56171541.0,
      "step": 32370
    },
    {
      "entropy": 5.7471682071685795,
      "epoch": 2.5188873759968877,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004360699182132685,
      "loss": 4.9901,
      "mean_token_accuracy": 0.204765222966671,
      "num_tokens": 56180309.0,
      "step": 32375
    },
    {
      "entropy": 5.789842319488526,
      "epoch": 2.5192764053686054,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043605056493208825,
      "loss": 5.0628,
      "mean_token_accuracy": 0.20925644934177398,
      "num_tokens": 56188916.0,
      "step": 32380
    },
    {
      "entropy": 5.678201484680176,
      "epoch": 2.5196654347403227,
      "grad_norm": 1.15625,
      "learning_rate": 0.000436031209207222,
      "loss": 4.8876,
      "mean_token_accuracy": 0.2129327893257141,
      "num_tokens": 56197358.0,
      "step": 32385
    },
    {
      "entropy": 5.694745874404907,
      "epoch": 2.5200544641120404,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043601185103896346,
      "loss": 5.0198,
      "mean_token_accuracy": 0.19967437982559205,
      "num_tokens": 56206060.0,
      "step": 32390
    },
    {
      "entropy": 5.704558324813843,
      "epoch": 2.520443493483758,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043599249042760636,
      "loss": 4.9574,
      "mean_token_accuracy": 0.19709437638521193,
      "num_tokens": 56214539.0,
      "step": 32395
    },
    {
      "entropy": 5.66460018157959,
      "epoch": 2.5208325228554758,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043597312737344446,
      "loss": 4.8599,
      "mean_token_accuracy": 0.21213049292564393,
      "num_tokens": 56222386.0,
      "step": 32400
    },
    {
      "entropy": 5.6303221702575685,
      "epoch": 2.521221552227193,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043595376187677157,
      "loss": 4.9558,
      "mean_token_accuracy": 0.20472346395254135,
      "num_tokens": 56231036.0,
      "step": 32405
    },
    {
      "entropy": 5.746415758132935,
      "epoch": 2.5216105815989107,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043593439393788147,
      "loss": 5.024,
      "mean_token_accuracy": 0.2045419156551361,
      "num_tokens": 56239494.0,
      "step": 32410
    },
    {
      "entropy": 5.777943563461304,
      "epoch": 2.5219996109706284,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004359150235570682,
      "loss": 5.0377,
      "mean_token_accuracy": 0.20532775819301605,
      "num_tokens": 56248075.0,
      "step": 32415
    },
    {
      "entropy": 5.72051477432251,
      "epoch": 2.5223886403423457,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004358956507346255,
      "loss": 5.0495,
      "mean_token_accuracy": 0.2032727673649788,
      "num_tokens": 56256630.0,
      "step": 32420
    },
    {
      "entropy": 5.687564659118652,
      "epoch": 2.5227776697140634,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004358762754708474,
      "loss": 4.8752,
      "mean_token_accuracy": 0.21148403882980346,
      "num_tokens": 56264922.0,
      "step": 32425
    },
    {
      "entropy": 5.78585000038147,
      "epoch": 2.523166699085781,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043585689776602795,
      "loss": 5.1095,
      "mean_token_accuracy": 0.20018740445375444,
      "num_tokens": 56273996.0,
      "step": 32430
    },
    {
      "entropy": 5.8147501945495605,
      "epoch": 2.5235557284574988,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043583751762046104,
      "loss": 5.0484,
      "mean_token_accuracy": 0.2066384106874466,
      "num_tokens": 56283583.0,
      "step": 32435
    },
    {
      "entropy": 5.780672693252564,
      "epoch": 2.523944757829216,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004358181350344408,
      "loss": 4.9893,
      "mean_token_accuracy": 0.2115368887782097,
      "num_tokens": 56292197.0,
      "step": 32440
    },
    {
      "entropy": 5.783722066879273,
      "epoch": 2.5243337872009337,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004357987500082613,
      "loss": 5.0566,
      "mean_token_accuracy": 0.201078599691391,
      "num_tokens": 56301177.0,
      "step": 32445
    },
    {
      "entropy": 5.789465093612671,
      "epoch": 2.524722816572651,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043577936254221675,
      "loss": 5.0121,
      "mean_token_accuracy": 0.20161508470773698,
      "num_tokens": 56310024.0,
      "step": 32450
    },
    {
      "entropy": 5.804666376113891,
      "epoch": 2.5251118459443687,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004357599726366013,
      "loss": 5.0182,
      "mean_token_accuracy": 0.19669306874275208,
      "num_tokens": 56318190.0,
      "step": 32455
    },
    {
      "entropy": 5.690086841583252,
      "epoch": 2.5255008753160864,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004357405802917091,
      "loss": 4.8663,
      "mean_token_accuracy": 0.2109483078122139,
      "num_tokens": 56326443.0,
      "step": 32460
    },
    {
      "entropy": 5.699089574813843,
      "epoch": 2.525889904687804,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004357211855078345,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20818455517292023,
      "num_tokens": 56335263.0,
      "step": 32465
    },
    {
      "entropy": 5.78942346572876,
      "epoch": 2.5262789340595218,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043570178828527157,
      "loss": 5.0373,
      "mean_token_accuracy": 0.20858552008867265,
      "num_tokens": 56343202.0,
      "step": 32470
    },
    {
      "entropy": 5.716124057769775,
      "epoch": 2.526667963431239,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004356823886243149,
      "loss": 4.9253,
      "mean_token_accuracy": 0.20922564268112182,
      "num_tokens": 56352150.0,
      "step": 32475
    },
    {
      "entropy": 5.656986904144287,
      "epoch": 2.5270569928029567,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043566298652525875,
      "loss": 4.8984,
      "mean_token_accuracy": 0.21658487915992736,
      "num_tokens": 56360460.0,
      "step": 32480
    },
    {
      "entropy": 5.666233682632447,
      "epoch": 2.527446022174674,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043564358198839746,
      "loss": 5.0599,
      "mean_token_accuracy": 0.20606822222471238,
      "num_tokens": 56369889.0,
      "step": 32485
    },
    {
      "entropy": 5.755233907699585,
      "epoch": 2.5278350515463917,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004356241750140256,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21050177216529847,
      "num_tokens": 56378500.0,
      "step": 32490
    },
    {
      "entropy": 5.82009220123291,
      "epoch": 2.5282240809181094,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043560476560243743,
      "loss": 5.0644,
      "mean_token_accuracy": 0.20119451135396957,
      "num_tokens": 56387390.0,
      "step": 32495
    },
    {
      "entropy": 5.811213064193725,
      "epoch": 2.528613110289827,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004355853537539276,
      "loss": 5.0546,
      "mean_token_accuracy": 0.2040078490972519,
      "num_tokens": 56395782.0,
      "step": 32500
    },
    {
      "entropy": 5.777859687805176,
      "epoch": 2.5290021396615443,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004355659394687906,
      "loss": 4.944,
      "mean_token_accuracy": 0.2110195592045784,
      "num_tokens": 56405023.0,
      "step": 32505
    },
    {
      "entropy": 5.835201454162598,
      "epoch": 2.529391169033262,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004355465227473211,
      "loss": 5.0801,
      "mean_token_accuracy": 0.2027295023202896,
      "num_tokens": 56413669.0,
      "step": 32510
    },
    {
      "entropy": 5.749133348464966,
      "epoch": 2.5297801984049797,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004355271035898136,
      "loss": 5.0256,
      "mean_token_accuracy": 0.20584987848997116,
      "num_tokens": 56422286.0,
      "step": 32515
    },
    {
      "entropy": 5.777972936630249,
      "epoch": 2.530169227776697,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004355076819965628,
      "loss": 5.0365,
      "mean_token_accuracy": 0.20526931285858155,
      "num_tokens": 56430891.0,
      "step": 32520
    },
    {
      "entropy": 5.777743864059448,
      "epoch": 2.5305582571484146,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043548825796786353,
      "loss": 5.0328,
      "mean_token_accuracy": 0.20165961235761642,
      "num_tokens": 56439231.0,
      "step": 32525
    },
    {
      "entropy": 5.643569898605347,
      "epoch": 2.5309472865201323,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043546883150401023,
      "loss": 4.7817,
      "mean_token_accuracy": 0.22766311466693878,
      "num_tokens": 56447508.0,
      "step": 32530
    },
    {
      "entropy": 5.6794380187988285,
      "epoch": 2.53133631589185,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043544940260529787,
      "loss": 4.9725,
      "mean_token_accuracy": 0.2097108319401741,
      "num_tokens": 56456049.0,
      "step": 32535
    },
    {
      "entropy": 5.731295585632324,
      "epoch": 2.5317253452635673,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004354299712720213,
      "loss": 5.0276,
      "mean_token_accuracy": 0.20618246048688887,
      "num_tokens": 56464775.0,
      "step": 32540
    },
    {
      "entropy": 5.743636226654052,
      "epoch": 2.532114374635285,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004354105375044752,
      "loss": 4.9759,
      "mean_token_accuracy": 0.209343783557415,
      "num_tokens": 56473835.0,
      "step": 32545
    },
    {
      "entropy": 5.781631135940552,
      "epoch": 2.5325034040070027,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043539110130295445,
      "loss": 4.9995,
      "mean_token_accuracy": 0.20376506745815276,
      "num_tokens": 56482488.0,
      "step": 32550
    },
    {
      "entropy": 5.726345109939575,
      "epoch": 2.53289243337872,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043537166266775405,
      "loss": 4.9386,
      "mean_token_accuracy": 0.20942093580961227,
      "num_tokens": 56491928.0,
      "step": 32555
    },
    {
      "entropy": 5.677356815338134,
      "epoch": 2.5332814627504376,
      "grad_norm": 1.390625,
      "learning_rate": 0.000435352221599169,
      "loss": 4.9136,
      "mean_token_accuracy": 0.2085426241159439,
      "num_tokens": 56500247.0,
      "step": 32560
    },
    {
      "entropy": 5.741963148117065,
      "epoch": 2.5336704921221553,
      "grad_norm": 1.25,
      "learning_rate": 0.0004353327780974941,
      "loss": 5.0199,
      "mean_token_accuracy": 0.2032533809542656,
      "num_tokens": 56509507.0,
      "step": 32565
    },
    {
      "entropy": 5.816419887542724,
      "epoch": 2.534059521493873,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004353133321630246,
      "loss": 5.0568,
      "mean_token_accuracy": 0.203626349568367,
      "num_tokens": 56517895.0,
      "step": 32570
    },
    {
      "entropy": 5.713904953002929,
      "epoch": 2.5344485508655903,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004352938837960554,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21091684699058533,
      "num_tokens": 56526894.0,
      "step": 32575
    },
    {
      "entropy": 5.752105474472046,
      "epoch": 2.534837580237308,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043527443299688166,
      "loss": 5.0125,
      "mean_token_accuracy": 0.20263675153255462,
      "num_tokens": 56535970.0,
      "step": 32580
    },
    {
      "entropy": 5.819946002960205,
      "epoch": 2.5352266096090252,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043525497976579843,
      "loss": 4.9976,
      "mean_token_accuracy": 0.20278133749961852,
      "num_tokens": 56544385.0,
      "step": 32585
    },
    {
      "entropy": 5.701795530319214,
      "epoch": 2.535615638980743,
      "grad_norm": 1.328125,
      "learning_rate": 0.000435235524103101,
      "loss": 4.9881,
      "mean_token_accuracy": 0.2026505947113037,
      "num_tokens": 56552645.0,
      "step": 32590
    },
    {
      "entropy": 5.847436046600341,
      "epoch": 2.5360046683524606,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004352160660090846,
      "loss": 5.0903,
      "mean_token_accuracy": 0.19805312305688857,
      "num_tokens": 56562223.0,
      "step": 32595
    },
    {
      "entropy": 5.798403882980347,
      "epoch": 2.5363936977241783,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004351966054840443,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20716869831085205,
      "num_tokens": 56571015.0,
      "step": 32600
    },
    {
      "entropy": 5.695578670501709,
      "epoch": 2.5367827270958956,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043517714252827556,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21334373205900192,
      "num_tokens": 56579566.0,
      "step": 32605
    },
    {
      "entropy": 5.798113250732422,
      "epoch": 2.5371717564676133,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004351576771420737,
      "loss": 5.0241,
      "mean_token_accuracy": 0.20136623382568358,
      "num_tokens": 56588231.0,
      "step": 32610
    },
    {
      "entropy": 5.776970291137696,
      "epoch": 2.537560785839331,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004351382093257339,
      "loss": 4.9992,
      "mean_token_accuracy": 0.2088244378566742,
      "num_tokens": 56596581.0,
      "step": 32615
    },
    {
      "entropy": 5.687603521347046,
      "epoch": 2.5379498152110482,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004351187390795517,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21420602798461913,
      "num_tokens": 56605440.0,
      "step": 32620
    },
    {
      "entropy": 5.748534917831421,
      "epoch": 2.538338844582766,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043509926640382244,
      "loss": 5.0903,
      "mean_token_accuracy": 0.2027679890394211,
      "num_tokens": 56614819.0,
      "step": 32625
    },
    {
      "entropy": 5.728839635848999,
      "epoch": 2.5387278739544836,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004350797912988417,
      "loss": 5.0456,
      "mean_token_accuracy": 0.2071196585893631,
      "num_tokens": 56623372.0,
      "step": 32630
    },
    {
      "entropy": 5.7032088279724125,
      "epoch": 2.5391169033262013,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004350603137649049,
      "loss": 4.9484,
      "mean_token_accuracy": 0.2035582482814789,
      "num_tokens": 56631768.0,
      "step": 32635
    },
    {
      "entropy": 5.728284502029419,
      "epoch": 2.5395059326979186,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004350408338023075,
      "loss": 4.9772,
      "mean_token_accuracy": 0.20788359194993972,
      "num_tokens": 56640135.0,
      "step": 32640
    },
    {
      "entropy": 5.694622659683228,
      "epoch": 2.5398949620696363,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004350213514113453,
      "loss": 4.8769,
      "mean_token_accuracy": 0.21244669556617737,
      "num_tokens": 56648007.0,
      "step": 32645
    },
    {
      "entropy": 5.786375665664673,
      "epoch": 2.540283991441354,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043500186659231377,
      "loss": 5.0727,
      "mean_token_accuracy": 0.20343949049711227,
      "num_tokens": 56656926.0,
      "step": 32650
    },
    {
      "entropy": 5.821308326721192,
      "epoch": 2.5406730208130712,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004349823793455086,
      "loss": 5.0943,
      "mean_token_accuracy": 0.2006352052092552,
      "num_tokens": 56665404.0,
      "step": 32655
    },
    {
      "entropy": 5.756999921798706,
      "epoch": 2.541062050184789,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043496288967122547,
      "loss": 5.0063,
      "mean_token_accuracy": 0.20426109582185745,
      "num_tokens": 56673922.0,
      "step": 32660
    },
    {
      "entropy": 5.746078014373779,
      "epoch": 2.5414510795565066,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043494339756976007,
      "loss": 4.9741,
      "mean_token_accuracy": 0.2039829134941101,
      "num_tokens": 56682062.0,
      "step": 32665
    },
    {
      "entropy": 5.791140174865722,
      "epoch": 2.5418401089282243,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004349239030414082,
      "loss": 5.0921,
      "mean_token_accuracy": 0.19989509582519532,
      "num_tokens": 56690571.0,
      "step": 32670
    },
    {
      "entropy": 5.778707885742188,
      "epoch": 2.5422291382999416,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043490440608646557,
      "loss": 5.0208,
      "mean_token_accuracy": 0.2057028204202652,
      "num_tokens": 56699599.0,
      "step": 32675
    },
    {
      "entropy": 5.756120729446411,
      "epoch": 2.5426181676716593,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043488490670522817,
      "loss": 4.9498,
      "mean_token_accuracy": 0.2124951884150505,
      "num_tokens": 56708120.0,
      "step": 32680
    },
    {
      "entropy": 5.698273086547852,
      "epoch": 2.5430071970433765,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004348654048979918,
      "loss": 4.97,
      "mean_token_accuracy": 0.20708248168230056,
      "num_tokens": 56716339.0,
      "step": 32685
    },
    {
      "entropy": 5.7037496089935305,
      "epoch": 2.543396226415094,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004348459006650523,
      "loss": 4.8779,
      "mean_token_accuracy": 0.2205042839050293,
      "num_tokens": 56724157.0,
      "step": 32690
    },
    {
      "entropy": 5.756705331802368,
      "epoch": 2.543785255786812,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043482639400670576,
      "loss": 5.0321,
      "mean_token_accuracy": 0.19947439581155776,
      "num_tokens": 56733075.0,
      "step": 32695
    },
    {
      "entropy": 5.671201944351196,
      "epoch": 2.5441742851585296,
      "grad_norm": 1.1875,
      "learning_rate": 0.000434806884923248,
      "loss": 4.92,
      "mean_token_accuracy": 0.2153444766998291,
      "num_tokens": 56742184.0,
      "step": 32700
    },
    {
      "entropy": 5.7458521842956545,
      "epoch": 2.5445633145302473,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004347873734149752,
      "loss": 4.9093,
      "mean_token_accuracy": 0.20738874226808549,
      "num_tokens": 56750559.0,
      "step": 32705
    },
    {
      "entropy": 5.769989919662476,
      "epoch": 2.5449523439019646,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043476785948218325,
      "loss": 5.0645,
      "mean_token_accuracy": 0.19844296723604202,
      "num_tokens": 56760031.0,
      "step": 32710
    },
    {
      "entropy": 5.7385557174682615,
      "epoch": 2.5453413732736823,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043474834312516835,
      "loss": 5.0385,
      "mean_token_accuracy": 0.20062004625797272,
      "num_tokens": 56768496.0,
      "step": 32715
    },
    {
      "entropy": 5.751664781570435,
      "epoch": 2.5457304026453995,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004347288243442266,
      "loss": 5.0049,
      "mean_token_accuracy": 0.2069281443953514,
      "num_tokens": 56776876.0,
      "step": 32720
    },
    {
      "entropy": 5.764461994171143,
      "epoch": 2.546119432017117,
      "grad_norm": 1.25,
      "learning_rate": 0.0004347093031396543,
      "loss": 4.9753,
      "mean_token_accuracy": 0.2123896673321724,
      "num_tokens": 56785661.0,
      "step": 32725
    },
    {
      "entropy": 5.71371374130249,
      "epoch": 2.546508461388835,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004346897795117474,
      "loss": 4.9641,
      "mean_token_accuracy": 0.20882887095212938,
      "num_tokens": 56794818.0,
      "step": 32730
    },
    {
      "entropy": 5.712417078018189,
      "epoch": 2.5468974907605526,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004346702534608023,
      "loss": 5.013,
      "mean_token_accuracy": 0.19923704415559768,
      "num_tokens": 56804528.0,
      "step": 32735
    },
    {
      "entropy": 5.795131254196167,
      "epoch": 2.54728652013227,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004346507249871153,
      "loss": 4.9839,
      "mean_token_accuracy": 0.2123407244682312,
      "num_tokens": 56812766.0,
      "step": 32740
    },
    {
      "entropy": 5.75357027053833,
      "epoch": 2.5476755495039876,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004346311940909826,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2088528424501419,
      "num_tokens": 56821108.0,
      "step": 32745
    },
    {
      "entropy": 5.772972297668457,
      "epoch": 2.5480645788757053,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043461166077270056,
      "loss": 5.0729,
      "mean_token_accuracy": 0.19937997162342072,
      "num_tokens": 56829060.0,
      "step": 32750
    },
    {
      "entropy": 5.791413116455078,
      "epoch": 2.5484536082474225,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043459212503256566,
      "loss": 5.0342,
      "mean_token_accuracy": 0.20933077186346055,
      "num_tokens": 56837440.0,
      "step": 32755
    },
    {
      "entropy": 5.790433263778686,
      "epoch": 2.54884263761914,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004345725868708743,
      "loss": 5.0006,
      "mean_token_accuracy": 0.2121829867362976,
      "num_tokens": 56845485.0,
      "step": 32760
    },
    {
      "entropy": 5.775847339630127,
      "epoch": 2.549231666990858,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043455304628792294,
      "loss": 4.977,
      "mean_token_accuracy": 0.20994850844144822,
      "num_tokens": 56854498.0,
      "step": 32765
    },
    {
      "entropy": 5.705957412719727,
      "epoch": 2.5496206963625756,
      "grad_norm": 1.203125,
      "learning_rate": 0.000434533503284008,
      "loss": 4.9742,
      "mean_token_accuracy": 0.2112821877002716,
      "num_tokens": 56862910.0,
      "step": 32770
    },
    {
      "entropy": 5.729395723342895,
      "epoch": 2.550009725734293,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004345139578594261,
      "loss": 4.9683,
      "mean_token_accuracy": 0.20883767008781434,
      "num_tokens": 56872235.0,
      "step": 32775
    },
    {
      "entropy": 5.7738926887512205,
      "epoch": 2.5503987551060106,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004344944100144738,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20581137388944626,
      "num_tokens": 56881506.0,
      "step": 32780
    },
    {
      "entropy": 5.76079535484314,
      "epoch": 2.550787784477728,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043447485974944765,
      "loss": 4.9686,
      "mean_token_accuracy": 0.21169120222330093,
      "num_tokens": 56890218.0,
      "step": 32785
    },
    {
      "entropy": 5.748682451248169,
      "epoch": 2.5511768138494455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043445530706464437,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20601194947957993,
      "num_tokens": 56898037.0,
      "step": 32790
    },
    {
      "entropy": 5.664990377426148,
      "epoch": 2.551565843221163,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004344357519603606,
      "loss": 4.9091,
      "mean_token_accuracy": 0.206167571246624,
      "num_tokens": 56906892.0,
      "step": 32795
    },
    {
      "entropy": 5.701299953460693,
      "epoch": 2.551954872592881,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004344161944368931,
      "loss": 4.8287,
      "mean_token_accuracy": 0.21554235219955445,
      "num_tokens": 56915286.0,
      "step": 32800
    },
    {
      "entropy": 5.705565643310547,
      "epoch": 2.5523439019645986,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004343966344945385,
      "loss": 4.9436,
      "mean_token_accuracy": 0.20639772862195968,
      "num_tokens": 56923670.0,
      "step": 32805
    },
    {
      "entropy": 5.819622659683228,
      "epoch": 2.552732931336316,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043437707213359375,
      "loss": 5.0852,
      "mean_token_accuracy": 0.19876272827386857,
      "num_tokens": 56932092.0,
      "step": 32810
    },
    {
      "entropy": 5.75649061203003,
      "epoch": 2.5531219607080335,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043435750735435554,
      "loss": 5.0451,
      "mean_token_accuracy": 0.2101953238248825,
      "num_tokens": 56940938.0,
      "step": 32815
    },
    {
      "entropy": 5.786232614517212,
      "epoch": 2.553510990079751,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043433794015712085,
      "loss": 5.0153,
      "mean_token_accuracy": 0.20385306626558303,
      "num_tokens": 56950001.0,
      "step": 32820
    },
    {
      "entropy": 5.7990052700042725,
      "epoch": 2.5539000194514685,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004343183705421865,
      "loss": 5.0562,
      "mean_token_accuracy": 0.2047049656510353,
      "num_tokens": 56958898.0,
      "step": 32825
    },
    {
      "entropy": 5.772749280929565,
      "epoch": 2.554289048823186,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004342987985098495,
      "loss": 4.9686,
      "mean_token_accuracy": 0.2073871001601219,
      "num_tokens": 56967586.0,
      "step": 32830
    },
    {
      "entropy": 5.666131210327149,
      "epoch": 2.554678078194904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043427922406040676,
      "loss": 4.903,
      "mean_token_accuracy": 0.20844308882951737,
      "num_tokens": 56975781.0,
      "step": 32835
    },
    {
      "entropy": 5.811225652694702,
      "epoch": 2.555067107566621,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043425964719415525,
      "loss": 5.0413,
      "mean_token_accuracy": 0.20668812543153764,
      "num_tokens": 56984230.0,
      "step": 32840
    },
    {
      "entropy": 5.762484979629517,
      "epoch": 2.555456136938339,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043424006791139213,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20323142409324646,
      "num_tokens": 56992100.0,
      "step": 32845
    },
    {
      "entropy": 5.771996021270752,
      "epoch": 2.5558451663100565,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043422048621241443,
      "loss": 5.0785,
      "mean_token_accuracy": 0.20102738440036774,
      "num_tokens": 57000407.0,
      "step": 32850
    },
    {
      "entropy": 5.815082550048828,
      "epoch": 2.556234195681774,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043420090209751933,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20778660476207733,
      "num_tokens": 57008781.0,
      "step": 32855
    },
    {
      "entropy": 5.752675580978393,
      "epoch": 2.5566232250534915,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004341813155670039,
      "loss": 4.9474,
      "mean_token_accuracy": 0.21140379458665848,
      "num_tokens": 57017568.0,
      "step": 32860
    },
    {
      "entropy": 5.752951955795288,
      "epoch": 2.557012254425209,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004341617266211654,
      "loss": 4.8913,
      "mean_token_accuracy": 0.21710657179355622,
      "num_tokens": 57025674.0,
      "step": 32865
    },
    {
      "entropy": 5.732583332061767,
      "epoch": 2.557401283796927,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000434142135260301,
      "loss": 4.977,
      "mean_token_accuracy": 0.2106477662920952,
      "num_tokens": 57033657.0,
      "step": 32870
    },
    {
      "entropy": 5.729851198196411,
      "epoch": 2.557790313168644,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000434122541484708,
      "loss": 5.0705,
      "mean_token_accuracy": 0.20223496705293656,
      "num_tokens": 57043057.0,
      "step": 32875
    },
    {
      "entropy": 5.7974385738372805,
      "epoch": 2.558179342540362,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004341029452946837,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20021673142910004,
      "num_tokens": 57051721.0,
      "step": 32880
    },
    {
      "entropy": 5.822482109069824,
      "epoch": 2.558568371912079,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004340833466905254,
      "loss": 5.0184,
      "mean_token_accuracy": 0.20185206532478334,
      "num_tokens": 57060223.0,
      "step": 32885
    },
    {
      "entropy": 5.797485303878784,
      "epoch": 2.558957401283797,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043406374567253055,
      "loss": 5.0068,
      "mean_token_accuracy": 0.20618628412485124,
      "num_tokens": 57068408.0,
      "step": 32890
    },
    {
      "entropy": 5.776030588150024,
      "epoch": 2.5593464306555145,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004340441422409965,
      "loss": 5.0757,
      "mean_token_accuracy": 0.1946803167462349,
      "num_tokens": 57076784.0,
      "step": 32895
    },
    {
      "entropy": 5.79780912399292,
      "epoch": 2.559735460027232,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004340245363962209,
      "loss": 5.0363,
      "mean_token_accuracy": 0.20086259245872498,
      "num_tokens": 57084745.0,
      "step": 32900
    },
    {
      "entropy": 5.709837102890015,
      "epoch": 2.56012448939895,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043400492813850083,
      "loss": 4.9623,
      "mean_token_accuracy": 0.21104331612586974,
      "num_tokens": 57093395.0,
      "step": 32905
    },
    {
      "entropy": 5.681382465362549,
      "epoch": 2.560513518770667,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043398531746813426,
      "loss": 4.9316,
      "mean_token_accuracy": 0.20191646963357926,
      "num_tokens": 57101665.0,
      "step": 32910
    },
    {
      "entropy": 5.6577447891235355,
      "epoch": 2.560902548142385,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043396570438541845,
      "loss": 4.9092,
      "mean_token_accuracy": 0.20565275102853775,
      "num_tokens": 57110328.0,
      "step": 32915
    },
    {
      "entropy": 5.756406545639038,
      "epoch": 2.561291577514102,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004339460888906512,
      "loss": 5.03,
      "mean_token_accuracy": 0.19820039123296737,
      "num_tokens": 57118834.0,
      "step": 32920
    },
    {
      "entropy": 5.852702999114991,
      "epoch": 2.5616806068858198,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004339264709841299,
      "loss": 5.1006,
      "mean_token_accuracy": 0.19865436851978302,
      "num_tokens": 57127672.0,
      "step": 32925
    },
    {
      "entropy": 5.716109085083008,
      "epoch": 2.5620696362575375,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043390685066615244,
      "loss": 4.9501,
      "mean_token_accuracy": 0.21160395443439484,
      "num_tokens": 57136209.0,
      "step": 32930
    },
    {
      "entropy": 5.754217147827148,
      "epoch": 2.562458665629255,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004338872279370164,
      "loss": 4.9955,
      "mean_token_accuracy": 0.20592164546251296,
      "num_tokens": 57144427.0,
      "step": 32935
    },
    {
      "entropy": 5.793454504013061,
      "epoch": 2.5628476950009724,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004338676027970196,
      "loss": 5.119,
      "mean_token_accuracy": 0.19316186904907226,
      "num_tokens": 57152860.0,
      "step": 32940
    },
    {
      "entropy": 5.82431321144104,
      "epoch": 2.56323672437269,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043384797524645977,
      "loss": 4.9901,
      "mean_token_accuracy": 0.20951569825410843,
      "num_tokens": 57162397.0,
      "step": 32945
    },
    {
      "entropy": 5.759331130981446,
      "epoch": 2.563625753744408,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004338283452856348,
      "loss": 4.968,
      "mean_token_accuracy": 0.21310473680496217,
      "num_tokens": 57171178.0,
      "step": 32950
    },
    {
      "entropy": 5.784869050979614,
      "epoch": 2.564014783116125,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004338087129148425,
      "loss": 4.9861,
      "mean_token_accuracy": 0.2079272076487541,
      "num_tokens": 57178715.0,
      "step": 32955
    },
    {
      "entropy": 5.828079080581665,
      "epoch": 2.5644038124878428,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043378907813438066,
      "loss": 5.0604,
      "mean_token_accuracy": 0.19179490357637405,
      "num_tokens": 57187239.0,
      "step": 32960
    },
    {
      "entropy": 5.775489044189453,
      "epoch": 2.5647928418595605,
      "grad_norm": 1.25,
      "learning_rate": 0.00043376944094454734,
      "loss": 4.9999,
      "mean_token_accuracy": 0.19840618520975112,
      "num_tokens": 57195655.0,
      "step": 32965
    },
    {
      "entropy": 5.7567229747772215,
      "epoch": 2.565181871231278,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043374980134564044,
      "loss": 5.0157,
      "mean_token_accuracy": 0.19842608273029327,
      "num_tokens": 57205026.0,
      "step": 32970
    },
    {
      "entropy": 5.743293380737304,
      "epoch": 2.5655709006029954,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043373015933795804,
      "loss": 4.9893,
      "mean_token_accuracy": 0.20903294980525972,
      "num_tokens": 57213344.0,
      "step": 32975
    },
    {
      "entropy": 5.796473693847656,
      "epoch": 2.565959929974713,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004337105149217981,
      "loss": 4.9733,
      "mean_token_accuracy": 0.2071865677833557,
      "num_tokens": 57222174.0,
      "step": 32980
    },
    {
      "entropy": 5.713343334197998,
      "epoch": 2.566348959346431,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043369086809745875,
      "loss": 4.9915,
      "mean_token_accuracy": 0.20441901236772536,
      "num_tokens": 57231099.0,
      "step": 32985
    },
    {
      "entropy": 5.7745014190673825,
      "epoch": 2.566737988718148,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043367121886523795,
      "loss": 5.0214,
      "mean_token_accuracy": 0.1989772692322731,
      "num_tokens": 57239143.0,
      "step": 32990
    },
    {
      "entropy": 5.727962303161621,
      "epoch": 2.5671270180898658,
      "grad_norm": 1.171875,
      "learning_rate": 0.000433651567225434,
      "loss": 5.003,
      "mean_token_accuracy": 0.205684033036232,
      "num_tokens": 57247971.0,
      "step": 32995
    },
    {
      "entropy": 5.769332408905029,
      "epoch": 2.5675160474615835,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004336319131783451,
      "loss": 4.9918,
      "mean_token_accuracy": 0.2089997798204422,
      "num_tokens": 57256257.0,
      "step": 33000
    },
    {
      "epoch": 2.5675160474615835,
      "eval_entropy": 5.521635795993456,
      "eval_loss": 5.075793266296387,
      "eval_mean_token_accuracy": 0.21213480656369588,
      "eval_num_tokens": 57256257.0,
      "eval_runtime": 21.6325,
      "eval_samples_per_second": 1921.088,
      "eval_steps_per_second": 240.148,
      "step": 33000
    },
    {
      "entropy": 5.786584091186524,
      "epoch": 2.567905076833301,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043361225672426933,
      "loss": 5.0105,
      "mean_token_accuracy": 0.20311107337474824,
      "num_tokens": 57264729.0,
      "step": 33005
    },
    {
      "entropy": 5.787122440338135,
      "epoch": 2.5682941062050184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004335925978635051,
      "loss": 5.0551,
      "mean_token_accuracy": 0.20045190155506135,
      "num_tokens": 57273370.0,
      "step": 33010
    },
    {
      "entropy": 5.767630481719971,
      "epoch": 2.568683135576736,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043357293659635057,
      "loss": 5.0319,
      "mean_token_accuracy": 0.205415678024292,
      "num_tokens": 57282466.0,
      "step": 33015
    },
    {
      "entropy": 5.735296058654785,
      "epoch": 2.5690721649484534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004335532729231041,
      "loss": 4.975,
      "mean_token_accuracy": 0.20214279890060424,
      "num_tokens": 57291015.0,
      "step": 33020
    },
    {
      "entropy": 5.806283187866211,
      "epoch": 2.569461194320171,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043353360684406415,
      "loss": 5.0644,
      "mean_token_accuracy": 0.2003934934735298,
      "num_tokens": 57300941.0,
      "step": 33025
    },
    {
      "entropy": 5.8102943897247314,
      "epoch": 2.5698502236918888,
      "grad_norm": 1.359375,
      "learning_rate": 0.000433513938359529,
      "loss": 5.0518,
      "mean_token_accuracy": 0.19729073196649552,
      "num_tokens": 57310412.0,
      "step": 33030
    },
    {
      "entropy": 5.8451512336730955,
      "epoch": 2.5702392530636065,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004334942674697971,
      "loss": 5.0955,
      "mean_token_accuracy": 0.204061596095562,
      "num_tokens": 57319262.0,
      "step": 33035
    },
    {
      "entropy": 5.769824361801147,
      "epoch": 2.5706282824353237,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043347459417516696,
      "loss": 5.0367,
      "mean_token_accuracy": 0.20008276253938675,
      "num_tokens": 57327279.0,
      "step": 33040
    },
    {
      "entropy": 5.7166553974151615,
      "epoch": 2.5710173118070414,
      "grad_norm": 1.25,
      "learning_rate": 0.0004334549184759371,
      "loss": 4.9083,
      "mean_token_accuracy": 0.21907222419977188,
      "num_tokens": 57335498.0,
      "step": 33045
    },
    {
      "entropy": 5.712505149841308,
      "epoch": 2.571406341178759,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004334352403724062,
      "loss": 5.0651,
      "mean_token_accuracy": 0.20013367086648942,
      "num_tokens": 57345134.0,
      "step": 33050
    },
    {
      "entropy": 5.771085977554321,
      "epoch": 2.5717953705504764,
      "grad_norm": 1.25,
      "learning_rate": 0.00043341555986487254,
      "loss": 5.0833,
      "mean_token_accuracy": 0.2036922663450241,
      "num_tokens": 57353451.0,
      "step": 33055
    },
    {
      "entropy": 5.791321229934693,
      "epoch": 2.572184399922194,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004333958769536349,
      "loss": 5.0212,
      "mean_token_accuracy": 0.20156901478767394,
      "num_tokens": 57362990.0,
      "step": 33060
    },
    {
      "entropy": 5.78676176071167,
      "epoch": 2.5725734292939118,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004333761916389921,
      "loss": 4.9869,
      "mean_token_accuracy": 0.20758597254753114,
      "num_tokens": 57371572.0,
      "step": 33065
    },
    {
      "entropy": 5.656745052337646,
      "epoch": 2.5729624586656294,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004333565039212425,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2067430466413498,
      "num_tokens": 57380558.0,
      "step": 33070
    },
    {
      "entropy": 5.691905212402344,
      "epoch": 2.5733514880373467,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004333368138006851,
      "loss": 4.8769,
      "mean_token_accuracy": 0.21097035855054855,
      "num_tokens": 57389636.0,
      "step": 33075
    },
    {
      "entropy": 5.73447847366333,
      "epoch": 2.5737405174090644,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004333171212776185,
      "loss": 4.9368,
      "mean_token_accuracy": 0.21167907118797302,
      "num_tokens": 57397546.0,
      "step": 33080
    },
    {
      "entropy": 5.793683099746704,
      "epoch": 2.574129546780782,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004332974263523416,
      "loss": 5.0973,
      "mean_token_accuracy": 0.20549457222223283,
      "num_tokens": 57406981.0,
      "step": 33085
    },
    {
      "entropy": 5.799549293518067,
      "epoch": 2.5745185761524993,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043327772902515327,
      "loss": 5.0448,
      "mean_token_accuracy": 0.2054416060447693,
      "num_tokens": 57416016.0,
      "step": 33090
    },
    {
      "entropy": 5.796039152145386,
      "epoch": 2.574907605524217,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004332580292963523,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19865683764219283,
      "num_tokens": 57425086.0,
      "step": 33095
    },
    {
      "entropy": 5.806149291992187,
      "epoch": 2.5752966348959347,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004332383271662376,
      "loss": 5.0395,
      "mean_token_accuracy": 0.2071179047226906,
      "num_tokens": 57433982.0,
      "step": 33100
    },
    {
      "entropy": 5.7628051280975345,
      "epoch": 2.5756856642676524,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043321862263510816,
      "loss": 4.8917,
      "mean_token_accuracy": 0.2116219624876976,
      "num_tokens": 57442922.0,
      "step": 33105
    },
    {
      "entropy": 5.774821043014526,
      "epoch": 2.5760746936393697,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004331989157032629,
      "loss": 5.0577,
      "mean_token_accuracy": 0.20238651484251022,
      "num_tokens": 57452373.0,
      "step": 33110
    },
    {
      "entropy": 5.761926794052124,
      "epoch": 2.5764637230110874,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043317920637100097,
      "loss": 4.9753,
      "mean_token_accuracy": 0.20934057533740996,
      "num_tokens": 57460741.0,
      "step": 33115
    },
    {
      "entropy": 5.717100667953491,
      "epoch": 2.5768527523828046,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004331594946386213,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20892391949892045,
      "num_tokens": 57469488.0,
      "step": 33120
    },
    {
      "entropy": 5.7583746910095215,
      "epoch": 2.5772417817545223,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043313978050642303,
      "loss": 5.0213,
      "mean_token_accuracy": 0.19943802803754807,
      "num_tokens": 57478560.0,
      "step": 33125
    },
    {
      "entropy": 5.744273948669433,
      "epoch": 2.57763081112624,
      "grad_norm": 1.25,
      "learning_rate": 0.0004331200639747053,
      "loss": 4.9279,
      "mean_token_accuracy": 0.2104368641972542,
      "num_tokens": 57486670.0,
      "step": 33130
    },
    {
      "entropy": 5.7799601554870605,
      "epoch": 2.5780198404979577,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004331003450437674,
      "loss": 5.0255,
      "mean_token_accuracy": 0.20352811366319656,
      "num_tokens": 57495676.0,
      "step": 33135
    },
    {
      "entropy": 5.702581977844238,
      "epoch": 2.5784088698696754,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004330806237139083,
      "loss": 4.9184,
      "mean_token_accuracy": 0.21573910862207413,
      "num_tokens": 57504422.0,
      "step": 33140
    },
    {
      "entropy": 5.748302602767945,
      "epoch": 2.5787978992413927,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004330608999854273,
      "loss": 4.9225,
      "mean_token_accuracy": 0.20659206956624984,
      "num_tokens": 57514555.0,
      "step": 33145
    },
    {
      "entropy": 5.725775241851807,
      "epoch": 2.5791869286131104,
      "grad_norm": 1.25,
      "learning_rate": 0.0004330411738586238,
      "loss": 4.8486,
      "mean_token_accuracy": 0.21359531581401825,
      "num_tokens": 57523088.0,
      "step": 33150
    },
    {
      "entropy": 5.678217506408691,
      "epoch": 2.5795759579848276,
      "grad_norm": 1.3125,
      "learning_rate": 0.000433021445333797,
      "loss": 4.9741,
      "mean_token_accuracy": 0.20969132483005523,
      "num_tokens": 57530890.0,
      "step": 33155
    },
    {
      "entropy": 5.769487190246582,
      "epoch": 2.5799649873565453,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043300171441124633,
      "loss": 5.0251,
      "mean_token_accuracy": 0.20796672999858856,
      "num_tokens": 57539834.0,
      "step": 33160
    },
    {
      "entropy": 5.7724262237548825,
      "epoch": 2.580354016728263,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004329819810912711,
      "loss": 4.994,
      "mean_token_accuracy": 0.20428797751665115,
      "num_tokens": 57548962.0,
      "step": 33165
    },
    {
      "entropy": 5.69587664604187,
      "epoch": 2.5807430460999807,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043296224537417075,
      "loss": 4.8795,
      "mean_token_accuracy": 0.2207977995276451,
      "num_tokens": 57557460.0,
      "step": 33170
    },
    {
      "entropy": 5.707780504226685,
      "epoch": 2.581132075471698,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043294250726024473,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2213783472776413,
      "num_tokens": 57566053.0,
      "step": 33175
    },
    {
      "entropy": 5.731279039382935,
      "epoch": 2.5815211048434157,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004329227667497926,
      "loss": 4.9496,
      "mean_token_accuracy": 0.20581699162721634,
      "num_tokens": 57575268.0,
      "step": 33180
    },
    {
      "entropy": 5.7321959972381595,
      "epoch": 2.5819101342151334,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043290302384311373,
      "loss": 4.9311,
      "mean_token_accuracy": 0.20782824158668517,
      "num_tokens": 57583518.0,
      "step": 33185
    },
    {
      "entropy": 5.690702342987061,
      "epoch": 2.5822991635868506,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043288327854050794,
      "loss": 4.913,
      "mean_token_accuracy": 0.2113048955798149,
      "num_tokens": 57592300.0,
      "step": 33190
    },
    {
      "entropy": 5.741007900238037,
      "epoch": 2.5826881929585683,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043286353084227467,
      "loss": 5.0721,
      "mean_token_accuracy": 0.20033103376626968,
      "num_tokens": 57600694.0,
      "step": 33195
    },
    {
      "entropy": 5.842280292510987,
      "epoch": 2.583077222330286,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043284378074871354,
      "loss": 5.0731,
      "mean_token_accuracy": 0.20883223414421082,
      "num_tokens": 57609213.0,
      "step": 33200
    },
    {
      "entropy": 5.753599309921265,
      "epoch": 2.5834662517020037,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004328240282601243,
      "loss": 4.989,
      "mean_token_accuracy": 0.20347818583250046,
      "num_tokens": 57617466.0,
      "step": 33205
    },
    {
      "entropy": 5.704377317428589,
      "epoch": 2.583855281073721,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004328042733768066,
      "loss": 5.0037,
      "mean_token_accuracy": 0.2028016373515129,
      "num_tokens": 57626620.0,
      "step": 33210
    },
    {
      "entropy": 5.860623788833618,
      "epoch": 2.5842443104454387,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043278451609906027,
      "loss": 5.0371,
      "mean_token_accuracy": 0.20837210565805436,
      "num_tokens": 57635230.0,
      "step": 33215
    },
    {
      "entropy": 5.810592746734619,
      "epoch": 2.584633339817156,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043276475642718503,
      "loss": 5.0409,
      "mean_token_accuracy": 0.2046953797340393,
      "num_tokens": 57644444.0,
      "step": 33220
    },
    {
      "entropy": 5.7788252353668215,
      "epoch": 2.5850223691888736,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004327449943614808,
      "loss": 5.0389,
      "mean_token_accuracy": 0.2011866971850395,
      "num_tokens": 57653282.0,
      "step": 33225
    },
    {
      "entropy": 5.718972110748291,
      "epoch": 2.5854113985605913,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043272522990224727,
      "loss": 5.0094,
      "mean_token_accuracy": 0.20761703103780746,
      "num_tokens": 57661153.0,
      "step": 33230
    },
    {
      "entropy": 5.775798988342285,
      "epoch": 2.585800427932309,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004327054630497844,
      "loss": 5.0705,
      "mean_token_accuracy": 0.19694240838289262,
      "num_tokens": 57669613.0,
      "step": 33235
    },
    {
      "entropy": 5.790027236938476,
      "epoch": 2.5861894573040267,
      "grad_norm": 1.25,
      "learning_rate": 0.00043268569380439223,
      "loss": 5.0282,
      "mean_token_accuracy": 0.20255554467439651,
      "num_tokens": 57678174.0,
      "step": 33240
    },
    {
      "entropy": 5.754463052749633,
      "epoch": 2.586578486675744,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004326659221663705,
      "loss": 5.0079,
      "mean_token_accuracy": 0.2098563000559807,
      "num_tokens": 57687117.0,
      "step": 33245
    },
    {
      "entropy": 5.804340887069702,
      "epoch": 2.5869675160474617,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004326461481360194,
      "loss": 5.0748,
      "mean_token_accuracy": 0.2016873374581337,
      "num_tokens": 57695073.0,
      "step": 33250
    },
    {
      "entropy": 5.711058855056763,
      "epoch": 2.587356545419179,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043262637171363906,
      "loss": 4.8802,
      "mean_token_accuracy": 0.21483015567064284,
      "num_tokens": 57703856.0,
      "step": 33255
    },
    {
      "entropy": 5.799820613861084,
      "epoch": 2.5877455747908966,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043260659289952926,
      "loss": 4.8863,
      "mean_token_accuracy": 0.21601614505052566,
      "num_tokens": 57712072.0,
      "step": 33260
    },
    {
      "entropy": 5.721677827835083,
      "epoch": 2.5881346041626143,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043258681169399034,
      "loss": 5.0077,
      "mean_token_accuracy": 0.20544838309288024,
      "num_tokens": 57721203.0,
      "step": 33265
    },
    {
      "entropy": 5.810277843475342,
      "epoch": 2.588523633534332,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043256702809732234,
      "loss": 5.0993,
      "mean_token_accuracy": 0.19455747902393342,
      "num_tokens": 57730591.0,
      "step": 33270
    },
    {
      "entropy": 5.801410913467407,
      "epoch": 2.5889126629060493,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004325472421098255,
      "loss": 4.9941,
      "mean_token_accuracy": 0.20934539288282394,
      "num_tokens": 57739390.0,
      "step": 33275
    },
    {
      "entropy": 5.740540313720703,
      "epoch": 2.589301692277767,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043252745373180006,
      "loss": 4.9476,
      "mean_token_accuracy": 0.20520153790712356,
      "num_tokens": 57748008.0,
      "step": 33280
    },
    {
      "entropy": 5.8652002811431885,
      "epoch": 2.5896907216494847,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004325076629635462,
      "loss": 5.2114,
      "mean_token_accuracy": 0.18855730444192886,
      "num_tokens": 57757471.0,
      "step": 33285
    },
    {
      "entropy": 5.80700159072876,
      "epoch": 2.590079751021202,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043248786980536424,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20105296075344087,
      "num_tokens": 57766534.0,
      "step": 33290
    },
    {
      "entropy": 5.705951166152954,
      "epoch": 2.5904687803929196,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004324680742575545,
      "loss": 4.8913,
      "mean_token_accuracy": 0.2158153012394905,
      "num_tokens": 57775011.0,
      "step": 33295
    },
    {
      "entropy": 5.746581029891968,
      "epoch": 2.5908578097646373,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043244827632041744,
      "loss": 5.0421,
      "mean_token_accuracy": 0.20172712206840515,
      "num_tokens": 57783606.0,
      "step": 33300
    },
    {
      "entropy": 5.762721872329712,
      "epoch": 2.591246839136355,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004324284759942534,
      "loss": 4.9576,
      "mean_token_accuracy": 0.2127518355846405,
      "num_tokens": 57792170.0,
      "step": 33305
    },
    {
      "entropy": 5.685492706298828,
      "epoch": 2.5916358685080723,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004324086732793627,
      "loss": 4.9068,
      "mean_token_accuracy": 0.20639525800943376,
      "num_tokens": 57801379.0,
      "step": 33310
    },
    {
      "entropy": 5.7005432605743405,
      "epoch": 2.59202489787979,
      "grad_norm": 1.25,
      "learning_rate": 0.00043238886817604605,
      "loss": 4.9954,
      "mean_token_accuracy": 0.20646542310714722,
      "num_tokens": 57810135.0,
      "step": 33315
    },
    {
      "entropy": 5.708997964859009,
      "epoch": 2.5924139272515077,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004323690606846039,
      "loss": 4.9409,
      "mean_token_accuracy": 0.20837097316980363,
      "num_tokens": 57818314.0,
      "step": 33320
    },
    {
      "entropy": 5.752008104324341,
      "epoch": 2.592802956623225,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043234925080533657,
      "loss": 4.9981,
      "mean_token_accuracy": 0.2048935040831566,
      "num_tokens": 57826883.0,
      "step": 33325
    },
    {
      "entropy": 5.728763103485107,
      "epoch": 2.5931919859949426,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004323294385385448,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2116884931921959,
      "num_tokens": 57835215.0,
      "step": 33330
    },
    {
      "entropy": 5.692485666275024,
      "epoch": 2.5935810153666603,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004323096238845293,
      "loss": 4.9865,
      "mean_token_accuracy": 0.21171266585588455,
      "num_tokens": 57844002.0,
      "step": 33335
    },
    {
      "entropy": 5.79583773612976,
      "epoch": 2.593970044738378,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004322898068435906,
      "loss": 5.0458,
      "mean_token_accuracy": 0.1998705506324768,
      "num_tokens": 57852453.0,
      "step": 33340
    },
    {
      "entropy": 5.757982778549194,
      "epoch": 2.5943590741100953,
      "grad_norm": 1.25,
      "learning_rate": 0.0004322699874160294,
      "loss": 4.9272,
      "mean_token_accuracy": 0.20676076859235765,
      "num_tokens": 57860807.0,
      "step": 33345
    },
    {
      "entropy": 5.772418975830078,
      "epoch": 2.594748103481813,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043225016560214654,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20712264776229858,
      "num_tokens": 57869768.0,
      "step": 33350
    },
    {
      "entropy": 5.809991312026978,
      "epoch": 2.59513713285353,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043223034140224266,
      "loss": 5.0156,
      "mean_token_accuracy": 0.20447915196418762,
      "num_tokens": 57878089.0,
      "step": 33355
    },
    {
      "entropy": 5.754279041290284,
      "epoch": 2.595526162225248,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004322105148166187,
      "loss": 5.0198,
      "mean_token_accuracy": 0.20274076014757156,
      "num_tokens": 57886468.0,
      "step": 33360
    },
    {
      "entropy": 5.7886889457702635,
      "epoch": 2.5959151915969656,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043219068584557526,
      "loss": 5.0758,
      "mean_token_accuracy": 0.19407573491334915,
      "num_tokens": 57895993.0,
      "step": 33365
    },
    {
      "entropy": 5.747131490707398,
      "epoch": 2.5963042209686833,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004321708544894134,
      "loss": 4.9244,
      "mean_token_accuracy": 0.2132062003016472,
      "num_tokens": 57904464.0,
      "step": 33370
    },
    {
      "entropy": 5.720435762405396,
      "epoch": 2.5966932503404005,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043215102074843403,
      "loss": 5.0508,
      "mean_token_accuracy": 0.2010599195957184,
      "num_tokens": 57912711.0,
      "step": 33375
    },
    {
      "entropy": 5.719974136352539,
      "epoch": 2.5970822797121182,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043213118462293796,
      "loss": 4.988,
      "mean_token_accuracy": 0.20771399587392808,
      "num_tokens": 57920926.0,
      "step": 33380
    },
    {
      "entropy": 5.760038995742798,
      "epoch": 2.597471309083836,
      "grad_norm": 1.25,
      "learning_rate": 0.0004321113461132264,
      "loss": 4.9448,
      "mean_token_accuracy": 0.21141133904457093,
      "num_tokens": 57930346.0,
      "step": 33385
    },
    {
      "entropy": 5.78417558670044,
      "epoch": 2.597860338455553,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043209150521960016,
      "loss": 5.0282,
      "mean_token_accuracy": 0.21219132244586944,
      "num_tokens": 57938578.0,
      "step": 33390
    },
    {
      "entropy": 5.7820686340332035,
      "epoch": 2.598249367827271,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043207166194236037,
      "loss": 5.1018,
      "mean_token_accuracy": 0.19686743766069412,
      "num_tokens": 57947505.0,
      "step": 33395
    },
    {
      "entropy": 5.795369291305542,
      "epoch": 2.5986383971989886,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004320518162818082,
      "loss": 4.9529,
      "mean_token_accuracy": 0.20554561614990235,
      "num_tokens": 57955806.0,
      "step": 33400
    },
    {
      "entropy": 5.803036880493164,
      "epoch": 2.5990274265707063,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004320319682382445,
      "loss": 5.1363,
      "mean_token_accuracy": 0.1929910734295845,
      "num_tokens": 57964275.0,
      "step": 33405
    },
    {
      "entropy": 5.758026742935181,
      "epoch": 2.5994164559424235,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043201211781197075,
      "loss": 5.034,
      "mean_token_accuracy": 0.21025671511888505,
      "num_tokens": 57973616.0,
      "step": 33410
    },
    {
      "entropy": 5.804161071777344,
      "epoch": 2.5998054853141412,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043199226500328805,
      "loss": 5.102,
      "mean_token_accuracy": 0.20368774682283403,
      "num_tokens": 57982809.0,
      "step": 33415
    },
    {
      "entropy": 5.789568996429443,
      "epoch": 2.600194514685859,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004319724098124975,
      "loss": 4.9285,
      "mean_token_accuracy": 0.21287197172641753,
      "num_tokens": 57991239.0,
      "step": 33420
    },
    {
      "entropy": 5.779268646240235,
      "epoch": 2.600583544057576,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043195255223990055,
      "loss": 4.9898,
      "mean_token_accuracy": 0.21082137972116471,
      "num_tokens": 57999213.0,
      "step": 33425
    },
    {
      "entropy": 5.761600017547607,
      "epoch": 2.600972573429294,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004319326922857984,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20612243711948394,
      "num_tokens": 58008289.0,
      "step": 33430
    },
    {
      "entropy": 5.716359663009643,
      "epoch": 2.6013616028010116,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004319128299504925,
      "loss": 4.9623,
      "mean_token_accuracy": 0.2100016176700592,
      "num_tokens": 58017495.0,
      "step": 33435
    },
    {
      "entropy": 5.756489610671997,
      "epoch": 2.6017506321727293,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043189296523428407,
      "loss": 5.0276,
      "mean_token_accuracy": 0.20782245695590973,
      "num_tokens": 58026368.0,
      "step": 33440
    },
    {
      "entropy": 5.794899606704712,
      "epoch": 2.6021396615444465,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043187309813747464,
      "loss": 5.0744,
      "mean_token_accuracy": 0.19933004081249237,
      "num_tokens": 58035086.0,
      "step": 33445
    },
    {
      "entropy": 5.708502388000488,
      "epoch": 2.6025286909161642,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043185322866036565,
      "loss": 4.9838,
      "mean_token_accuracy": 0.20658310502767563,
      "num_tokens": 58044050.0,
      "step": 33450
    },
    {
      "entropy": 5.742864894866943,
      "epoch": 2.6029177202878815,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043183335680325866,
      "loss": 4.9529,
      "mean_token_accuracy": 0.2101571872830391,
      "num_tokens": 58052257.0,
      "step": 33455
    },
    {
      "entropy": 5.849968194961548,
      "epoch": 2.603306749659599,
      "grad_norm": 1.25,
      "learning_rate": 0.00043181348256645497,
      "loss": 5.1124,
      "mean_token_accuracy": 0.2022900626063347,
      "num_tokens": 58060742.0,
      "step": 33460
    },
    {
      "entropy": 5.741953182220459,
      "epoch": 2.603695779031317,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043179360595025637,
      "loss": 4.9918,
      "mean_token_accuracy": 0.19760675281286239,
      "num_tokens": 58069697.0,
      "step": 33465
    },
    {
      "entropy": 5.751530027389526,
      "epoch": 2.6040848084030346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004317737269549643,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21261819154024125,
      "num_tokens": 58077875.0,
      "step": 33470
    },
    {
      "entropy": 5.821722984313965,
      "epoch": 2.604473837774752,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004317538455808805,
      "loss": 5.0737,
      "mean_token_accuracy": 0.20578673630952835,
      "num_tokens": 58086500.0,
      "step": 33475
    },
    {
      "entropy": 5.732499790191651,
      "epoch": 2.6048628671464695,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004317339618283065,
      "loss": 4.9823,
      "mean_token_accuracy": 0.21007787585258483,
      "num_tokens": 58095851.0,
      "step": 33480
    },
    {
      "entropy": 5.8182457447052,
      "epoch": 2.6052518965181872,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004317140756975442,
      "loss": 5.009,
      "mean_token_accuracy": 0.20688370615243912,
      "num_tokens": 58104462.0,
      "step": 33485
    },
    {
      "entropy": 5.828068256378174,
      "epoch": 2.6056409258899045,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004316941871888951,
      "loss": 5.0676,
      "mean_token_accuracy": 0.19650615155696868,
      "num_tokens": 58112698.0,
      "step": 33490
    },
    {
      "entropy": 5.8249047756195065,
      "epoch": 2.606029955261622,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043167429630266134,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20694387704133987,
      "num_tokens": 58121807.0,
      "step": 33495
    },
    {
      "entropy": 5.727827405929565,
      "epoch": 2.60641898463334,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004316544030391444,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21072512716054917,
      "num_tokens": 58129824.0,
      "step": 33500
    },
    {
      "entropy": 5.806965303421021,
      "epoch": 2.6068080140050576,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004316345073986461,
      "loss": 5.0071,
      "mean_token_accuracy": 0.21447204947471618,
      "num_tokens": 58138083.0,
      "step": 33505
    },
    {
      "entropy": 5.77633376121521,
      "epoch": 2.607197043376775,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004316146093814686,
      "loss": 4.961,
      "mean_token_accuracy": 0.21852541267871856,
      "num_tokens": 58146631.0,
      "step": 33510
    },
    {
      "entropy": 5.782254409790039,
      "epoch": 2.6075860727484925,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004315947089879137,
      "loss": 4.9903,
      "mean_token_accuracy": 0.21014781892299653,
      "num_tokens": 58154867.0,
      "step": 33515
    },
    {
      "entropy": 5.724263906478882,
      "epoch": 2.60797510212021,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043157480621828327,
      "loss": 5.026,
      "mean_token_accuracy": 0.2025931030511856,
      "num_tokens": 58163887.0,
      "step": 33520
    },
    {
      "entropy": 5.698496723175049,
      "epoch": 2.6083641314919275,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043155490107287935,
      "loss": 4.8827,
      "mean_token_accuracy": 0.21423107534646987,
      "num_tokens": 58172731.0,
      "step": 33525
    },
    {
      "entropy": 5.684683895111084,
      "epoch": 2.608753160863645,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004315349935520041,
      "loss": 4.936,
      "mean_token_accuracy": 0.2159273609519005,
      "num_tokens": 58181619.0,
      "step": 33530
    },
    {
      "entropy": 5.727381706237793,
      "epoch": 2.609142190235363,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004315150836559594,
      "loss": 5.0223,
      "mean_token_accuracy": 0.20945177525281905,
      "num_tokens": 58190560.0,
      "step": 33535
    },
    {
      "entropy": 5.81807279586792,
      "epoch": 2.6095312196070806,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004314951713850474,
      "loss": 5.0222,
      "mean_token_accuracy": 0.20207540094852447,
      "num_tokens": 58199212.0,
      "step": 33540
    },
    {
      "entropy": 5.849285125732422,
      "epoch": 2.609920248978798,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043147525673957026,
      "loss": 5.005,
      "mean_token_accuracy": 0.20072565823793412,
      "num_tokens": 58207704.0,
      "step": 33545
    },
    {
      "entropy": 5.783019113540649,
      "epoch": 2.6103092783505155,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043145533971983025,
      "loss": 5.0711,
      "mean_token_accuracy": 0.19894753098487855,
      "num_tokens": 58217238.0,
      "step": 33550
    },
    {
      "entropy": 5.760096406936645,
      "epoch": 2.6106983077222328,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043143542032612935,
      "loss": 5.0327,
      "mean_token_accuracy": 0.20654432475566864,
      "num_tokens": 58226732.0,
      "step": 33555
    },
    {
      "entropy": 5.726895999908447,
      "epoch": 2.6110873370939505,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004314154985587701,
      "loss": 5.0085,
      "mean_token_accuracy": 0.21440177261829377,
      "num_tokens": 58234602.0,
      "step": 33560
    },
    {
      "entropy": 5.740339565277099,
      "epoch": 2.611476366465668,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043139557441805457,
      "loss": 5.0488,
      "mean_token_accuracy": 0.19463347792625427,
      "num_tokens": 58242557.0,
      "step": 33565
    },
    {
      "entropy": 5.750267601013183,
      "epoch": 2.611865395837386,
      "grad_norm": 1.40625,
      "learning_rate": 0.00043137564790428503,
      "loss": 5.0443,
      "mean_token_accuracy": 0.2065536007285118,
      "num_tokens": 58251805.0,
      "step": 33570
    },
    {
      "entropy": 5.748099613189697,
      "epoch": 2.6122544252091036,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043135571901776397,
      "loss": 4.9806,
      "mean_token_accuracy": 0.20032902657985688,
      "num_tokens": 58261241.0,
      "step": 33575
    },
    {
      "entropy": 5.797948265075684,
      "epoch": 2.612643454580821,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004313357877587937,
      "loss": 4.9848,
      "mean_token_accuracy": 0.2141013652086258,
      "num_tokens": 58269965.0,
      "step": 33580
    },
    {
      "entropy": 5.7683295726776125,
      "epoch": 2.6130324839525385,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004313158541276768,
      "loss": 4.9832,
      "mean_token_accuracy": 0.20454055219888687,
      "num_tokens": 58278420.0,
      "step": 33585
    },
    {
      "entropy": 5.730152702331543,
      "epoch": 2.6134215133242558,
      "grad_norm": 1.125,
      "learning_rate": 0.0004312959181247156,
      "loss": 4.9337,
      "mean_token_accuracy": 0.20773289501667022,
      "num_tokens": 58287384.0,
      "step": 33590
    },
    {
      "entropy": 5.727198600769043,
      "epoch": 2.6138105426959735,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004312759797502126,
      "loss": 4.9498,
      "mean_token_accuracy": 0.20301904231309892,
      "num_tokens": 58295312.0,
      "step": 33595
    },
    {
      "entropy": 5.740700149536133,
      "epoch": 2.614199572067691,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043125603900447026,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20670076608657836,
      "num_tokens": 58303387.0,
      "step": 33600
    },
    {
      "entropy": 5.766790866851807,
      "epoch": 2.614588601439409,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004312360958877913,
      "loss": 4.9498,
      "mean_token_accuracy": 0.21195856481790543,
      "num_tokens": 58312368.0,
      "step": 33605
    },
    {
      "entropy": 5.6911005020141605,
      "epoch": 2.614977630811126,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004312161504004782,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20431155264377593,
      "num_tokens": 58320639.0,
      "step": 33610
    },
    {
      "entropy": 5.658743762969971,
      "epoch": 2.615366660182844,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043119620254283375,
      "loss": 4.8951,
      "mean_token_accuracy": 0.2098271831870079,
      "num_tokens": 58329180.0,
      "step": 33615
    },
    {
      "entropy": 5.7568535804748535,
      "epoch": 2.6157556895545615,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043117625231516054,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21423605978488922,
      "num_tokens": 58337837.0,
      "step": 33620
    },
    {
      "entropy": 5.778844118118286,
      "epoch": 2.6161447189262788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004311562997177612,
      "loss": 4.91,
      "mean_token_accuracy": 0.21796731352806092,
      "num_tokens": 58347000.0,
      "step": 33625
    },
    {
      "entropy": 5.607855653762817,
      "epoch": 2.6165337482979965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004311363447509386,
      "loss": 4.9472,
      "mean_token_accuracy": 0.2084742933511734,
      "num_tokens": 58355895.0,
      "step": 33630
    },
    {
      "entropy": 5.671793413162232,
      "epoch": 2.616922777669714,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004311163874149955,
      "loss": 5.0169,
      "mean_token_accuracy": 0.20318219810724258,
      "num_tokens": 58364653.0,
      "step": 33635
    },
    {
      "entropy": 5.728887367248535,
      "epoch": 2.617311807041432,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043109642771023464,
      "loss": 4.8833,
      "mean_token_accuracy": 0.2168629512190819,
      "num_tokens": 58374157.0,
      "step": 33640
    },
    {
      "entropy": 5.823047828674317,
      "epoch": 2.617700836413149,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004310764656369591,
      "loss": 5.0232,
      "mean_token_accuracy": 0.20537955015897752,
      "num_tokens": 58382870.0,
      "step": 33645
    },
    {
      "entropy": 5.730980014801025,
      "epoch": 2.618089865784867,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004310565011954715,
      "loss": 4.9346,
      "mean_token_accuracy": 0.21156704276800156,
      "num_tokens": 58391016.0,
      "step": 33650
    },
    {
      "entropy": 5.758261489868164,
      "epoch": 2.618478895156584,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004310365343860748,
      "loss": 5.0338,
      "mean_token_accuracy": 0.2100912645459175,
      "num_tokens": 58399595.0,
      "step": 33655
    },
    {
      "entropy": 5.860752725601197,
      "epoch": 2.6188679245283017,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043101656520907225,
      "loss": 5.1263,
      "mean_token_accuracy": 0.1984937980771065,
      "num_tokens": 58409071.0,
      "step": 33660
    },
    {
      "entropy": 5.775374031066894,
      "epoch": 2.6192569539000194,
      "grad_norm": 1.25,
      "learning_rate": 0.0004309965936647665,
      "loss": 5.0443,
      "mean_token_accuracy": 0.20149798393249513,
      "num_tokens": 58418313.0,
      "step": 33665
    },
    {
      "entropy": 5.7613341331481935,
      "epoch": 2.619645983271737,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004309766197534608,
      "loss": 5.0169,
      "mean_token_accuracy": 0.19764334112405776,
      "num_tokens": 58426263.0,
      "step": 33670
    },
    {
      "entropy": 5.769893074035645,
      "epoch": 2.620035012643455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043095664347545816,
      "loss": 5.026,
      "mean_token_accuracy": 0.19989794492721558,
      "num_tokens": 58435449.0,
      "step": 33675
    },
    {
      "entropy": 5.79514799118042,
      "epoch": 2.620424042015172,
      "grad_norm": 1.25,
      "learning_rate": 0.0004309366648310616,
      "loss": 5.0463,
      "mean_token_accuracy": 0.20278458595275878,
      "num_tokens": 58444015.0,
      "step": 33680
    },
    {
      "entropy": 5.7270814895629885,
      "epoch": 2.62081307138689,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043091668382057443,
      "loss": 4.9684,
      "mean_token_accuracy": 0.20447409749031067,
      "num_tokens": 58452826.0,
      "step": 33685
    },
    {
      "entropy": 5.797095727920532,
      "epoch": 2.621202100758607,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043089670044429973,
      "loss": 5.0789,
      "mean_token_accuracy": 0.20720056742429732,
      "num_tokens": 58462175.0,
      "step": 33690
    },
    {
      "entropy": 5.68622612953186,
      "epoch": 2.6215911301303247,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043087671470254076,
      "loss": 4.8876,
      "mean_token_accuracy": 0.21023828983306886,
      "num_tokens": 58470714.0,
      "step": 33695
    },
    {
      "entropy": 5.694883155822754,
      "epoch": 2.6219801595020424,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043085672659560077,
      "loss": 4.9102,
      "mean_token_accuracy": 0.20954437851905822,
      "num_tokens": 58478839.0,
      "step": 33700
    },
    {
      "entropy": 5.786400365829468,
      "epoch": 2.62236918887376,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043083673612378295,
      "loss": 5.0622,
      "mean_token_accuracy": 0.20183077603578567,
      "num_tokens": 58488184.0,
      "step": 33705
    },
    {
      "entropy": 5.797252035140991,
      "epoch": 2.6227582182454774,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043081674328739066,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20082711130380632,
      "num_tokens": 58497202.0,
      "step": 33710
    },
    {
      "entropy": 5.737247562408447,
      "epoch": 2.623147247617195,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043079674808672744,
      "loss": 4.9042,
      "mean_token_accuracy": 0.20578330308198928,
      "num_tokens": 58505885.0,
      "step": 33715
    },
    {
      "entropy": 5.740895366668701,
      "epoch": 2.623536276988913,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004307767505220964,
      "loss": 4.9165,
      "mean_token_accuracy": 0.21376111656427382,
      "num_tokens": 58513807.0,
      "step": 33720
    },
    {
      "entropy": 5.681150579452515,
      "epoch": 2.62392530636063,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004307567505938012,
      "loss": 4.9061,
      "mean_token_accuracy": 0.2109743759036064,
      "num_tokens": 58522426.0,
      "step": 33725
    },
    {
      "entropy": 5.714715576171875,
      "epoch": 2.6243143357323477,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004307367483021452,
      "loss": 4.978,
      "mean_token_accuracy": 0.20549404621124268,
      "num_tokens": 58531038.0,
      "step": 33730
    },
    {
      "entropy": 5.762574338912964,
      "epoch": 2.6247033651040654,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000430716743647432,
      "loss": 4.9733,
      "mean_token_accuracy": 0.21315242648124694,
      "num_tokens": 58539373.0,
      "step": 33735
    },
    {
      "entropy": 5.697098016738892,
      "epoch": 2.625092394475783,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000430696736629965,
      "loss": 4.8452,
      "mean_token_accuracy": 0.2152157172560692,
      "num_tokens": 58547295.0,
      "step": 33740
    },
    {
      "entropy": 5.741835498809815,
      "epoch": 2.6254814238475004,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004306767272500478,
      "loss": 5.0712,
      "mean_token_accuracy": 0.20154183655977248,
      "num_tokens": 58555614.0,
      "step": 33745
    },
    {
      "entropy": 5.724033975601197,
      "epoch": 2.625870453219218,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043065671550798415,
      "loss": 4.9642,
      "mean_token_accuracy": 0.20256564468145372,
      "num_tokens": 58564444.0,
      "step": 33750
    },
    {
      "entropy": 5.708144569396973,
      "epoch": 2.626259482590936,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043063670140407753,
      "loss": 4.9325,
      "mean_token_accuracy": 0.20851137191057206,
      "num_tokens": 58573446.0,
      "step": 33755
    },
    {
      "entropy": 5.81295075416565,
      "epoch": 2.626648511962653,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004306166849386317,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20683825463056565,
      "num_tokens": 58582065.0,
      "step": 33760
    },
    {
      "entropy": 5.648973894119263,
      "epoch": 2.6270375413343707,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043059666611195037,
      "loss": 4.8756,
      "mean_token_accuracy": 0.21208566278219224,
      "num_tokens": 58590222.0,
      "step": 33765
    },
    {
      "entropy": 5.739373970031738,
      "epoch": 2.6274265707060884,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004305766449243372,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20121004581451415,
      "num_tokens": 58598745.0,
      "step": 33770
    },
    {
      "entropy": 5.81614351272583,
      "epoch": 2.627815600077806,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004305566213760962,
      "loss": 4.975,
      "mean_token_accuracy": 0.2091350093483925,
      "num_tokens": 58607199.0,
      "step": 33775
    },
    {
      "entropy": 5.799896717071533,
      "epoch": 2.6282046294495234,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043053659546753094,
      "loss": 5.0887,
      "mean_token_accuracy": 0.19685013294219972,
      "num_tokens": 58616593.0,
      "step": 33780
    },
    {
      "entropy": 5.724701166152954,
      "epoch": 2.628593658821241,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004305165671989455,
      "loss": 4.9859,
      "mean_token_accuracy": 0.20650027245283126,
      "num_tokens": 58626165.0,
      "step": 33785
    },
    {
      "entropy": 5.757542562484741,
      "epoch": 2.6289826881929583,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004304965365706437,
      "loss": 4.9637,
      "mean_token_accuracy": 0.21062172949314117,
      "num_tokens": 58634392.0,
      "step": 33790
    },
    {
      "entropy": 5.7461389064788815,
      "epoch": 2.629371717564676,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004304765035829294,
      "loss": 5.0616,
      "mean_token_accuracy": 0.20463840514421464,
      "num_tokens": 58644584.0,
      "step": 33795
    },
    {
      "entropy": 5.7736328601837155,
      "epoch": 2.6297607469363937,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043045646823610664,
      "loss": 5.0706,
      "mean_token_accuracy": 0.20260827988386154,
      "num_tokens": 58653601.0,
      "step": 33800
    },
    {
      "entropy": 5.770496225357055,
      "epoch": 2.6301497763081114,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043043643053047935,
      "loss": 4.9892,
      "mean_token_accuracy": 0.21079366356134416,
      "num_tokens": 58661947.0,
      "step": 33805
    },
    {
      "entropy": 5.808750200271606,
      "epoch": 2.6305388056798287,
      "grad_norm": 1.25,
      "learning_rate": 0.0004304163904663517,
      "loss": 5.1001,
      "mean_token_accuracy": 0.19874742180109023,
      "num_tokens": 58670537.0,
      "step": 33810
    },
    {
      "entropy": 5.778955030441284,
      "epoch": 2.6309278350515464,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043039634804402767,
      "loss": 4.9327,
      "mean_token_accuracy": 0.20697752386331558,
      "num_tokens": 58679491.0,
      "step": 33815
    },
    {
      "entropy": 5.738336420059204,
      "epoch": 2.631316864423264,
      "grad_norm": 1.25,
      "learning_rate": 0.0004303763032638114,
      "loss": 4.967,
      "mean_token_accuracy": 0.20661523342132568,
      "num_tokens": 58689078.0,
      "step": 33820
    },
    {
      "entropy": 5.733738708496094,
      "epoch": 2.6317058937949813,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004303562561260071,
      "loss": 4.9392,
      "mean_token_accuracy": 0.20538584142923355,
      "num_tokens": 58697113.0,
      "step": 33825
    },
    {
      "entropy": 5.781717824935913,
      "epoch": 2.632094923166699,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043033620663091883,
      "loss": 5.0297,
      "mean_token_accuracy": 0.2081252321600914,
      "num_tokens": 58705368.0,
      "step": 33830
    },
    {
      "entropy": 5.768449163436889,
      "epoch": 2.6324839525384167,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004303161547788509,
      "loss": 4.9888,
      "mean_token_accuracy": 0.20570405274629594,
      "num_tokens": 58713828.0,
      "step": 33835
    },
    {
      "entropy": 5.714462089538574,
      "epoch": 2.6328729819101344,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004302961005701074,
      "loss": 4.9522,
      "mean_token_accuracy": 0.2038877710700035,
      "num_tokens": 58721896.0,
      "step": 33840
    },
    {
      "entropy": 5.830992364883423,
      "epoch": 2.6332620112818517,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043027604400499295,
      "loss": 5.0257,
      "mean_token_accuracy": 0.20067147314548492,
      "num_tokens": 58729861.0,
      "step": 33845
    },
    {
      "entropy": 5.7352429866790775,
      "epoch": 2.6336510406535694,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043025598508381155,
      "loss": 4.916,
      "mean_token_accuracy": 0.2150746241211891,
      "num_tokens": 58738023.0,
      "step": 33850
    },
    {
      "entropy": 5.719020891189575,
      "epoch": 2.634040070025287,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004302359238068677,
      "loss": 4.9929,
      "mean_token_accuracy": 0.20194027572870255,
      "num_tokens": 58746678.0,
      "step": 33855
    },
    {
      "entropy": 5.6561439514160154,
      "epoch": 2.6344290993970043,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043021586017446585,
      "loss": 4.9014,
      "mean_token_accuracy": 0.2147592782974243,
      "num_tokens": 58754746.0,
      "step": 33860
    },
    {
      "entropy": 5.7043579578399655,
      "epoch": 2.634818128768722,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043019579418691027,
      "loss": 4.9865,
      "mean_token_accuracy": 0.20470765382051467,
      "num_tokens": 58762972.0,
      "step": 33865
    },
    {
      "entropy": 5.830435085296631,
      "epoch": 2.6352071581404397,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004301757258445056,
      "loss": 5.043,
      "mean_token_accuracy": 0.20073170065879822,
      "num_tokens": 58772529.0,
      "step": 33870
    },
    {
      "entropy": 5.693178462982178,
      "epoch": 2.6355961875121574,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004301556551475563,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2114935740828514,
      "num_tokens": 58781239.0,
      "step": 33875
    },
    {
      "entropy": 5.7335515975952145,
      "epoch": 2.6359852168838747,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004301355820963669,
      "loss": 4.9737,
      "mean_token_accuracy": 0.2073205143213272,
      "num_tokens": 58789876.0,
      "step": 33880
    },
    {
      "entropy": 5.736359310150147,
      "epoch": 2.6363742462555924,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004301155066912419,
      "loss": 5.0156,
      "mean_token_accuracy": 0.20825667083263397,
      "num_tokens": 58798383.0,
      "step": 33885
    },
    {
      "entropy": 5.790340280532837,
      "epoch": 2.6367632756273096,
      "grad_norm": 1.296875,
      "learning_rate": 0.000430095428932486,
      "loss": 5.0253,
      "mean_token_accuracy": 0.20657783299684523,
      "num_tokens": 58806804.0,
      "step": 33890
    },
    {
      "entropy": 5.760024166107177,
      "epoch": 2.6371523049990273,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043007534882040385,
      "loss": 5.0046,
      "mean_token_accuracy": 0.2038985714316368,
      "num_tokens": 58815755.0,
      "step": 33895
    },
    {
      "entropy": 5.782485866546631,
      "epoch": 2.637541334370745,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004300552663553001,
      "loss": 4.9876,
      "mean_token_accuracy": 0.20811828523874282,
      "num_tokens": 58824547.0,
      "step": 33900
    },
    {
      "entropy": 5.7784322738647464,
      "epoch": 2.6379303637424627,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004300351815374795,
      "loss": 5.0193,
      "mean_token_accuracy": 0.20673235654830932,
      "num_tokens": 58832471.0,
      "step": 33905
    },
    {
      "entropy": 5.765452575683594,
      "epoch": 2.63831939311418,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004300150943672467,
      "loss": 5.0755,
      "mean_token_accuracy": 0.20904047340154647,
      "num_tokens": 58841367.0,
      "step": 33910
    },
    {
      "entropy": 5.7486978530883786,
      "epoch": 2.6387084224858977,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004299950048449067,
      "loss": 4.9915,
      "mean_token_accuracy": 0.2039876252412796,
      "num_tokens": 58849946.0,
      "step": 33915
    },
    {
      "entropy": 5.828851985931396,
      "epoch": 2.6390974518576154,
      "grad_norm": 1.25,
      "learning_rate": 0.0004299749129707641,
      "loss": 5.0742,
      "mean_token_accuracy": 0.20084165036678314,
      "num_tokens": 58858288.0,
      "step": 33920
    },
    {
      "entropy": 5.855333375930786,
      "epoch": 2.6394864812293326,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004299548187451239,
      "loss": 5.0483,
      "mean_token_accuracy": 0.20018096268177032,
      "num_tokens": 58866968.0,
      "step": 33925
    },
    {
      "entropy": 5.785285711288452,
      "epoch": 2.6398755106010503,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042993472216829097,
      "loss": 4.9492,
      "mean_token_accuracy": 0.2095009833574295,
      "num_tokens": 58875635.0,
      "step": 33930
    },
    {
      "entropy": 5.788784980773926,
      "epoch": 2.640264539972768,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042991462324057025,
      "loss": 4.9789,
      "mean_token_accuracy": 0.21303447782993318,
      "num_tokens": 58884080.0,
      "step": 33935
    },
    {
      "entropy": 5.700630617141724,
      "epoch": 2.6406535693444857,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004298945219622667,
      "loss": 4.8982,
      "mean_token_accuracy": 0.2099815234541893,
      "num_tokens": 58891770.0,
      "step": 33940
    },
    {
      "entropy": 5.727335453033447,
      "epoch": 2.641042598716203,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042987441833368525,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20298945009708405,
      "num_tokens": 58900446.0,
      "step": 33945
    },
    {
      "entropy": 5.788039875030518,
      "epoch": 2.6414316280879206,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042985431235513104,
      "loss": 5.0038,
      "mean_token_accuracy": 0.2093222126364708,
      "num_tokens": 58909423.0,
      "step": 33950
    },
    {
      "entropy": 5.8286150932312015,
      "epoch": 2.6418206574596383,
      "grad_norm": 1.25,
      "learning_rate": 0.00042983420402690917,
      "loss": 5.0596,
      "mean_token_accuracy": 0.19979308247566224,
      "num_tokens": 58918175.0,
      "step": 33955
    },
    {
      "entropy": 5.7000058650970455,
      "epoch": 2.6422096868313556,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042981409334932457,
      "loss": 4.981,
      "mean_token_accuracy": 0.2027666911482811,
      "num_tokens": 58926837.0,
      "step": 33960
    },
    {
      "entropy": 5.77056188583374,
      "epoch": 2.6425987162030733,
      "grad_norm": 1.25,
      "learning_rate": 0.0004297939803226826,
      "loss": 5.0374,
      "mean_token_accuracy": 0.2035613551735878,
      "num_tokens": 58934994.0,
      "step": 33965
    },
    {
      "entropy": 5.716078615188598,
      "epoch": 2.642987745574791,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004297738649472884,
      "loss": 4.9848,
      "mean_token_accuracy": 0.20521089434623718,
      "num_tokens": 58943937.0,
      "step": 33970
    },
    {
      "entropy": 5.785591173171997,
      "epoch": 2.6433767749465087,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042975374722344713,
      "loss": 5.0158,
      "mean_token_accuracy": 0.1968502178788185,
      "num_tokens": 58952360.0,
      "step": 33975
    },
    {
      "entropy": 5.7958756446838375,
      "epoch": 2.643765804318226,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042973362715146397,
      "loss": 4.9613,
      "mean_token_accuracy": 0.21146545112133025,
      "num_tokens": 58961846.0,
      "step": 33980
    },
    {
      "entropy": 5.798363828659058,
      "epoch": 2.6441548336899436,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004297135047316444,
      "loss": 5.0238,
      "mean_token_accuracy": 0.20462702959775925,
      "num_tokens": 58970215.0,
      "step": 33985
    },
    {
      "entropy": 5.719180059432984,
      "epoch": 2.644543863061661,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004296933799642936,
      "loss": 4.9234,
      "mean_token_accuracy": 0.20716263502836227,
      "num_tokens": 58978623.0,
      "step": 33990
    },
    {
      "entropy": 5.7968532085418705,
      "epoch": 2.6449328924333786,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000429673252849717,
      "loss": 4.9762,
      "mean_token_accuracy": 0.21215085834264755,
      "num_tokens": 58986616.0,
      "step": 33995
    },
    {
      "entropy": 5.782397222518921,
      "epoch": 2.6453219218050963,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042965312338822005,
      "loss": 5.04,
      "mean_token_accuracy": 0.201270455121994,
      "num_tokens": 58996121.0,
      "step": 34000
    },
    {
      "entropy": 5.746051931381226,
      "epoch": 2.645710951176814,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000429632991580108,
      "loss": 4.9425,
      "mean_token_accuracy": 0.2059352442622185,
      "num_tokens": 59004272.0,
      "step": 34005
    },
    {
      "entropy": 5.7122009754180905,
      "epoch": 2.6460999805485317,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004296128574256864,
      "loss": 4.9925,
      "mean_token_accuracy": 0.2080587089061737,
      "num_tokens": 59012513.0,
      "step": 34010
    },
    {
      "entropy": 5.752615070343017,
      "epoch": 2.646489009920249,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042959272092526086,
      "loss": 4.9787,
      "mean_token_accuracy": 0.20590208023786544,
      "num_tokens": 59021071.0,
      "step": 34015
    },
    {
      "entropy": 5.834737730026245,
      "epoch": 2.6468780392919666,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042957258207913687,
      "loss": 5.0161,
      "mean_token_accuracy": 0.20381605625152588,
      "num_tokens": 59030154.0,
      "step": 34020
    },
    {
      "entropy": 5.817493343353272,
      "epoch": 2.647267068663684,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042955244088761985,
      "loss": 5.0606,
      "mean_token_accuracy": 0.2063338950276375,
      "num_tokens": 59038962.0,
      "step": 34025
    },
    {
      "entropy": 5.737457227706909,
      "epoch": 2.6476560980354016,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004295322973510156,
      "loss": 4.9898,
      "mean_token_accuracy": 0.20038558840751647,
      "num_tokens": 59046856.0,
      "step": 34030
    },
    {
      "entropy": 5.781431579589844,
      "epoch": 2.6480451274071193,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004295121514696297,
      "loss": 5.0165,
      "mean_token_accuracy": 0.20398189574480058,
      "num_tokens": 59055124.0,
      "step": 34035
    },
    {
      "entropy": 5.807726955413818,
      "epoch": 2.648434156778837,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042949200324376784,
      "loss": 5.0456,
      "mean_token_accuracy": 0.20543297976255417,
      "num_tokens": 59064104.0,
      "step": 34040
    },
    {
      "entropy": 5.726795291900634,
      "epoch": 2.6488231861505542,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004294718526737357,
      "loss": 4.9524,
      "mean_token_accuracy": 0.20853398591279984,
      "num_tokens": 59071991.0,
      "step": 34045
    },
    {
      "entropy": 5.70872049331665,
      "epoch": 2.649212215522272,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004294516997598391,
      "loss": 4.9693,
      "mean_token_accuracy": 0.21636675596237182,
      "num_tokens": 59080519.0,
      "step": 34050
    },
    {
      "entropy": 5.737383842468262,
      "epoch": 2.6496012448939896,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004294315445023838,
      "loss": 4.9846,
      "mean_token_accuracy": 0.20309363156557084,
      "num_tokens": 59089555.0,
      "step": 34055
    },
    {
      "entropy": 5.781348943710327,
      "epoch": 2.649990274265707,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042941138690167556,
      "loss": 4.9735,
      "mean_token_accuracy": 0.21209764927625657,
      "num_tokens": 59098287.0,
      "step": 34060
    },
    {
      "entropy": 5.673504114151001,
      "epoch": 2.6503793036374246,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004293912269580204,
      "loss": 4.9775,
      "mean_token_accuracy": 0.21377580761909484,
      "num_tokens": 59106463.0,
      "step": 34065
    },
    {
      "entropy": 5.817241907119751,
      "epoch": 2.6507683330091423,
      "grad_norm": 1.203125,
      "learning_rate": 0.000429371064671724,
      "loss": 5.0344,
      "mean_token_accuracy": 0.20153960585594177,
      "num_tokens": 59115324.0,
      "step": 34070
    },
    {
      "entropy": 5.7594386577606205,
      "epoch": 2.65115736238086,
      "grad_norm": 1.25,
      "learning_rate": 0.00042935090004309247,
      "loss": 4.9412,
      "mean_token_accuracy": 0.20840903073549272,
      "num_tokens": 59123438.0,
      "step": 34075
    },
    {
      "entropy": 5.802229022979736,
      "epoch": 2.6515463917525772,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042933073307243165,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20748104751110077,
      "num_tokens": 59131436.0,
      "step": 34080
    },
    {
      "entropy": 5.714805603027344,
      "epoch": 2.651935421124295,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042931056376004767,
      "loss": 5.0567,
      "mean_token_accuracy": 0.21004955023527144,
      "num_tokens": 59139981.0,
      "step": 34085
    },
    {
      "entropy": 5.721381664276123,
      "epoch": 2.652324450496012,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004292903921062465,
      "loss": 4.9589,
      "mean_token_accuracy": 0.21093981117010116,
      "num_tokens": 59148833.0,
      "step": 34090
    },
    {
      "entropy": 5.776584148406982,
      "epoch": 2.65271347986773,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042927021811133416,
      "loss": 5.0368,
      "mean_token_accuracy": 0.1987171545624733,
      "num_tokens": 59157597.0,
      "step": 34095
    },
    {
      "entropy": 5.782119941711426,
      "epoch": 2.6531025092394476,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004292500417756167,
      "loss": 4.969,
      "mean_token_accuracy": 0.2091366469860077,
      "num_tokens": 59166286.0,
      "step": 34100
    },
    {
      "entropy": 5.7783575534820555,
      "epoch": 2.6534915386111653,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042922986309940054,
      "loss": 5.0221,
      "mean_token_accuracy": 0.2092737764120102,
      "num_tokens": 59174019.0,
      "step": 34105
    },
    {
      "entropy": 5.757407236099243,
      "epoch": 2.653880567982883,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042920968208299165,
      "loss": 4.9923,
      "mean_token_accuracy": 0.20546508729457855,
      "num_tokens": 59183885.0,
      "step": 34110
    },
    {
      "entropy": 5.78719129562378,
      "epoch": 2.6542695973546,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004291894987266963,
      "loss": 4.9174,
      "mean_token_accuracy": 0.20796530246734618,
      "num_tokens": 59192196.0,
      "step": 34115
    },
    {
      "entropy": 5.769192695617676,
      "epoch": 2.654658626726318,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042916931303082064,
      "loss": 4.9882,
      "mean_token_accuracy": 0.20765392780303954,
      "num_tokens": 59200269.0,
      "step": 34120
    },
    {
      "entropy": 5.788691234588623,
      "epoch": 2.655047656098035,
      "grad_norm": 1.25,
      "learning_rate": 0.00042914912499567113,
      "loss": 5.0551,
      "mean_token_accuracy": 0.1997659608721733,
      "num_tokens": 59208690.0,
      "step": 34125
    },
    {
      "entropy": 5.807337713241577,
      "epoch": 2.655436685469753,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042912893462155395,
      "loss": 5.0103,
      "mean_token_accuracy": 0.20489755272865295,
      "num_tokens": 59217222.0,
      "step": 34130
    },
    {
      "entropy": 5.803229856491089,
      "epoch": 2.6558257148414706,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042910874190877545,
      "loss": 5.0237,
      "mean_token_accuracy": 0.20683650225400924,
      "num_tokens": 59226161.0,
      "step": 34135
    },
    {
      "entropy": 5.810457229614258,
      "epoch": 2.6562147442131883,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004290885468576422,
      "loss": 4.9886,
      "mean_token_accuracy": 0.20274344682693482,
      "num_tokens": 59234239.0,
      "step": 34140
    },
    {
      "entropy": 5.742700910568237,
      "epoch": 2.6566037735849055,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004290683494684604,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21063126623630524,
      "num_tokens": 59243390.0,
      "step": 34145
    },
    {
      "entropy": 5.728374290466308,
      "epoch": 2.656992802956623,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004290481497415367,
      "loss": 5.0125,
      "mean_token_accuracy": 0.21027939021587372,
      "num_tokens": 59252854.0,
      "step": 34150
    },
    {
      "entropy": 5.730996322631836,
      "epoch": 2.657381832328341,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004290279476771775,
      "loss": 4.9585,
      "mean_token_accuracy": 0.2089229166507721,
      "num_tokens": 59261505.0,
      "step": 34155
    },
    {
      "entropy": 5.781494092941284,
      "epoch": 2.657770861700058,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004290077432756894,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20536563247442247,
      "num_tokens": 59269209.0,
      "step": 34160
    },
    {
      "entropy": 5.774490642547607,
      "epoch": 2.658159891071776,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004289875365373788,
      "loss": 5.0026,
      "mean_token_accuracy": 0.1972527414560318,
      "num_tokens": 59277951.0,
      "step": 34165
    },
    {
      "entropy": 5.670295381546021,
      "epoch": 2.6585489204434936,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042896732746255256,
      "loss": 4.8796,
      "mean_token_accuracy": 0.21668145507574083,
      "num_tokens": 59286266.0,
      "step": 34170
    },
    {
      "entropy": 5.761726760864258,
      "epoch": 2.6589379498152113,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042894711605151714,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20081547200679778,
      "num_tokens": 59294633.0,
      "step": 34175
    },
    {
      "entropy": 5.783517694473266,
      "epoch": 2.6593269791869285,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042892690230457924,
      "loss": 4.9585,
      "mean_token_accuracy": 0.21711023449897765,
      "num_tokens": 59303526.0,
      "step": 34180
    },
    {
      "entropy": 5.799003934860229,
      "epoch": 2.659716008558646,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042890668622204566,
      "loss": 4.9503,
      "mean_token_accuracy": 0.2143409878015518,
      "num_tokens": 59311921.0,
      "step": 34185
    },
    {
      "entropy": 5.819645595550537,
      "epoch": 2.660105037930364,
      "grad_norm": 1.265625,
      "learning_rate": 0.000428886467804223,
      "loss": 5.119,
      "mean_token_accuracy": 0.20340732634067535,
      "num_tokens": 59320750.0,
      "step": 34190
    },
    {
      "entropy": 5.710008668899536,
      "epoch": 2.660494067302081,
      "grad_norm": 1.125,
      "learning_rate": 0.00042886624705141825,
      "loss": 4.918,
      "mean_token_accuracy": 0.20686909556388855,
      "num_tokens": 59330036.0,
      "step": 34195
    },
    {
      "entropy": 5.777923727035523,
      "epoch": 2.660883096673799,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042884602396393796,
      "loss": 5.0774,
      "mean_token_accuracy": 0.19690336883068085,
      "num_tokens": 59339022.0,
      "step": 34200
    },
    {
      "entropy": 5.767139577865601,
      "epoch": 2.6612721260455166,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004288257985420892,
      "loss": 4.9312,
      "mean_token_accuracy": 0.20684129297733306,
      "num_tokens": 59347599.0,
      "step": 34205
    },
    {
      "entropy": 5.864638757705689,
      "epoch": 2.6616611554172342,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004288055707861788,
      "loss": 5.0431,
      "mean_token_accuracy": 0.20078569501638413,
      "num_tokens": 59356453.0,
      "step": 34210
    },
    {
      "entropy": 5.733516263961792,
      "epoch": 2.6620501847889515,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042878534069651364,
      "loss": 4.9921,
      "mean_token_accuracy": 0.21365273594856263,
      "num_tokens": 59364617.0,
      "step": 34215
    },
    {
      "entropy": 5.767459869384766,
      "epoch": 2.662439214160669,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004287651082734007,
      "loss": 5.0145,
      "mean_token_accuracy": 0.20334753543138503,
      "num_tokens": 59374144.0,
      "step": 34220
    },
    {
      "entropy": 5.7932545185089115,
      "epoch": 2.6628282435323865,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000428744873517147,
      "loss": 5.0876,
      "mean_token_accuracy": 0.19351752549409867,
      "num_tokens": 59382572.0,
      "step": 34225
    },
    {
      "entropy": 5.7979217052459715,
      "epoch": 2.663217272904104,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004287246364280596,
      "loss": 5.0858,
      "mean_token_accuracy": 0.1991134151816368,
      "num_tokens": 59391582.0,
      "step": 34230
    },
    {
      "entropy": 5.740156793594361,
      "epoch": 2.663606302275822,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004287043970064455,
      "loss": 4.9852,
      "mean_token_accuracy": 0.2120040699839592,
      "num_tokens": 59400760.0,
      "step": 34235
    },
    {
      "entropy": 5.85084080696106,
      "epoch": 2.6639953316475395,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004286841552526118,
      "loss": 5.1228,
      "mean_token_accuracy": 0.2027067020535469,
      "num_tokens": 59409427.0,
      "step": 34240
    },
    {
      "entropy": 5.756965494155883,
      "epoch": 2.664384361019257,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042866391116686567,
      "loss": 5.0072,
      "mean_token_accuracy": 0.20642681866884233,
      "num_tokens": 59417538.0,
      "step": 34245
    },
    {
      "entropy": 5.803336811065674,
      "epoch": 2.6647733903909745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004286436647495142,
      "loss": 4.9753,
      "mean_token_accuracy": 0.20432389080524443,
      "num_tokens": 59426499.0,
      "step": 34250
    },
    {
      "entropy": 5.853385734558105,
      "epoch": 2.665162419762692,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042862341600086477,
      "loss": 5.0675,
      "mean_token_accuracy": 0.19707028716802596,
      "num_tokens": 59434921.0,
      "step": 34255
    },
    {
      "entropy": 5.722882604598999,
      "epoch": 2.6655514491344094,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042860316492122447,
      "loss": 4.8551,
      "mean_token_accuracy": 0.2123110696673393,
      "num_tokens": 59443930.0,
      "step": 34260
    },
    {
      "entropy": 5.750018978118897,
      "epoch": 2.665940478506127,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004285829115109006,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21142388433218,
      "num_tokens": 59453335.0,
      "step": 34265
    },
    {
      "entropy": 5.768413019180298,
      "epoch": 2.666329507877845,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004285626557702005,
      "loss": 5.0263,
      "mean_token_accuracy": 0.20362842231988906,
      "num_tokens": 59461937.0,
      "step": 34270
    },
    {
      "entropy": 5.823128795623779,
      "epoch": 2.6667185372495625,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004285423976994316,
      "loss": 4.9652,
      "mean_token_accuracy": 0.2073827564716339,
      "num_tokens": 59470858.0,
      "step": 34275
    },
    {
      "entropy": 5.7865955352783205,
      "epoch": 2.66710756662128,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042852213729890117,
      "loss": 5.0226,
      "mean_token_accuracy": 0.2045006662607193,
      "num_tokens": 59478902.0,
      "step": 34280
    },
    {
      "entropy": 5.702309513092041,
      "epoch": 2.6674965959929975,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004285018745689166,
      "loss": 4.934,
      "mean_token_accuracy": 0.21013825982809067,
      "num_tokens": 59487281.0,
      "step": 34285
    },
    {
      "entropy": 5.75648946762085,
      "epoch": 2.667885625364715,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004284816095097855,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20621083974838256,
      "num_tokens": 59496304.0,
      "step": 34290
    },
    {
      "entropy": 5.6896919250488285,
      "epoch": 2.6682746547364324,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004284613421218152,
      "loss": 4.9135,
      "mean_token_accuracy": 0.21539425253868102,
      "num_tokens": 59504577.0,
      "step": 34295
    },
    {
      "entropy": 5.692991590499878,
      "epoch": 2.66866368410815,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042844107240531324,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20654021501541137,
      "num_tokens": 59513468.0,
      "step": 34300
    },
    {
      "entropy": 5.79981369972229,
      "epoch": 2.669052713479868,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004284208003605873,
      "loss": 5.0113,
      "mean_token_accuracy": 0.200614857673645,
      "num_tokens": 59522516.0,
      "step": 34305
    },
    {
      "entropy": 5.797414398193359,
      "epoch": 2.6694417428515855,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000428400525987945,
      "loss": 4.9576,
      "mean_token_accuracy": 0.20651436150074004,
      "num_tokens": 59531434.0,
      "step": 34310
    },
    {
      "entropy": 5.816468048095703,
      "epoch": 2.669830772223303,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042838024928769375,
      "loss": 4.9992,
      "mean_token_accuracy": 0.2122481048107147,
      "num_tokens": 59539967.0,
      "step": 34315
    },
    {
      "entropy": 5.815037727355957,
      "epoch": 2.6702198015950205,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042835997026014145,
      "loss": 5.115,
      "mean_token_accuracy": 0.19873297810554505,
      "num_tokens": 59549558.0,
      "step": 34320
    },
    {
      "entropy": 5.726599311828613,
      "epoch": 2.6706088309667377,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004283396889055957,
      "loss": 4.9481,
      "mean_token_accuracy": 0.2147311896085739,
      "num_tokens": 59557520.0,
      "step": 34325
    },
    {
      "entropy": 5.851682806015015,
      "epoch": 2.6709978603384554,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004283194052243641,
      "loss": 5.0727,
      "mean_token_accuracy": 0.20618268400430678,
      "num_tokens": 59565753.0,
      "step": 34330
    },
    {
      "entropy": 5.834269285202026,
      "epoch": 2.671386889710173,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042829911921675456,
      "loss": 5.0742,
      "mean_token_accuracy": 0.2059593141078949,
      "num_tokens": 59574522.0,
      "step": 34335
    },
    {
      "entropy": 5.731231784820556,
      "epoch": 2.671775919081891,
      "grad_norm": 1.203125,
      "learning_rate": 0.000428278830883075,
      "loss": 4.9562,
      "mean_token_accuracy": 0.2056015819311142,
      "num_tokens": 59583733.0,
      "step": 34340
    },
    {
      "entropy": 5.791233682632447,
      "epoch": 2.6721649484536085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042825854022363307,
      "loss": 5.0905,
      "mean_token_accuracy": 0.20382305830717087,
      "num_tokens": 59593009.0,
      "step": 34345
    },
    {
      "entropy": 5.780006074905396,
      "epoch": 2.6725539778253258,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042823824723873676,
      "loss": 4.9974,
      "mean_token_accuracy": 0.2115011304616928,
      "num_tokens": 59601770.0,
      "step": 34350
    },
    {
      "entropy": 5.759178161621094,
      "epoch": 2.6729430071970435,
      "grad_norm": 1.25,
      "learning_rate": 0.00042821795192869407,
      "loss": 4.9421,
      "mean_token_accuracy": 0.21230160146951677,
      "num_tokens": 59610397.0,
      "step": 34355
    },
    {
      "entropy": 5.785210514068604,
      "epoch": 2.6733320365687607,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004281976542938127,
      "loss": 4.9987,
      "mean_token_accuracy": 0.2062266156077385,
      "num_tokens": 59618864.0,
      "step": 34360
    },
    {
      "entropy": 5.728514957427978,
      "epoch": 2.6737210659404784,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042817735433440076,
      "loss": 4.98,
      "mean_token_accuracy": 0.20489100515842437,
      "num_tokens": 59627469.0,
      "step": 34365
    },
    {
      "entropy": 5.714287614822387,
      "epoch": 2.674110095312196,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004281570520507663,
      "loss": 4.9758,
      "mean_token_accuracy": 0.20644194334745408,
      "num_tokens": 59636125.0,
      "step": 34370
    },
    {
      "entropy": 5.744770622253418,
      "epoch": 2.674499124683914,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042813674744321737,
      "loss": 4.9475,
      "mean_token_accuracy": 0.2105255275964737,
      "num_tokens": 59644586.0,
      "step": 34375
    },
    {
      "entropy": 5.8100052833557125,
      "epoch": 2.674888154055631,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000428116440512062,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20611904114484786,
      "num_tokens": 59653644.0,
      "step": 34380
    },
    {
      "entropy": 5.729713821411133,
      "epoch": 2.6752771834273488,
      "grad_norm": 1.25,
      "learning_rate": 0.0004280961312576084,
      "loss": 5.0297,
      "mean_token_accuracy": 0.19709552824497223,
      "num_tokens": 59662270.0,
      "step": 34385
    },
    {
      "entropy": 5.7554456233978275,
      "epoch": 2.6756662127990665,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004280758196801646,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20997452437877656,
      "num_tokens": 59670544.0,
      "step": 34390
    },
    {
      "entropy": 5.80449652671814,
      "epoch": 2.6760552421707837,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042805550578003894,
      "loss": 5.0725,
      "mean_token_accuracy": 0.20609840154647827,
      "num_tokens": 59679985.0,
      "step": 34395
    },
    {
      "entropy": 5.775054883956909,
      "epoch": 2.6764442715425014,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004280351895575396,
      "loss": 5.048,
      "mean_token_accuracy": 0.204373799264431,
      "num_tokens": 59689831.0,
      "step": 34400
    },
    {
      "entropy": 5.79603943824768,
      "epoch": 2.676833300914219,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004280148710129748,
      "loss": 5.046,
      "mean_token_accuracy": 0.20629473626613617,
      "num_tokens": 59697400.0,
      "step": 34405
    },
    {
      "entropy": 5.754294967651367,
      "epoch": 2.677222330285937,
      "grad_norm": 1.25,
      "learning_rate": 0.00042799455014665296,
      "loss": 4.98,
      "mean_token_accuracy": 0.20895417928695678,
      "num_tokens": 59706137.0,
      "step": 34410
    },
    {
      "entropy": 5.814772653579712,
      "epoch": 2.677611359657654,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042797422695888226,
      "loss": 5.0591,
      "mean_token_accuracy": 0.19948384761810303,
      "num_tokens": 59715454.0,
      "step": 34415
    },
    {
      "entropy": 5.801063108444214,
      "epoch": 2.6780003890293718,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004279539014499712,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21211555749177932,
      "num_tokens": 59723801.0,
      "step": 34420
    },
    {
      "entropy": 5.786560726165772,
      "epoch": 2.678389418401089,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004279335736202281,
      "loss": 5.0142,
      "mean_token_accuracy": 0.2042987510561943,
      "num_tokens": 59732899.0,
      "step": 34425
    },
    {
      "entropy": 5.744022989273072,
      "epoch": 2.6787784477728067,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004279132434699615,
      "loss": 5.0584,
      "mean_token_accuracy": 0.20824712216854097,
      "num_tokens": 59741693.0,
      "step": 34430
    },
    {
      "entropy": 5.748468971252441,
      "epoch": 2.6791674771445244,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004278929109994798,
      "loss": 4.9097,
      "mean_token_accuracy": 0.2120950147509575,
      "num_tokens": 59749583.0,
      "step": 34435
    },
    {
      "entropy": 5.807040071487426,
      "epoch": 2.679556506516242,
      "grad_norm": 1.25,
      "learning_rate": 0.00042787257620909144,
      "loss": 5.0845,
      "mean_token_accuracy": 0.19569815397262574,
      "num_tokens": 59758638.0,
      "step": 34440
    },
    {
      "entropy": 5.727781581878662,
      "epoch": 2.67994553588796,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042785223909910515,
      "loss": 4.9296,
      "mean_token_accuracy": 0.212016062438488,
      "num_tokens": 59767746.0,
      "step": 34445
    },
    {
      "entropy": 5.777770614624023,
      "epoch": 2.680334565259677,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004278318996698294,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20249080657958984,
      "num_tokens": 59776426.0,
      "step": 34450
    },
    {
      "entropy": 5.780412244796753,
      "epoch": 2.6807235946313948,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004278115579215728,
      "loss": 5.0137,
      "mean_token_accuracy": 0.20517564862966536,
      "num_tokens": 59785032.0,
      "step": 34455
    },
    {
      "entropy": 5.775725317001343,
      "epoch": 2.681112624003112,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004277912138546441,
      "loss": 5.0328,
      "mean_token_accuracy": 0.20419394969940186,
      "num_tokens": 59794510.0,
      "step": 34460
    },
    {
      "entropy": 5.729579210281372,
      "epoch": 2.6815016533748297,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004277708674693519,
      "loss": 4.9635,
      "mean_token_accuracy": 0.2043824464082718,
      "num_tokens": 59802889.0,
      "step": 34465
    },
    {
      "entropy": 5.7661572933197025,
      "epoch": 2.6818906827465474,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042775051876600483,
      "loss": 4.9655,
      "mean_token_accuracy": 0.2051308274269104,
      "num_tokens": 59811940.0,
      "step": 34470
    },
    {
      "entropy": 5.766362524032592,
      "epoch": 2.682279712118265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004277301677449119,
      "loss": 4.9839,
      "mean_token_accuracy": 0.2079742893576622,
      "num_tokens": 59820465.0,
      "step": 34475
    },
    {
      "entropy": 5.814667701721191,
      "epoch": 2.6826687414899824,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004277098144063818,
      "loss": 5.0867,
      "mean_token_accuracy": 0.20039381384849547,
      "num_tokens": 59829611.0,
      "step": 34480
    },
    {
      "entropy": 5.829601716995239,
      "epoch": 2.6830577708617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042768945875072326,
      "loss": 5.0192,
      "mean_token_accuracy": 0.20279199182987212,
      "num_tokens": 59839073.0,
      "step": 34485
    },
    {
      "entropy": 5.826048278808594,
      "epoch": 2.6834468002334178,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004276691007782451,
      "loss": 5.0809,
      "mean_token_accuracy": 0.19903241842985153,
      "num_tokens": 59847653.0,
      "step": 34490
    },
    {
      "entropy": 5.790249252319336,
      "epoch": 2.683835829605135,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004276487404892565,
      "loss": 5.0394,
      "mean_token_accuracy": 0.196772937476635,
      "num_tokens": 59855534.0,
      "step": 34495
    },
    {
      "entropy": 5.767035293579101,
      "epoch": 2.6842248589768527,
      "grad_norm": 1.375,
      "learning_rate": 0.00042762837788406615,
      "loss": 4.9765,
      "mean_token_accuracy": 0.20635103732347487,
      "num_tokens": 59864862.0,
      "step": 34500
    },
    {
      "entropy": 5.76398344039917,
      "epoch": 2.6846138883485704,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004276080129629831,
      "loss": 5.0088,
      "mean_token_accuracy": 0.2057476446032524,
      "num_tokens": 59873100.0,
      "step": 34505
    },
    {
      "entropy": 5.783201885223389,
      "epoch": 2.685002917720288,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042758764572631644,
      "loss": 4.9594,
      "mean_token_accuracy": 0.21393164694309236,
      "num_tokens": 59881500.0,
      "step": 34510
    },
    {
      "entropy": 5.746620225906372,
      "epoch": 2.6853919470920053,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042756727617437504,
      "loss": 4.9726,
      "mean_token_accuracy": 0.2087078019976616,
      "num_tokens": 59890168.0,
      "step": 34515
    },
    {
      "entropy": 5.787930107116699,
      "epoch": 2.685780976463723,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042754690430746805,
      "loss": 5.0211,
      "mean_token_accuracy": 0.20230022370815276,
      "num_tokens": 59898805.0,
      "step": 34520
    },
    {
      "entropy": 5.81476845741272,
      "epoch": 2.6861700058354403,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004275265301259046,
      "loss": 5.0387,
      "mean_token_accuracy": 0.19938917458057404,
      "num_tokens": 59907012.0,
      "step": 34525
    },
    {
      "entropy": 5.819837427139282,
      "epoch": 2.686559035207158,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042750615362999395,
      "loss": 4.9879,
      "mean_token_accuracy": 0.21245453655719757,
      "num_tokens": 59915117.0,
      "step": 34530
    },
    {
      "entropy": 5.780423641204834,
      "epoch": 2.6869480645788757,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042748577482004495,
      "loss": 5.0241,
      "mean_token_accuracy": 0.20278410911560057,
      "num_tokens": 59923625.0,
      "step": 34535
    },
    {
      "entropy": 5.810448837280274,
      "epoch": 2.6873370939505934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042746539369636715,
      "loss": 5.0208,
      "mean_token_accuracy": 0.20138627141714097,
      "num_tokens": 59932456.0,
      "step": 34540
    },
    {
      "entropy": 5.754693937301636,
      "epoch": 2.687726123322311,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042744501025926966,
      "loss": 4.9668,
      "mean_token_accuracy": 0.20627070516347884,
      "num_tokens": 59941210.0,
      "step": 34545
    },
    {
      "entropy": 5.798054981231689,
      "epoch": 2.6881151526940283,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004274246245090618,
      "loss": 5.0793,
      "mean_token_accuracy": 0.1944602519273758,
      "num_tokens": 59949646.0,
      "step": 34550
    },
    {
      "entropy": 5.717916488647461,
      "epoch": 2.688504182065746,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004274042364460529,
      "loss": 4.8759,
      "mean_token_accuracy": 0.21776845753192903,
      "num_tokens": 59958263.0,
      "step": 34555
    },
    {
      "entropy": 5.812463426589966,
      "epoch": 2.6888932114374633,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042738384607055214,
      "loss": 5.0798,
      "mean_token_accuracy": 0.20343097746372224,
      "num_tokens": 59967222.0,
      "step": 34560
    },
    {
      "entropy": 5.782292699813842,
      "epoch": 2.689282240809181,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042736345338286917,
      "loss": 4.9496,
      "mean_token_accuracy": 0.20751760005950928,
      "num_tokens": 59974898.0,
      "step": 34565
    },
    {
      "entropy": 5.746344089508057,
      "epoch": 2.6896712701808987,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004273430583833133,
      "loss": 4.9283,
      "mean_token_accuracy": 0.2110963985323906,
      "num_tokens": 59983593.0,
      "step": 34570
    },
    {
      "entropy": 5.735291194915772,
      "epoch": 2.6900602995526164,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000427322661072194,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21232431679964064,
      "num_tokens": 59991360.0,
      "step": 34575
    },
    {
      "entropy": 5.782443809509277,
      "epoch": 2.6904493289243336,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004273022614498208,
      "loss": 5.0037,
      "mean_token_accuracy": 0.2048299267888069,
      "num_tokens": 59999123.0,
      "step": 34580
    },
    {
      "entropy": 5.717710494995117,
      "epoch": 2.6908383582960513,
      "grad_norm": 1.25,
      "learning_rate": 0.0004272818595165031,
      "loss": 4.9002,
      "mean_token_accuracy": 0.2105396270751953,
      "num_tokens": 60007666.0,
      "step": 34585
    },
    {
      "entropy": 5.772450971603393,
      "epoch": 2.691227387667769,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004272614552725506,
      "loss": 5.0891,
      "mean_token_accuracy": 0.19817533940076829,
      "num_tokens": 60016813.0,
      "step": 34590
    },
    {
      "entropy": 5.804533863067627,
      "epoch": 2.6916164170394863,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042724104871827293,
      "loss": 4.9974,
      "mean_token_accuracy": 0.2032974898815155,
      "num_tokens": 60025243.0,
      "step": 34595
    },
    {
      "entropy": 5.807737541198731,
      "epoch": 2.692005446411204,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004272206398539795,
      "loss": 5.0985,
      "mean_token_accuracy": 0.20094842165708543,
      "num_tokens": 60033705.0,
      "step": 34600
    },
    {
      "entropy": 5.794867038726807,
      "epoch": 2.6923944757829217,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042720022867998026,
      "loss": 4.9864,
      "mean_token_accuracy": 0.2139917179942131,
      "num_tokens": 60042073.0,
      "step": 34605
    },
    {
      "entropy": 5.8213622093200685,
      "epoch": 2.6927835051546394,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004271798151965848,
      "loss": 5.0659,
      "mean_token_accuracy": 0.19883980751037597,
      "num_tokens": 60050286.0,
      "step": 34610
    },
    {
      "entropy": 5.721873807907104,
      "epoch": 2.6931725345263566,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004271593994041029,
      "loss": 4.8599,
      "mean_token_accuracy": 0.20827201306819915,
      "num_tokens": 60057812.0,
      "step": 34615
    },
    {
      "entropy": 5.6534449577331545,
      "epoch": 2.6935615638980743,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004271389813028443,
      "loss": 4.8976,
      "mean_token_accuracy": 0.21934927850961686,
      "num_tokens": 60066636.0,
      "step": 34620
    },
    {
      "entropy": 5.671154546737671,
      "epoch": 2.693950593269792,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004271185608931187,
      "loss": 4.9954,
      "mean_token_accuracy": 0.204618401825428,
      "num_tokens": 60075605.0,
      "step": 34625
    },
    {
      "entropy": 5.7769664287567135,
      "epoch": 2.6943396226415093,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042709813817523617,
      "loss": 4.8862,
      "mean_token_accuracy": 0.21259747445583344,
      "num_tokens": 60083947.0,
      "step": 34630
    },
    {
      "entropy": 5.723951578140259,
      "epoch": 2.694728652013227,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042707771314950645,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21884589344263078,
      "num_tokens": 60092518.0,
      "step": 34635
    },
    {
      "entropy": 5.719964408874512,
      "epoch": 2.6951176813849447,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042705728581623944,
      "loss": 4.9642,
      "mean_token_accuracy": 0.20667190849781036,
      "num_tokens": 60100829.0,
      "step": 34640
    },
    {
      "entropy": 5.795005655288696,
      "epoch": 2.6955067107566624,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004270368561757452,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20318998247385026,
      "num_tokens": 60109423.0,
      "step": 34645
    },
    {
      "entropy": 5.706937837600708,
      "epoch": 2.6958957401283796,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004270164242283336,
      "loss": 4.8986,
      "mean_token_accuracy": 0.211423921585083,
      "num_tokens": 60117938.0,
      "step": 34650
    },
    {
      "entropy": 5.779661560058594,
      "epoch": 2.6962847695000973,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004269959899743148,
      "loss": 4.9971,
      "mean_token_accuracy": 0.20823391526937485,
      "num_tokens": 60126411.0,
      "step": 34655
    },
    {
      "entropy": 5.813962030410766,
      "epoch": 2.6966737988718146,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004269755534139987,
      "loss": 5.0555,
      "mean_token_accuracy": 0.19742758125066756,
      "num_tokens": 60134811.0,
      "step": 34660
    },
    {
      "entropy": 5.777877521514893,
      "epoch": 2.6970628282435323,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042695511454769557,
      "loss": 4.9773,
      "mean_token_accuracy": 0.20441885590553283,
      "num_tokens": 60143387.0,
      "step": 34665
    },
    {
      "entropy": 5.801006746292114,
      "epoch": 2.69745185761525,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042693467337571534,
      "loss": 5.0135,
      "mean_token_accuracy": 0.2021132379770279,
      "num_tokens": 60152729.0,
      "step": 34670
    },
    {
      "entropy": 5.725065088272094,
      "epoch": 2.6978408869869677,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004269142298983684,
      "loss": 4.9588,
      "mean_token_accuracy": 0.20173279941082,
      "num_tokens": 60160927.0,
      "step": 34675
    },
    {
      "entropy": 5.750795650482178,
      "epoch": 2.698229916358685,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042689378411596467,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21146584749221803,
      "num_tokens": 60169252.0,
      "step": 34680
    },
    {
      "entropy": 5.746347856521607,
      "epoch": 2.6986189457304026,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004268733360288146,
      "loss": 4.9763,
      "mean_token_accuracy": 0.20956062525510788,
      "num_tokens": 60178511.0,
      "step": 34685
    },
    {
      "entropy": 5.735502862930298,
      "epoch": 2.6990079751021203,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042685288563722835,
      "loss": 4.9779,
      "mean_token_accuracy": 0.20367707014083863,
      "num_tokens": 60187619.0,
      "step": 34690
    },
    {
      "entropy": 5.846250009536743,
      "epoch": 2.6993970044738376,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004268324329415163,
      "loss": 5.109,
      "mean_token_accuracy": 0.2001197963953018,
      "num_tokens": 60196176.0,
      "step": 34695
    },
    {
      "entropy": 5.783370590209961,
      "epoch": 2.6997860338455553,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004268119779419887,
      "loss": 5.0095,
      "mean_token_accuracy": 0.20562014132738113,
      "num_tokens": 60204960.0,
      "step": 34700
    },
    {
      "entropy": 5.7230946063995365,
      "epoch": 2.700175063217273,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042679152063895603,
      "loss": 4.9422,
      "mean_token_accuracy": 0.2094443067908287,
      "num_tokens": 60212974.0,
      "step": 34705
    },
    {
      "entropy": 5.75137939453125,
      "epoch": 2.7005640925889907,
      "grad_norm": 1.25,
      "learning_rate": 0.00042677106103272867,
      "loss": 4.9824,
      "mean_token_accuracy": 0.20360343009233475,
      "num_tokens": 60221789.0,
      "step": 34710
    },
    {
      "entropy": 5.7842326164245605,
      "epoch": 2.700953121960708,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000426750599123617,
      "loss": 5.0194,
      "mean_token_accuracy": 0.20283101499080658,
      "num_tokens": 60230349.0,
      "step": 34715
    },
    {
      "entropy": 5.799442768096924,
      "epoch": 2.7013421513324256,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042673013491193145,
      "loss": 4.9621,
      "mean_token_accuracy": 0.20963309854269027,
      "num_tokens": 60238711.0,
      "step": 34720
    },
    {
      "entropy": 5.768613004684449,
      "epoch": 2.7017311807041433,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042670966839798263,
      "loss": 4.9871,
      "mean_token_accuracy": 0.20390649735927582,
      "num_tokens": 60247210.0,
      "step": 34725
    },
    {
      "entropy": 5.798851346969604,
      "epoch": 2.7021202100758606,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004266891995820811,
      "loss": 5.0066,
      "mean_token_accuracy": 0.20574664324522018,
      "num_tokens": 60255916.0,
      "step": 34730
    },
    {
      "entropy": 5.774350833892822,
      "epoch": 2.7025092394475783,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042666872846453734,
      "loss": 4.9944,
      "mean_token_accuracy": 0.2069736123085022,
      "num_tokens": 60264138.0,
      "step": 34735
    },
    {
      "entropy": 5.800201177597046,
      "epoch": 2.702898268819296,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042664825504566206,
      "loss": 5.0325,
      "mean_token_accuracy": 0.19991787970066072,
      "num_tokens": 60272918.0,
      "step": 34740
    },
    {
      "entropy": 5.705561637878418,
      "epoch": 2.7032872981910137,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000426627779325766,
      "loss": 4.7845,
      "mean_token_accuracy": 0.2252723827958107,
      "num_tokens": 60281328.0,
      "step": 34745
    },
    {
      "entropy": 5.7498589038848875,
      "epoch": 2.703676327562731,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004266073013051595,
      "loss": 5.031,
      "mean_token_accuracy": 0.20136120468378066,
      "num_tokens": 60290005.0,
      "step": 34750
    },
    {
      "entropy": 5.728151607513428,
      "epoch": 2.7040653569344486,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004265868209841537,
      "loss": 4.9215,
      "mean_token_accuracy": 0.20578172504901887,
      "num_tokens": 60298225.0,
      "step": 34755
    },
    {
      "entropy": 5.757472276687622,
      "epoch": 2.704454386306166,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004265663383630591,
      "loss": 5.006,
      "mean_token_accuracy": 0.20349553674459459,
      "num_tokens": 60306700.0,
      "step": 34760
    },
    {
      "entropy": 5.779485893249512,
      "epoch": 2.7048434156778836,
      "grad_norm": 1.375,
      "learning_rate": 0.00042654585344218657,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20820042937994004,
      "num_tokens": 60314405.0,
      "step": 34765
    },
    {
      "entropy": 5.788700532913208,
      "epoch": 2.7052324450496013,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004265253662218469,
      "loss": 4.9293,
      "mean_token_accuracy": 0.21110981404781343,
      "num_tokens": 60323159.0,
      "step": 34770
    },
    {
      "entropy": 5.837171936035157,
      "epoch": 2.705621474421319,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042650487670235093,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20528038442134858,
      "num_tokens": 60331669.0,
      "step": 34775
    },
    {
      "entropy": 5.777014255523682,
      "epoch": 2.7060105037930366,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004264843848840097,
      "loss": 4.997,
      "mean_token_accuracy": 0.20669656693935395,
      "num_tokens": 60340841.0,
      "step": 34780
    },
    {
      "entropy": 5.688830375671387,
      "epoch": 2.706399533164754,
      "grad_norm": 1.1875,
      "learning_rate": 0.000426463890767134,
      "loss": 4.8858,
      "mean_token_accuracy": 0.21295156925916672,
      "num_tokens": 60349263.0,
      "step": 34785
    },
    {
      "entropy": 5.777228212356567,
      "epoch": 2.7067885625364716,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004264433943520349,
      "loss": 5.0506,
      "mean_token_accuracy": 0.2017501875758171,
      "num_tokens": 60358214.0,
      "step": 34790
    },
    {
      "entropy": 5.795553159713745,
      "epoch": 2.707177591908189,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042642289563902326,
      "loss": 4.9879,
      "mean_token_accuracy": 0.207388336956501,
      "num_tokens": 60367392.0,
      "step": 34795
    },
    {
      "entropy": 5.760332536697388,
      "epoch": 2.7075666212799065,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042640239462841023,
      "loss": 4.9916,
      "mean_token_accuracy": 0.2058045208454132,
      "num_tokens": 60376223.0,
      "step": 34800
    },
    {
      "entropy": 5.839930486679077,
      "epoch": 2.7079556506516242,
      "grad_norm": 1.25,
      "learning_rate": 0.00042638189132050675,
      "loss": 5.1189,
      "mean_token_accuracy": 0.19161401689052582,
      "num_tokens": 60385102.0,
      "step": 34805
    },
    {
      "entropy": 5.7897789478302,
      "epoch": 2.708344680023342,
      "grad_norm": 1.25,
      "learning_rate": 0.00042636138571562415,
      "loss": 4.9788,
      "mean_token_accuracy": 0.20586809664964675,
      "num_tokens": 60393115.0,
      "step": 34810
    },
    {
      "entropy": 5.760279178619385,
      "epoch": 2.708733709395059,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042634087781407337,
      "loss": 5.0977,
      "mean_token_accuracy": 0.20114169269800186,
      "num_tokens": 60402185.0,
      "step": 34815
    },
    {
      "entropy": 5.883836030960083,
      "epoch": 2.709122738766777,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042632036761616564,
      "loss": 5.0122,
      "mean_token_accuracy": 0.20251393169164658,
      "num_tokens": 60410420.0,
      "step": 34820
    },
    {
      "entropy": 5.8173009872436525,
      "epoch": 2.7095117681384946,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042629985512221224,
      "loss": 4.9659,
      "mean_token_accuracy": 0.21368215084075928,
      "num_tokens": 60419300.0,
      "step": 34825
    },
    {
      "entropy": 5.802827548980713,
      "epoch": 2.709900797510212,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004262793403325243,
      "loss": 4.9886,
      "mean_token_accuracy": 0.21659869700670242,
      "num_tokens": 60428108.0,
      "step": 34830
    },
    {
      "entropy": 5.738508605957032,
      "epoch": 2.7102898268819295,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042625882324741316,
      "loss": 4.9852,
      "mean_token_accuracy": 0.20390155613422395,
      "num_tokens": 60436829.0,
      "step": 34835
    },
    {
      "entropy": 5.800957536697387,
      "epoch": 2.7106788562536472,
      "grad_norm": 1.25,
      "learning_rate": 0.0004262383038671902,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20480151176452638,
      "num_tokens": 60445681.0,
      "step": 34840
    },
    {
      "entropy": 5.8467278480529785,
      "epoch": 2.711067885625365,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004262177821921666,
      "loss": 5.0348,
      "mean_token_accuracy": 0.20910994708538055,
      "num_tokens": 60454263.0,
      "step": 34845
    },
    {
      "entropy": 5.818915224075317,
      "epoch": 2.711456914997082,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042619725822265394,
      "loss": 5.1074,
      "mean_token_accuracy": 0.19942054897546768,
      "num_tokens": 60463569.0,
      "step": 34850
    },
    {
      "entropy": 5.727205085754394,
      "epoch": 2.7118459443688,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042617673195896345,
      "loss": 4.9369,
      "mean_token_accuracy": 0.2060912162065506,
      "num_tokens": 60471458.0,
      "step": 34855
    },
    {
      "entropy": 5.823347806930542,
      "epoch": 2.712234973740517,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042615620340140674,
      "loss": 5.061,
      "mean_token_accuracy": 0.20362839996814727,
      "num_tokens": 60480581.0,
      "step": 34860
    },
    {
      "entropy": 5.81991868019104,
      "epoch": 2.712624003112235,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042613567255029525,
      "loss": 5.0461,
      "mean_token_accuracy": 0.200668540596962,
      "num_tokens": 60488879.0,
      "step": 34865
    },
    {
      "entropy": 5.773199605941772,
      "epoch": 2.7130130324839525,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042611513940594037,
      "loss": 5.0006,
      "mean_token_accuracy": 0.20738014578819275,
      "num_tokens": 60497294.0,
      "step": 34870
    },
    {
      "entropy": 5.754495763778687,
      "epoch": 2.7134020618556702,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004260946039686539,
      "loss": 4.9375,
      "mean_token_accuracy": 0.21688595712184905,
      "num_tokens": 60506449.0,
      "step": 34875
    },
    {
      "entropy": 5.83273229598999,
      "epoch": 2.713791091227388,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042607406623874726,
      "loss": 5.0147,
      "mean_token_accuracy": 0.2067901834845543,
      "num_tokens": 60515318.0,
      "step": 34880
    },
    {
      "entropy": 5.756293916702271,
      "epoch": 2.714180120599105,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004260535262165322,
      "loss": 5.0739,
      "mean_token_accuracy": 0.20100174099206924,
      "num_tokens": 60525425.0,
      "step": 34885
    },
    {
      "entropy": 5.761459255218506,
      "epoch": 2.714569149970823,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042603298390232015,
      "loss": 5.012,
      "mean_token_accuracy": 0.2013502299785614,
      "num_tokens": 60535028.0,
      "step": 34890
    },
    {
      "entropy": 5.7942695140838625,
      "epoch": 2.71495817934254,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042601243929642305,
      "loss": 4.9545,
      "mean_token_accuracy": 0.21280696243047714,
      "num_tokens": 60543076.0,
      "step": 34895
    },
    {
      "entropy": 5.779795455932617,
      "epoch": 2.715347208714258,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042599189239915253,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20492010414600373,
      "num_tokens": 60551697.0,
      "step": 34900
    },
    {
      "entropy": 5.731320190429687,
      "epoch": 2.7157362380859755,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004259713432108204,
      "loss": 4.9448,
      "mean_token_accuracy": 0.20635045319795609,
      "num_tokens": 60559910.0,
      "step": 34905
    },
    {
      "entropy": 5.861095714569092,
      "epoch": 2.7161252674576932,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004259507917317384,
      "loss": 5.0901,
      "mean_token_accuracy": 0.1997839853167534,
      "num_tokens": 60568717.0,
      "step": 34910
    },
    {
      "entropy": 5.845838832855224,
      "epoch": 2.7165142968294105,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042593023796221843,
      "loss": 5.0112,
      "mean_token_accuracy": 0.20307700484991073,
      "num_tokens": 60576707.0,
      "step": 34915
    },
    {
      "entropy": 5.7930680274963375,
      "epoch": 2.716903326201128,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004259096819025723,
      "loss": 4.9692,
      "mean_token_accuracy": 0.20571343153715133,
      "num_tokens": 60585097.0,
      "step": 34920
    },
    {
      "entropy": 5.75047459602356,
      "epoch": 2.717292355572846,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042588912355311196,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20903354734182358,
      "num_tokens": 60594026.0,
      "step": 34925
    },
    {
      "entropy": 5.768545484542846,
      "epoch": 2.717681384944563,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042586856291414925,
      "loss": 4.9827,
      "mean_token_accuracy": 0.20313398987054826,
      "num_tokens": 60602281.0,
      "step": 34930
    },
    {
      "entropy": 5.74733600616455,
      "epoch": 2.718070414316281,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042584799998599633,
      "loss": 4.9657,
      "mean_token_accuracy": 0.2090936481952667,
      "num_tokens": 60610329.0,
      "step": 34935
    },
    {
      "entropy": 5.756964492797851,
      "epoch": 2.7184594436879985,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004258274347689651,
      "loss": 5.0493,
      "mean_token_accuracy": 0.20791924744844437,
      "num_tokens": 60618251.0,
      "step": 34940
    },
    {
      "entropy": 5.859614419937134,
      "epoch": 2.718848473059716,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042580686726336766,
      "loss": 5.0256,
      "mean_token_accuracy": 0.19635428339242936,
      "num_tokens": 60626479.0,
      "step": 34945
    },
    {
      "entropy": 5.810048246383667,
      "epoch": 2.7192375024314335,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042578629746951597,
      "loss": 4.9714,
      "mean_token_accuracy": 0.2110928937792778,
      "num_tokens": 60635375.0,
      "step": 34950
    },
    {
      "entropy": 5.781143760681152,
      "epoch": 2.719626531803151,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004257657253877222,
      "loss": 5.0088,
      "mean_token_accuracy": 0.20841973572969436,
      "num_tokens": 60644136.0,
      "step": 34955
    },
    {
      "entropy": 5.831383848190308,
      "epoch": 2.720015561174869,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042574515101829856,
      "loss": 5.0839,
      "mean_token_accuracy": 0.21121550351381302,
      "num_tokens": 60653258.0,
      "step": 34960
    },
    {
      "entropy": 5.809576988220215,
      "epoch": 2.720404590546586,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042572457436155717,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20745818465948104,
      "num_tokens": 60662320.0,
      "step": 34965
    },
    {
      "entropy": 5.753822088241577,
      "epoch": 2.720793619918304,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042570399541781023,
      "loss": 4.9732,
      "mean_token_accuracy": 0.2098812848329544,
      "num_tokens": 60671669.0,
      "step": 34970
    },
    {
      "entropy": 5.790738153457641,
      "epoch": 2.7211826492900215,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042568341418737016,
      "loss": 5.1303,
      "mean_token_accuracy": 0.19656278043985367,
      "num_tokens": 60680391.0,
      "step": 34975
    },
    {
      "entropy": 5.804852104187011,
      "epoch": 2.721571678661739,
      "grad_norm": 1.15625,
      "learning_rate": 0.000425662830670549,
      "loss": 5.0362,
      "mean_token_accuracy": 0.2017245277762413,
      "num_tokens": 60690150.0,
      "step": 34980
    },
    {
      "entropy": 5.8036939144134525,
      "epoch": 2.7219607080334565,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042564224486765925,
      "loss": 4.9039,
      "mean_token_accuracy": 0.21682696044445038,
      "num_tokens": 60698950.0,
      "step": 34985
    },
    {
      "entropy": 5.878361988067627,
      "epoch": 2.722349737405174,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004256216567790131,
      "loss": 5.0397,
      "mean_token_accuracy": 0.19976284801959993,
      "num_tokens": 60707859.0,
      "step": 34990
    },
    {
      "entropy": 5.817148876190186,
      "epoch": 2.7227387667768914,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004256010664049232,
      "loss": 5.0141,
      "mean_token_accuracy": 0.20796209275722505,
      "num_tokens": 60716621.0,
      "step": 34995
    },
    {
      "entropy": 5.8363272190094,
      "epoch": 2.723127796148609,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004255804737457017,
      "loss": 5.132,
      "mean_token_accuracy": 0.19978080242872237,
      "num_tokens": 60725695.0,
      "step": 35000
    },
    {
      "entropy": 5.785414171218872,
      "epoch": 2.723516825520327,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042555987880166127,
      "loss": 4.9752,
      "mean_token_accuracy": 0.2088310167193413,
      "num_tokens": 60733978.0,
      "step": 35005
    },
    {
      "entropy": 5.773030090332031,
      "epoch": 2.7239058548920445,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004255392815731142,
      "loss": 4.9912,
      "mean_token_accuracy": 0.2068332239985466,
      "num_tokens": 60742391.0,
      "step": 35010
    },
    {
      "entropy": 5.781872892379761,
      "epoch": 2.7242948842637618,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042551868206037324,
      "loss": 4.9639,
      "mean_token_accuracy": 0.21750523149967194,
      "num_tokens": 60749808.0,
      "step": 35015
    },
    {
      "entropy": 5.748452615737915,
      "epoch": 2.7246839136354795,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042549808026375084,
      "loss": 5.0929,
      "mean_token_accuracy": 0.20555961281061172,
      "num_tokens": 60759902.0,
      "step": 35020
    },
    {
      "entropy": 5.7715624332427975,
      "epoch": 2.725072943007197,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042547747618355966,
      "loss": 5.0351,
      "mean_token_accuracy": 0.19903212636709214,
      "num_tokens": 60768542.0,
      "step": 35025
    },
    {
      "entropy": 5.7900224208831785,
      "epoch": 2.7254619723789144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004254568698201122,
      "loss": 4.9658,
      "mean_token_accuracy": 0.2161123663187027,
      "num_tokens": 60777077.0,
      "step": 35030
    },
    {
      "entropy": 5.772587871551513,
      "epoch": 2.725851001750632,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042543626117372133,
      "loss": 4.936,
      "mean_token_accuracy": 0.20868244022130966,
      "num_tokens": 60786520.0,
      "step": 35035
    },
    {
      "entropy": 5.830029153823853,
      "epoch": 2.72624003112235,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042541565024469945,
      "loss": 5.0382,
      "mean_token_accuracy": 0.20879154205322265,
      "num_tokens": 60795270.0,
      "step": 35040
    },
    {
      "entropy": 5.816509771347046,
      "epoch": 2.7266290604940675,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042539503703335963,
      "loss": 5.018,
      "mean_token_accuracy": 0.19818613380193711,
      "num_tokens": 60804047.0,
      "step": 35045
    },
    {
      "entropy": 5.769230937957763,
      "epoch": 2.7270180898657848,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042537442154001455,
      "loss": 4.9689,
      "mean_token_accuracy": 0.20399209409952163,
      "num_tokens": 60812376.0,
      "step": 35050
    },
    {
      "entropy": 5.851846361160279,
      "epoch": 2.7274071192375025,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042535380376497684,
      "loss": 5.1318,
      "mean_token_accuracy": 0.1975157544016838,
      "num_tokens": 60822075.0,
      "step": 35055
    },
    {
      "entropy": 5.817855167388916,
      "epoch": 2.72779614860922,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004253331837085595,
      "loss": 5.0359,
      "mean_token_accuracy": 0.20430450588464738,
      "num_tokens": 60830729.0,
      "step": 35060
    },
    {
      "entropy": 5.71275520324707,
      "epoch": 2.7281851779809374,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042531256137107545,
      "loss": 4.8958,
      "mean_token_accuracy": 0.21357685476541519,
      "num_tokens": 60840429.0,
      "step": 35065
    },
    {
      "entropy": 5.790626096725464,
      "epoch": 2.728574207352655,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004252919367528375,
      "loss": 4.889,
      "mean_token_accuracy": 0.21457489132881163,
      "num_tokens": 60848967.0,
      "step": 35070
    },
    {
      "entropy": 5.771803474426269,
      "epoch": 2.728963236724373,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042527130985415857,
      "loss": 4.9688,
      "mean_token_accuracy": 0.2074423089623451,
      "num_tokens": 60856718.0,
      "step": 35075
    },
    {
      "entropy": 5.739031839370727,
      "epoch": 2.7293522660960905,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004252506806753517,
      "loss": 4.89,
      "mean_token_accuracy": 0.21835147142410277,
      "num_tokens": 60865338.0,
      "step": 35080
    },
    {
      "entropy": 5.733790302276612,
      "epoch": 2.7297412954678077,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042523004921673,
      "loss": 5.0098,
      "mean_token_accuracy": 0.20610920041799546,
      "num_tokens": 60873887.0,
      "step": 35085
    },
    {
      "entropy": 5.804330587387085,
      "epoch": 2.7301303248395254,
      "grad_norm": 1.25,
      "learning_rate": 0.0004252094154786063,
      "loss": 5.0576,
      "mean_token_accuracy": 0.2022850975394249,
      "num_tokens": 60882755.0,
      "step": 35090
    },
    {
      "entropy": 5.766473197937012,
      "epoch": 2.7305193542112427,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004251887794612939,
      "loss": 4.9001,
      "mean_token_accuracy": 0.21452568620443344,
      "num_tokens": 60891191.0,
      "step": 35095
    },
    {
      "entropy": 5.830274868011474,
      "epoch": 2.7309083835829604,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004251681411651057,
      "loss": 5.088,
      "mean_token_accuracy": 0.2016287237405777,
      "num_tokens": 60899634.0,
      "step": 35100
    },
    {
      "entropy": 5.801162576675415,
      "epoch": 2.731297412954678,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042514750059035504,
      "loss": 5.0663,
      "mean_token_accuracy": 0.2055305927991867,
      "num_tokens": 60908468.0,
      "step": 35105
    },
    {
      "entropy": 5.7757611751556395,
      "epoch": 2.731686442326396,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004251268577373551,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20415472835302353,
      "num_tokens": 60917102.0,
      "step": 35110
    },
    {
      "entropy": 5.875105857849121,
      "epoch": 2.732075471698113,
      "grad_norm": 1.28125,
      "learning_rate": 0.000425106212606419,
      "loss": 5.0908,
      "mean_token_accuracy": 0.20788992643356324,
      "num_tokens": 60926410.0,
      "step": 35115
    },
    {
      "entropy": 5.805799198150635,
      "epoch": 2.7324645010698307,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042508556519786004,
      "loss": 4.9539,
      "mean_token_accuracy": 0.20563670992851257,
      "num_tokens": 60935068.0,
      "step": 35120
    },
    {
      "entropy": 5.822583198547363,
      "epoch": 2.7328535304415484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004250649155119915,
      "loss": 4.9886,
      "mean_token_accuracy": 0.20363896042108537,
      "num_tokens": 60943507.0,
      "step": 35125
    },
    {
      "entropy": 5.791417837142944,
      "epoch": 2.7332425598132657,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042504426354912666,
      "loss": 4.9729,
      "mean_token_accuracy": 0.20575102269649506,
      "num_tokens": 60952175.0,
      "step": 35130
    },
    {
      "entropy": 5.705262041091919,
      "epoch": 2.7336315891849834,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042502360930957904,
      "loss": 4.9635,
      "mean_token_accuracy": 0.21023884862661363,
      "num_tokens": 60962087.0,
      "step": 35135
    },
    {
      "entropy": 5.7573048114776615,
      "epoch": 2.734020618556701,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042500295279366196,
      "loss": 4.979,
      "mean_token_accuracy": 0.20761753767728805,
      "num_tokens": 60970142.0,
      "step": 35140
    },
    {
      "entropy": 5.792966270446778,
      "epoch": 2.734409647928419,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042498229400168874,
      "loss": 5.0447,
      "mean_token_accuracy": 0.2029822751879692,
      "num_tokens": 60979498.0,
      "step": 35145
    },
    {
      "entropy": 5.79144983291626,
      "epoch": 2.734798677300136,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000424961632933973,
      "loss": 5.0088,
      "mean_token_accuracy": 0.2044406861066818,
      "num_tokens": 60988382.0,
      "step": 35150
    },
    {
      "entropy": 5.801993799209595,
      "epoch": 2.7351877066718537,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004249409695908281,
      "loss": 5.0918,
      "mean_token_accuracy": 0.19698527455329895,
      "num_tokens": 60998191.0,
      "step": 35155
    },
    {
      "entropy": 5.779683685302734,
      "epoch": 2.7355767360435714,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004249203039725677,
      "loss": 5.0318,
      "mean_token_accuracy": 0.2073124974966049,
      "num_tokens": 61006417.0,
      "step": 35160
    },
    {
      "entropy": 5.741063261032105,
      "epoch": 2.7359657654152887,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042489963607950527,
      "loss": 4.8911,
      "mean_token_accuracy": 0.21582740247249604,
      "num_tokens": 61014562.0,
      "step": 35165
    },
    {
      "entropy": 5.759390830993652,
      "epoch": 2.7363547947870064,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042487896591195454,
      "loss": 4.9607,
      "mean_token_accuracy": 0.20245912075042724,
      "num_tokens": 61022139.0,
      "step": 35170
    },
    {
      "entropy": 5.765931129455566,
      "epoch": 2.736743824158724,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042485829347022893,
      "loss": 5.1222,
      "mean_token_accuracy": 0.1992650255560875,
      "num_tokens": 61031099.0,
      "step": 35175
    },
    {
      "entropy": 5.842671155929565,
      "epoch": 2.737132853530442,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042483761875464226,
      "loss": 5.0638,
      "mean_token_accuracy": 0.20458708852529525,
      "num_tokens": 61039136.0,
      "step": 35180
    },
    {
      "entropy": 5.763128709793091,
      "epoch": 2.737521882902159,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042481694176550825,
      "loss": 4.9701,
      "mean_token_accuracy": 0.2096380412578583,
      "num_tokens": 61047177.0,
      "step": 35185
    },
    {
      "entropy": 5.760892152786255,
      "epoch": 2.7379109122738767,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042479626250314066,
      "loss": 4.9651,
      "mean_token_accuracy": 0.20235661566257476,
      "num_tokens": 61055383.0,
      "step": 35190
    },
    {
      "entropy": 5.768124771118164,
      "epoch": 2.738299941645594,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042477558096785316,
      "loss": 4.9604,
      "mean_token_accuracy": 0.21137823313474655,
      "num_tokens": 61063831.0,
      "step": 35195
    },
    {
      "entropy": 5.862046527862549,
      "epoch": 2.7386889710173117,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042475489715995956,
      "loss": 5.0815,
      "mean_token_accuracy": 0.1968831464648247,
      "num_tokens": 61072328.0,
      "step": 35200
    },
    {
      "entropy": 5.762005376815796,
      "epoch": 2.7390780003890294,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042473421107977385,
      "loss": 4.908,
      "mean_token_accuracy": 0.2119050681591034,
      "num_tokens": 61081191.0,
      "step": 35205
    },
    {
      "entropy": 5.7792126655578615,
      "epoch": 2.739467029760747,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042471352272760974,
      "loss": 4.9979,
      "mean_token_accuracy": 0.21172756105661392,
      "num_tokens": 61089314.0,
      "step": 35210
    },
    {
      "entropy": 5.6863486766815186,
      "epoch": 2.7398560591324648,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004246928321037813,
      "loss": 4.9383,
      "mean_token_accuracy": 0.21088102012872695,
      "num_tokens": 61098322.0,
      "step": 35215
    },
    {
      "entropy": 5.901244640350342,
      "epoch": 2.740245088504182,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004246721392086023,
      "loss": 5.1303,
      "mean_token_accuracy": 0.19525084048509597,
      "num_tokens": 61106682.0,
      "step": 35220
    },
    {
      "entropy": 5.829957962036133,
      "epoch": 2.7406341178758997,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042465144404238686,
      "loss": 4.9998,
      "mean_token_accuracy": 0.20137352198362352,
      "num_tokens": 61115765.0,
      "step": 35225
    },
    {
      "entropy": 5.701172304153443,
      "epoch": 2.741023147247617,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042463074660544897,
      "loss": 4.8874,
      "mean_token_accuracy": 0.21093329936265945,
      "num_tokens": 61124464.0,
      "step": 35230
    },
    {
      "entropy": 5.805878162384033,
      "epoch": 2.7414121766193347,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004246100468981026,
      "loss": 5.0907,
      "mean_token_accuracy": 0.19932863861322403,
      "num_tokens": 61133386.0,
      "step": 35235
    },
    {
      "entropy": 5.906299495697022,
      "epoch": 2.7418012059910524,
      "grad_norm": 1.265625,
      "learning_rate": 0.000424589344920662,
      "loss": 5.0982,
      "mean_token_accuracy": 0.1961233749985695,
      "num_tokens": 61141873.0,
      "step": 35240
    },
    {
      "entropy": 5.810507965087891,
      "epoch": 2.74219023536277,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004245686406734411,
      "loss": 4.995,
      "mean_token_accuracy": 0.20480487048625945,
      "num_tokens": 61150113.0,
      "step": 35245
    },
    {
      "entropy": 5.711752700805664,
      "epoch": 2.7425792647344873,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004245479341567541,
      "loss": 4.9068,
      "mean_token_accuracy": 0.20872446447610854,
      "num_tokens": 61158941.0,
      "step": 35250
    },
    {
      "entropy": 5.7568871021270756,
      "epoch": 2.742968294106205,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004245272253709154,
      "loss": 4.992,
      "mean_token_accuracy": 0.21075983792543412,
      "num_tokens": 61167978.0,
      "step": 35255
    },
    {
      "entropy": 5.782249498367309,
      "epoch": 2.7433573234779227,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004245065143162389,
      "loss": 5.0867,
      "mean_token_accuracy": 0.20218978375196456,
      "num_tokens": 61176581.0,
      "step": 35260
    },
    {
      "entropy": 5.879281663894654,
      "epoch": 2.74374635284964,
      "grad_norm": 1.25,
      "learning_rate": 0.00042448580099303904,
      "loss": 5.1327,
      "mean_token_accuracy": 0.1960772380232811,
      "num_tokens": 61186072.0,
      "step": 35265
    },
    {
      "entropy": 5.910063219070435,
      "epoch": 2.7441353822213577,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042446508540163017,
      "loss": 5.0299,
      "mean_token_accuracy": 0.19997680485248565,
      "num_tokens": 61194632.0,
      "step": 35270
    },
    {
      "entropy": 5.729909801483155,
      "epoch": 2.7445244115930754,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004244443675423264,
      "loss": 4.9452,
      "mean_token_accuracy": 0.21285128295421601,
      "num_tokens": 61203705.0,
      "step": 35275
    },
    {
      "entropy": 5.739958620071411,
      "epoch": 2.744913440964793,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042442364741544227,
      "loss": 4.9608,
      "mean_token_accuracy": 0.20530932396650314,
      "num_tokens": 61212565.0,
      "step": 35280
    },
    {
      "entropy": 5.782778739929199,
      "epoch": 2.7453024703365103,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042440292502129214,
      "loss": 4.9945,
      "mean_token_accuracy": 0.20230991244316102,
      "num_tokens": 61221092.0,
      "step": 35285
    },
    {
      "entropy": 5.806141376495361,
      "epoch": 2.745691499708228,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004243822003601904,
      "loss": 5.0004,
      "mean_token_accuracy": 0.2026850923895836,
      "num_tokens": 61229880.0,
      "step": 35290
    },
    {
      "entropy": 5.779419946670532,
      "epoch": 2.7460805290799453,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004243614734324515,
      "loss": 4.9941,
      "mean_token_accuracy": 0.20756012052297593,
      "num_tokens": 61238890.0,
      "step": 35295
    },
    {
      "entropy": 5.777364110946655,
      "epoch": 2.746469558451663,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004243407442383901,
      "loss": 4.9878,
      "mean_token_accuracy": 0.20178650170564652,
      "num_tokens": 61248260.0,
      "step": 35300
    },
    {
      "entropy": 5.810210275650024,
      "epoch": 2.7468585878233807,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042432001277832045,
      "loss": 4.9683,
      "mean_token_accuracy": 0.206354358792305,
      "num_tokens": 61256345.0,
      "step": 35305
    },
    {
      "entropy": 5.801637649536133,
      "epoch": 2.7472476171950984,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004242992790525573,
      "loss": 4.996,
      "mean_token_accuracy": 0.20489351004362105,
      "num_tokens": 61264952.0,
      "step": 35310
    },
    {
      "entropy": 5.830531167984009,
      "epoch": 2.747636646566816,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004242785430614154,
      "loss": 5.0312,
      "mean_token_accuracy": 0.2030854269862175,
      "num_tokens": 61273596.0,
      "step": 35315
    },
    {
      "entropy": 5.782908201217651,
      "epoch": 2.7480256759385333,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042425780480520906,
      "loss": 4.9489,
      "mean_token_accuracy": 0.2097545251250267,
      "num_tokens": 61281799.0,
      "step": 35320
    },
    {
      "entropy": 5.7766059875488285,
      "epoch": 2.748414705310251,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042423706428425305,
      "loss": 4.9312,
      "mean_token_accuracy": 0.20895693451166153,
      "num_tokens": 61290295.0,
      "step": 35325
    },
    {
      "entropy": 5.748567199707031,
      "epoch": 2.7488037346819683,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004242163214988622,
      "loss": 5.0066,
      "mean_token_accuracy": 0.21176577657461165,
      "num_tokens": 61298745.0,
      "step": 35330
    },
    {
      "entropy": 5.868057441711426,
      "epoch": 2.749192764053686,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004241955764493512,
      "loss": 5.0947,
      "mean_token_accuracy": 0.1945350006222725,
      "num_tokens": 61307472.0,
      "step": 35335
    },
    {
      "entropy": 5.843629598617554,
      "epoch": 2.7495817934254037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042417482913603477,
      "loss": 5.0935,
      "mean_token_accuracy": 0.19627974778413773,
      "num_tokens": 61317633.0,
      "step": 35340
    },
    {
      "entropy": 5.793390464782715,
      "epoch": 2.7499708227971214,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042415407955922774,
      "loss": 5.0336,
      "mean_token_accuracy": 0.20190190970897676,
      "num_tokens": 61326836.0,
      "step": 35345
    },
    {
      "entropy": 5.829093265533447,
      "epoch": 2.7503598521688386,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042413332771924496,
      "loss": 5.0374,
      "mean_token_accuracy": 0.20668653696775435,
      "num_tokens": 61335817.0,
      "step": 35350
    },
    {
      "entropy": 5.850252914428711,
      "epoch": 2.7507488815405563,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004241125736164013,
      "loss": 5.1689,
      "mean_token_accuracy": 0.1976330801844597,
      "num_tokens": 61345030.0,
      "step": 35355
    },
    {
      "entropy": 5.795464324951172,
      "epoch": 2.751137910912274,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004240918172510117,
      "loss": 5.0243,
      "mean_token_accuracy": 0.20263781100511552,
      "num_tokens": 61353061.0,
      "step": 35360
    },
    {
      "entropy": 5.884984970092773,
      "epoch": 2.7515269402839913,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000424071058623391,
      "loss": 5.1214,
      "mean_token_accuracy": 0.20143271535634993,
      "num_tokens": 61361357.0,
      "step": 35365
    },
    {
      "entropy": 5.8445971488952635,
      "epoch": 2.751915969655709,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004240502977338543,
      "loss": 5.0956,
      "mean_token_accuracy": 0.19450919330120087,
      "num_tokens": 61369731.0,
      "step": 35370
    },
    {
      "entropy": 5.789604568481446,
      "epoch": 2.7523049990274266,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004240295345827166,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20761335790157318,
      "num_tokens": 61378212.0,
      "step": 35375
    },
    {
      "entropy": 5.780000925064087,
      "epoch": 2.7526940283991443,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004240087691702929,
      "loss": 4.9608,
      "mean_token_accuracy": 0.20778839886188508,
      "num_tokens": 61386743.0,
      "step": 35380
    },
    {
      "entropy": 5.808331632614136,
      "epoch": 2.7530830577708616,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004239880014968983,
      "loss": 5.0115,
      "mean_token_accuracy": 0.2014694780111313,
      "num_tokens": 61394513.0,
      "step": 35385
    },
    {
      "entropy": 5.7995765686035154,
      "epoch": 2.7534720871425793,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042396723156284786,
      "loss": 4.9961,
      "mean_token_accuracy": 0.2066247880458832,
      "num_tokens": 61403999.0,
      "step": 35390
    },
    {
      "entropy": 5.808501148223877,
      "epoch": 2.753861116514297,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004239464593684568,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21023458540439605,
      "num_tokens": 61413363.0,
      "step": 35395
    },
    {
      "entropy": 5.733277463912964,
      "epoch": 2.7542501458860142,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042392568491404043,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21105723232030868,
      "num_tokens": 61421749.0,
      "step": 35400
    },
    {
      "entropy": 5.783891868591309,
      "epoch": 2.754639175257732,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042390490819991367,
      "loss": 5.0743,
      "mean_token_accuracy": 0.20296269059181213,
      "num_tokens": 61430422.0,
      "step": 35405
    },
    {
      "entropy": 5.841206121444702,
      "epoch": 2.7550282046294496,
      "grad_norm": 1.1875,
      "learning_rate": 0.000423884129226392,
      "loss": 5.0069,
      "mean_token_accuracy": 0.20465365648269654,
      "num_tokens": 61439402.0,
      "step": 35410
    },
    {
      "entropy": 5.833832788467407,
      "epoch": 2.7554172340011673,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004238633479937906,
      "loss": 5.0802,
      "mean_token_accuracy": 0.2014279246330261,
      "num_tokens": 61448791.0,
      "step": 35415
    },
    {
      "entropy": 5.728889036178589,
      "epoch": 2.7558062633728846,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004238425645024249,
      "loss": 4.9178,
      "mean_token_accuracy": 0.20450449436903,
      "num_tokens": 61457181.0,
      "step": 35420
    },
    {
      "entropy": 5.8445783138275145,
      "epoch": 2.7561952927446023,
      "grad_norm": 1.25,
      "learning_rate": 0.0004238217787526102,
      "loss": 5.0487,
      "mean_token_accuracy": 0.20281512588262557,
      "num_tokens": 61465525.0,
      "step": 35425
    },
    {
      "entropy": 5.813938617706299,
      "epoch": 2.7565843221163195,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004238009907446618,
      "loss": 5.083,
      "mean_token_accuracy": 0.1973132610321045,
      "num_tokens": 61474805.0,
      "step": 35430
    },
    {
      "entropy": 5.722457504272461,
      "epoch": 2.7569733514880372,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042378020047889527,
      "loss": 4.9039,
      "mean_token_accuracy": 0.20890847593545914,
      "num_tokens": 61484147.0,
      "step": 35435
    },
    {
      "entropy": 5.7638537883758545,
      "epoch": 2.757362380859755,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042375940795562594,
      "loss": 4.8945,
      "mean_token_accuracy": 0.21161660254001619,
      "num_tokens": 61492396.0,
      "step": 35440
    },
    {
      "entropy": 5.738065910339356,
      "epoch": 2.7577514102314726,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042373861317516944,
      "loss": 4.9684,
      "mean_token_accuracy": 0.20043593347072602,
      "num_tokens": 61500941.0,
      "step": 35445
    },
    {
      "entropy": 5.787286758422852,
      "epoch": 2.75814043960319,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004237178161378412,
      "loss": 5.0016,
      "mean_token_accuracy": 0.20926357358694075,
      "num_tokens": 61510245.0,
      "step": 35450
    },
    {
      "entropy": 5.777422714233398,
      "epoch": 2.7585294689749076,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004236970168439568,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21251198053359985,
      "num_tokens": 61519070.0,
      "step": 35455
    },
    {
      "entropy": 5.721857357025146,
      "epoch": 2.7589184983466253,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042367621529383186,
      "loss": 4.8948,
      "mean_token_accuracy": 0.2135111078619957,
      "num_tokens": 61526931.0,
      "step": 35460
    },
    {
      "entropy": 5.854739618301392,
      "epoch": 2.7593075277183425,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042365541148778205,
      "loss": 5.1036,
      "mean_token_accuracy": 0.20403952598571778,
      "num_tokens": 61534645.0,
      "step": 35465
    },
    {
      "entropy": 5.752589130401612,
      "epoch": 2.7596965570900602,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042363460542612285,
      "loss": 4.9454,
      "mean_token_accuracy": 0.20633412152528763,
      "num_tokens": 61542770.0,
      "step": 35470
    },
    {
      "entropy": 5.776446342468262,
      "epoch": 2.760085586461778,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004236137971091702,
      "loss": 4.9997,
      "mean_token_accuracy": 0.20987797379493714,
      "num_tokens": 61551059.0,
      "step": 35475
    },
    {
      "entropy": 5.755829668045044,
      "epoch": 2.7604746158334956,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042359298653723963,
      "loss": 4.8881,
      "mean_token_accuracy": 0.21037476360797883,
      "num_tokens": 61559384.0,
      "step": 35480
    },
    {
      "entropy": 5.712637424468994,
      "epoch": 2.760863645205213,
      "grad_norm": 1.25,
      "learning_rate": 0.0004235721737106471,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2139532223343849,
      "num_tokens": 61568335.0,
      "step": 35485
    },
    {
      "entropy": 5.752257633209228,
      "epoch": 2.7612526745769306,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004235513586297082,
      "loss": 4.9655,
      "mean_token_accuracy": 0.20424951314926149,
      "num_tokens": 61577481.0,
      "step": 35490
    },
    {
      "entropy": 5.758753919601441,
      "epoch": 2.7616417039486483,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042353054129473894,
      "loss": 5.0128,
      "mean_token_accuracy": 0.2024803176522255,
      "num_tokens": 61585921.0,
      "step": 35495
    },
    {
      "entropy": 5.747140312194825,
      "epoch": 2.7620307333203655,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004235097217060552,
      "loss": 4.9653,
      "mean_token_accuracy": 0.20831135660409927,
      "num_tokens": 61594548.0,
      "step": 35500
    },
    {
      "entropy": 5.7798402309417725,
      "epoch": 2.7624197626920832,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004234888998639727,
      "loss": 5.0067,
      "mean_token_accuracy": 0.21114012002944946,
      "num_tokens": 61603165.0,
      "step": 35505
    },
    {
      "entropy": 5.772410249710083,
      "epoch": 2.762808792063801,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004234680757688076,
      "loss": 4.9199,
      "mean_token_accuracy": 0.21001650691032409,
      "num_tokens": 61611920.0,
      "step": 35510
    },
    {
      "entropy": 5.802515697479248,
      "epoch": 2.7631978214355186,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042344724942087564,
      "loss": 4.9841,
      "mean_token_accuracy": 0.20442851781845092,
      "num_tokens": 61620553.0,
      "step": 35515
    },
    {
      "entropy": 5.779647731781006,
      "epoch": 2.763586850807236,
      "grad_norm": 1.28125,
      "learning_rate": 0.000423426420820493,
      "loss": 4.9937,
      "mean_token_accuracy": 0.20715341418981553,
      "num_tokens": 61629319.0,
      "step": 35520
    },
    {
      "entropy": 5.7224067687988285,
      "epoch": 2.7639758801789536,
      "grad_norm": 1.25,
      "learning_rate": 0.00042340558996797564,
      "loss": 4.915,
      "mean_token_accuracy": 0.213382525742054,
      "num_tokens": 61638475.0,
      "step": 35525
    },
    {
      "entropy": 5.779732036590576,
      "epoch": 2.764364909550671,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042338475686363964,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20895246267318726,
      "num_tokens": 61647345.0,
      "step": 35530
    },
    {
      "entropy": 5.854640007019043,
      "epoch": 2.7647539389223885,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004233639215078012,
      "loss": 5.0239,
      "mean_token_accuracy": 0.20542342513799666,
      "num_tokens": 61655871.0,
      "step": 35535
    },
    {
      "entropy": 5.832192707061767,
      "epoch": 2.765142968294106,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042334308390077637,
      "loss": 5.0241,
      "mean_token_accuracy": 0.2040394812822342,
      "num_tokens": 61664429.0,
      "step": 35540
    },
    {
      "entropy": 5.704099369049072,
      "epoch": 2.765531997665824,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004233222440428814,
      "loss": 4.9558,
      "mean_token_accuracy": 0.21007076352834703,
      "num_tokens": 61671980.0,
      "step": 35545
    },
    {
      "entropy": 5.753475093841553,
      "epoch": 2.765921027037541,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004233014019344324,
      "loss": 4.9124,
      "mean_token_accuracy": 0.2124169021844864,
      "num_tokens": 61680140.0,
      "step": 35550
    },
    {
      "entropy": 5.770744895935058,
      "epoch": 2.766310056409259,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004232805575757457,
      "loss": 4.9431,
      "mean_token_accuracy": 0.20882345587015153,
      "num_tokens": 61688863.0,
      "step": 35555
    },
    {
      "entropy": 5.830897474288941,
      "epoch": 2.7666990857809766,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004232597109671375,
      "loss": 5.0065,
      "mean_token_accuracy": 0.20445864349603654,
      "num_tokens": 61697494.0,
      "step": 35560
    },
    {
      "entropy": 5.767996549606323,
      "epoch": 2.767088115152694,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004232388621089242,
      "loss": 5.0059,
      "mean_token_accuracy": 0.20509446859359742,
      "num_tokens": 61706456.0,
      "step": 35565
    },
    {
      "entropy": 5.793598318099976,
      "epoch": 2.7674771445244115,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042321801100142216,
      "loss": 5.0725,
      "mean_token_accuracy": 0.20837675631046296,
      "num_tokens": 61715123.0,
      "step": 35570
    },
    {
      "entropy": 5.8132486820220945,
      "epoch": 2.767866173896129,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042319715764494765,
      "loss": 4.9327,
      "mean_token_accuracy": 0.20981833189725876,
      "num_tokens": 61724129.0,
      "step": 35575
    },
    {
      "entropy": 5.8294614315032955,
      "epoch": 2.768255203267847,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042317630203981714,
      "loss": 5.0725,
      "mean_token_accuracy": 0.20007632225751876,
      "num_tokens": 61732646.0,
      "step": 35580
    },
    {
      "entropy": 5.824590635299683,
      "epoch": 2.768644232639564,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042315544418634714,
      "loss": 4.9903,
      "mean_token_accuracy": 0.20786398500204087,
      "num_tokens": 61741511.0,
      "step": 35585
    },
    {
      "entropy": 5.813971948623657,
      "epoch": 2.769033262011282,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042313458408485414,
      "loss": 5.0191,
      "mean_token_accuracy": 0.20370755046606065,
      "num_tokens": 61749703.0,
      "step": 35590
    },
    {
      "entropy": 5.77941575050354,
      "epoch": 2.7694222913829996,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042311372173565456,
      "loss": 4.9451,
      "mean_token_accuracy": 0.20955634713172913,
      "num_tokens": 61758943.0,
      "step": 35595
    },
    {
      "entropy": 5.794227981567383,
      "epoch": 2.769811320754717,
      "grad_norm": 1.328125,
      "learning_rate": 0.000423092857139065,
      "loss": 4.9748,
      "mean_token_accuracy": 0.2085822716355324,
      "num_tokens": 61767120.0,
      "step": 35600
    },
    {
      "entropy": 5.782001256942749,
      "epoch": 2.7702003501264345,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000423071990295402,
      "loss": 4.9642,
      "mean_token_accuracy": 0.20487237423658372,
      "num_tokens": 61774760.0,
      "step": 35605
    },
    {
      "entropy": 5.727087450027466,
      "epoch": 2.770589379498152,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004230511212049822,
      "loss": 4.9517,
      "mean_token_accuracy": 0.21297386288642883,
      "num_tokens": 61783792.0,
      "step": 35610
    },
    {
      "entropy": 5.817515087127686,
      "epoch": 2.77097840886987,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004230302498681224,
      "loss": 5.0754,
      "mean_token_accuracy": 0.2020127981901169,
      "num_tokens": 61791991.0,
      "step": 35615
    },
    {
      "entropy": 5.795999002456665,
      "epoch": 2.771367438241587,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004230093762851392,
      "loss": 4.9809,
      "mean_token_accuracy": 0.20746717005968093,
      "num_tokens": 61801813.0,
      "step": 35620
    },
    {
      "entropy": 5.782782173156738,
      "epoch": 2.771756467613305,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004229885004563491,
      "loss": 4.894,
      "mean_token_accuracy": 0.21372242122888566,
      "num_tokens": 61810140.0,
      "step": 35625
    },
    {
      "entropy": 5.738880968093872,
      "epoch": 2.772145496985022,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004229676223820692,
      "loss": 4.8402,
      "mean_token_accuracy": 0.21061637848615647,
      "num_tokens": 61818168.0,
      "step": 35630
    },
    {
      "entropy": 5.7481942653656,
      "epoch": 2.77253452635674,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042294674206261616,
      "loss": 4.9455,
      "mean_token_accuracy": 0.20810801684856414,
      "num_tokens": 61826640.0,
      "step": 35635
    },
    {
      "entropy": 5.744426298141479,
      "epoch": 2.7729235557284575,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004229258594983067,
      "loss": 5.0307,
      "mean_token_accuracy": 0.20852140486240386,
      "num_tokens": 61834787.0,
      "step": 35640
    },
    {
      "entropy": 5.83230242729187,
      "epoch": 2.773312585100175,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004229049746894578,
      "loss": 5.0235,
      "mean_token_accuracy": 0.20915850549936293,
      "num_tokens": 61842847.0,
      "step": 35645
    },
    {
      "entropy": 5.9205484867095945,
      "epoch": 2.773701614471893,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004228840876363864,
      "loss": 5.1547,
      "mean_token_accuracy": 0.20510663092136383,
      "num_tokens": 61851786.0,
      "step": 35650
    },
    {
      "entropy": 5.782312488555908,
      "epoch": 2.77409064384361,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004228631983394093,
      "loss": 4.9669,
      "mean_token_accuracy": 0.2137373849749565,
      "num_tokens": 61859917.0,
      "step": 35655
    },
    {
      "entropy": 5.780764818191528,
      "epoch": 2.774479673215328,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004228423067988434,
      "loss": 4.9897,
      "mean_token_accuracy": 0.20813019424676896,
      "num_tokens": 61868489.0,
      "step": 35660
    },
    {
      "entropy": 5.788909482955932,
      "epoch": 2.774868702587045,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042282141301500597,
      "loss": 5.0649,
      "mean_token_accuracy": 0.20003543943166732,
      "num_tokens": 61876896.0,
      "step": 35665
    },
    {
      "entropy": 5.833459901809692,
      "epoch": 2.775257731958763,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004228005169882138,
      "loss": 5.021,
      "mean_token_accuracy": 0.20489841550588608,
      "num_tokens": 61886110.0,
      "step": 35670
    },
    {
      "entropy": 5.792447233200074,
      "epoch": 2.7756467613304805,
      "grad_norm": 1.359375,
      "learning_rate": 0.00042277961871878413,
      "loss": 5.0186,
      "mean_token_accuracy": 0.2032855823636055,
      "num_tokens": 61894121.0,
      "step": 35675
    },
    {
      "entropy": 5.799462032318115,
      "epoch": 2.776035790702198,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004227587182070338,
      "loss": 4.988,
      "mean_token_accuracy": 0.20917194634675979,
      "num_tokens": 61903125.0,
      "step": 35680
    },
    {
      "entropy": 5.842939472198486,
      "epoch": 2.7764248200739154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004227378154532802,
      "loss": 5.0808,
      "mean_token_accuracy": 0.19859634786844255,
      "num_tokens": 61912368.0,
      "step": 35685
    },
    {
      "entropy": 5.7847424983978275,
      "epoch": 2.776813849445633,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004227169104578404,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20663607269525527,
      "num_tokens": 61921044.0,
      "step": 35690
    },
    {
      "entropy": 5.743499374389648,
      "epoch": 2.777202878817351,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042269600322103164,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20332723408937453,
      "num_tokens": 61928706.0,
      "step": 35695
    },
    {
      "entropy": 5.770125150680542,
      "epoch": 2.777591908189068,
      "grad_norm": 1.28125,
      "learning_rate": 0.000422675093743171,
      "loss": 4.994,
      "mean_token_accuracy": 0.19831535816192628,
      "num_tokens": 61936403.0,
      "step": 35700
    },
    {
      "entropy": 5.779916906356812,
      "epoch": 2.777980937560786,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042265418202457603,
      "loss": 4.9177,
      "mean_token_accuracy": 0.2170262172818184,
      "num_tokens": 61944970.0,
      "step": 35705
    },
    {
      "entropy": 5.7699440002441404,
      "epoch": 2.7783699669325035,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004226332680655637,
      "loss": 4.9721,
      "mean_token_accuracy": 0.2072661817073822,
      "num_tokens": 61953305.0,
      "step": 35710
    },
    {
      "entropy": 5.748608207702636,
      "epoch": 2.778758996304221,
      "grad_norm": 1.25,
      "learning_rate": 0.0004226123518664516,
      "loss": 4.9689,
      "mean_token_accuracy": 0.20977416783571243,
      "num_tokens": 61961098.0,
      "step": 35715
    },
    {
      "entropy": 5.793012571334839,
      "epoch": 2.7791480256759384,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004225914334275569,
      "loss": 4.9719,
      "mean_token_accuracy": 0.20499999523162843,
      "num_tokens": 61970862.0,
      "step": 35720
    },
    {
      "entropy": 5.7535638332366945,
      "epoch": 2.779537055047656,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004225705127491972,
      "loss": 4.8889,
      "mean_token_accuracy": 0.21185240000486374,
      "num_tokens": 61978967.0,
      "step": 35725
    },
    {
      "entropy": 5.7620415687561035,
      "epoch": 2.7799260844193734,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004225495898316899,
      "loss": 4.9923,
      "mean_token_accuracy": 0.20651823431253433,
      "num_tokens": 61987251.0,
      "step": 35730
    },
    {
      "entropy": 5.790960597991943,
      "epoch": 2.780315113791091,
      "grad_norm": 1.25,
      "learning_rate": 0.00042252866467535233,
      "loss": 5.0089,
      "mean_token_accuracy": 0.20955318808555604,
      "num_tokens": 61995203.0,
      "step": 35735
    },
    {
      "entropy": 5.78781943321228,
      "epoch": 2.780704143162809,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042250773728050214,
      "loss": 5.087,
      "mean_token_accuracy": 0.19624236971139908,
      "num_tokens": 62004156.0,
      "step": 35740
    },
    {
      "entropy": 5.777319622039795,
      "epoch": 2.7810931725345265,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004224868076474568,
      "loss": 4.9086,
      "mean_token_accuracy": 0.20853970646858216,
      "num_tokens": 62012919.0,
      "step": 35745
    },
    {
      "entropy": 5.772463846206665,
      "epoch": 2.781482201906244,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042246587577653394,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20359875559806823,
      "num_tokens": 62022622.0,
      "step": 35750
    },
    {
      "entropy": 5.808602237701416,
      "epoch": 2.7818712312779614,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004224449416680511,
      "loss": 5.0633,
      "mean_token_accuracy": 0.20240371972322463,
      "num_tokens": 62030747.0,
      "step": 35755
    },
    {
      "entropy": 5.803055763244629,
      "epoch": 2.782260260649679,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042242400532232593,
      "loss": 5.0128,
      "mean_token_accuracy": 0.20710589438676835,
      "num_tokens": 62039722.0,
      "step": 35760
    },
    {
      "entropy": 5.7718939781188965,
      "epoch": 2.7826492900213964,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042240306673967614,
      "loss": 4.9712,
      "mean_token_accuracy": 0.2050610974431038,
      "num_tokens": 62047907.0,
      "step": 35765
    },
    {
      "entropy": 5.771156930923462,
      "epoch": 2.783038319393114,
      "grad_norm": 1.25,
      "learning_rate": 0.00042238212592041945,
      "loss": 5.0385,
      "mean_token_accuracy": 0.20211889296770097,
      "num_tokens": 62056917.0,
      "step": 35770
    },
    {
      "entropy": 5.752791070938111,
      "epoch": 2.7834273487648318,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042236118286487364,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2109630212187767,
      "num_tokens": 62065396.0,
      "step": 35775
    },
    {
      "entropy": 5.730744743347168,
      "epoch": 2.7838163781365495,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042234023757335633,
      "loss": 5.0136,
      "mean_token_accuracy": 0.20145327001810073,
      "num_tokens": 62074202.0,
      "step": 35780
    },
    {
      "entropy": 5.7240242004394535,
      "epoch": 2.7842054075082667,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042231929004618547,
      "loss": 4.8524,
      "mean_token_accuracy": 0.22004143446683883,
      "num_tokens": 62082444.0,
      "step": 35785
    },
    {
      "entropy": 5.698032379150391,
      "epoch": 2.7845944368799844,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004222983402836789,
      "loss": 4.8754,
      "mean_token_accuracy": 0.21690282225608826,
      "num_tokens": 62090959.0,
      "step": 35790
    },
    {
      "entropy": 5.7842741966247555,
      "epoch": 2.784983466251702,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042227738828615435,
      "loss": 5.0521,
      "mean_token_accuracy": 0.20951877385377884,
      "num_tokens": 62099007.0,
      "step": 35795
    },
    {
      "entropy": 5.808579587936402,
      "epoch": 2.7853724956234194,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042225643405392993,
      "loss": 5.0163,
      "mean_token_accuracy": 0.20638728886842728,
      "num_tokens": 62107754.0,
      "step": 35800
    },
    {
      "entropy": 5.891208267211914,
      "epoch": 2.785761524995137,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042223547758732356,
      "loss": 5.0965,
      "mean_token_accuracy": 0.20191871672868728,
      "num_tokens": 62117183.0,
      "step": 35805
    },
    {
      "entropy": 5.83622727394104,
      "epoch": 2.7861505543668548,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042221451888665307,
      "loss": 5.0024,
      "mean_token_accuracy": 0.2032902091741562,
      "num_tokens": 62125713.0,
      "step": 35810
    },
    {
      "entropy": 5.791455078125,
      "epoch": 2.7865395837385725,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042219355795223657,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20954337269067763,
      "num_tokens": 62134361.0,
      "step": 35815
    },
    {
      "entropy": 5.789490365982056,
      "epoch": 2.7869286131102897,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042217259478439216,
      "loss": 4.9961,
      "mean_token_accuracy": 0.21363739520311356,
      "num_tokens": 62142912.0,
      "step": 35820
    },
    {
      "entropy": 5.800534009933472,
      "epoch": 2.7873176424820074,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004221516293834379,
      "loss": 4.998,
      "mean_token_accuracy": 0.21311627477407455,
      "num_tokens": 62150873.0,
      "step": 35825
    },
    {
      "entropy": 5.8178246974945065,
      "epoch": 2.787706671853725,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042213066174969177,
      "loss": 5.0746,
      "mean_token_accuracy": 0.19922704100608826,
      "num_tokens": 62159734.0,
      "step": 35830
    },
    {
      "entropy": 5.804581546783448,
      "epoch": 2.7880957012254424,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042210969188347217,
      "loss": 4.964,
      "mean_token_accuracy": 0.21121985018253325,
      "num_tokens": 62168319.0,
      "step": 35835
    },
    {
      "entropy": 5.81942458152771,
      "epoch": 2.78848473059716,
      "grad_norm": 1.25,
      "learning_rate": 0.0004220887197850971,
      "loss": 5.0209,
      "mean_token_accuracy": 0.19972504079341888,
      "num_tokens": 62177176.0,
      "step": 35840
    },
    {
      "entropy": 5.821565055847168,
      "epoch": 2.7888737599688778,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042206774545488476,
      "loss": 5.0392,
      "mean_token_accuracy": 0.1988060787320137,
      "num_tokens": 62185488.0,
      "step": 35845
    },
    {
      "entropy": 5.868545913696289,
      "epoch": 2.7892627893405955,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004220467688931535,
      "loss": 5.0584,
      "mean_token_accuracy": 0.2102896824479103,
      "num_tokens": 62193707.0,
      "step": 35850
    },
    {
      "entropy": 5.8591094493865965,
      "epoch": 2.7896518187123127,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004220257901002216,
      "loss": 4.9839,
      "mean_token_accuracy": 0.20450886934995652,
      "num_tokens": 62201820.0,
      "step": 35855
    },
    {
      "entropy": 5.90029296875,
      "epoch": 2.7900408480840304,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004220048090764073,
      "loss": 5.131,
      "mean_token_accuracy": 0.20247508585453033,
      "num_tokens": 62210239.0,
      "step": 35860
    },
    {
      "entropy": 5.852000617980957,
      "epoch": 2.7904298774557477,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042198382582202905,
      "loss": 5.0505,
      "mean_token_accuracy": 0.21087409257888795,
      "num_tokens": 62218818.0,
      "step": 35865
    },
    {
      "entropy": 5.820072603225708,
      "epoch": 2.7908189068274654,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004219628403374052,
      "loss": 5.0766,
      "mean_token_accuracy": 0.20060217529535293,
      "num_tokens": 62227091.0,
      "step": 35870
    },
    {
      "entropy": 5.7604471206665036,
      "epoch": 2.791207936199183,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004219418526228541,
      "loss": 5.0031,
      "mean_token_accuracy": 0.20945625603199006,
      "num_tokens": 62235838.0,
      "step": 35875
    },
    {
      "entropy": 5.817627000808716,
      "epoch": 2.7915969655709008,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004219208626786943,
      "loss": 4.9925,
      "mean_token_accuracy": 0.20540888458490372,
      "num_tokens": 62244081.0,
      "step": 35880
    },
    {
      "entropy": 5.785270881652832,
      "epoch": 2.791985994942618,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004218998705052443,
      "loss": 5.0812,
      "mean_token_accuracy": 0.19652227461338043,
      "num_tokens": 62253187.0,
      "step": 35885
    },
    {
      "entropy": 5.7822729587554935,
      "epoch": 2.7923750243143357,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042187887610282255,
      "loss": 5.0301,
      "mean_token_accuracy": 0.21476561725139617,
      "num_tokens": 62261004.0,
      "step": 35890
    },
    {
      "entropy": 5.809177541732788,
      "epoch": 2.7927640536860534,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004218578794717476,
      "loss": 5.0258,
      "mean_token_accuracy": 0.2068438097834587,
      "num_tokens": 62270196.0,
      "step": 35895
    },
    {
      "entropy": 5.75802264213562,
      "epoch": 2.7931530830577707,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042183688061233813,
      "loss": 4.9917,
      "mean_token_accuracy": 0.20206031650304795,
      "num_tokens": 62278637.0,
      "step": 35900
    },
    {
      "entropy": 5.742706775665283,
      "epoch": 2.7935421124294884,
      "grad_norm": 1.25,
      "learning_rate": 0.00042181587952491267,
      "loss": 4.949,
      "mean_token_accuracy": 0.21075053960084916,
      "num_tokens": 62287678.0,
      "step": 35905
    },
    {
      "entropy": 5.8102398872375485,
      "epoch": 2.793931141801206,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042179487620978997,
      "loss": 5.0048,
      "mean_token_accuracy": 0.21160924285650254,
      "num_tokens": 62296123.0,
      "step": 35910
    },
    {
      "entropy": 5.792109107971191,
      "epoch": 2.7943201711729238,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004217738706672886,
      "loss": 4.9906,
      "mean_token_accuracy": 0.20502732396125795,
      "num_tokens": 62304677.0,
      "step": 35915
    },
    {
      "entropy": 5.822752380371094,
      "epoch": 2.794709200544641,
      "grad_norm": 1.25,
      "learning_rate": 0.00042175286289772746,
      "loss": 4.9965,
      "mean_token_accuracy": 0.2074603408575058,
      "num_tokens": 62313944.0,
      "step": 35920
    },
    {
      "entropy": 5.795668458938598,
      "epoch": 2.7950982299163587,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004217318529014251,
      "loss": 4.9789,
      "mean_token_accuracy": 0.20077481418848037,
      "num_tokens": 62323053.0,
      "step": 35925
    },
    {
      "entropy": 5.807820081710815,
      "epoch": 2.7954872592880764,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004217108406787005,
      "loss": 5.0563,
      "mean_token_accuracy": 0.20044478327035903,
      "num_tokens": 62332642.0,
      "step": 35930
    },
    {
      "entropy": 5.840886020660401,
      "epoch": 2.7958762886597937,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004216898262298724,
      "loss": 5.0443,
      "mean_token_accuracy": 0.20167292803525924,
      "num_tokens": 62341700.0,
      "step": 35935
    },
    {
      "entropy": 5.844509744644165,
      "epoch": 2.7962653180315113,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004216688095552596,
      "loss": 5.0129,
      "mean_token_accuracy": 0.20108199268579482,
      "num_tokens": 62350828.0,
      "step": 35940
    },
    {
      "entropy": 5.70234055519104,
      "epoch": 2.796654347403229,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004216477906551811,
      "loss": 4.909,
      "mean_token_accuracy": 0.21718523502349854,
      "num_tokens": 62359494.0,
      "step": 35945
    },
    {
      "entropy": 5.761034297943115,
      "epoch": 2.7970433767749467,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042162676952995585,
      "loss": 4.9592,
      "mean_token_accuracy": 0.21320106834173203,
      "num_tokens": 62368272.0,
      "step": 35950
    },
    {
      "entropy": 5.749753427505493,
      "epoch": 2.797432406146664,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042160574617990267,
      "loss": 4.9862,
      "mean_token_accuracy": 0.2064133271574974,
      "num_tokens": 62376937.0,
      "step": 35955
    },
    {
      "entropy": 5.834462022781372,
      "epoch": 2.7978214355183817,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042158472060534066,
      "loss": 5.057,
      "mean_token_accuracy": 0.20001390874385833,
      "num_tokens": 62385765.0,
      "step": 35960
    },
    {
      "entropy": 5.7818211078643795,
      "epoch": 2.798210464890099,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042156369280658886,
      "loss": 4.9267,
      "mean_token_accuracy": 0.2095125988125801,
      "num_tokens": 62393640.0,
      "step": 35965
    },
    {
      "entropy": 5.805858993530274,
      "epoch": 2.7985994942618166,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004215426627839662,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21102497428655625,
      "num_tokens": 62402138.0,
      "step": 35970
    },
    {
      "entropy": 5.786659240722656,
      "epoch": 2.7989885236335343,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042152163053779196,
      "loss": 5.0346,
      "mean_token_accuracy": 0.20162624418735503,
      "num_tokens": 62410257.0,
      "step": 35975
    },
    {
      "entropy": 5.808211755752564,
      "epoch": 2.799377553005252,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004215005960683852,
      "loss": 5.0453,
      "mean_token_accuracy": 0.20985135138034822,
      "num_tokens": 62418409.0,
      "step": 35980
    },
    {
      "entropy": 5.763594245910644,
      "epoch": 2.7997665823769697,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004214795593760651,
      "loss": 4.9966,
      "mean_token_accuracy": 0.20591468513011932,
      "num_tokens": 62426921.0,
      "step": 35985
    },
    {
      "entropy": 5.820926189422607,
      "epoch": 2.800155611748687,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004214585204611508,
      "loss": 5.0791,
      "mean_token_accuracy": 0.2025665357708931,
      "num_tokens": 62435293.0,
      "step": 35990
    },
    {
      "entropy": 5.8100358009338375,
      "epoch": 2.8005446411204047,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042143747932396153,
      "loss": 5.045,
      "mean_token_accuracy": 0.20157251060009002,
      "num_tokens": 62443678.0,
      "step": 35995
    },
    {
      "entropy": 5.828720188140869,
      "epoch": 2.800933670492122,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004214164359648166,
      "loss": 5.0415,
      "mean_token_accuracy": 0.2002898246049881,
      "num_tokens": 62452092.0,
      "step": 36000
    },
    {
      "epoch": 2.800933670492122,
      "eval_entropy": 5.586891979019268,
      "eval_loss": 5.073980331420898,
      "eval_mean_token_accuracy": 0.21203724074897223,
      "eval_num_tokens": 62452092.0,
      "eval_runtime": 21.6259,
      "eval_samples_per_second": 1921.676,
      "eval_steps_per_second": 240.221,
      "step": 36000
    },
    {
      "entropy": 5.889638185501099,
      "epoch": 2.8013226998638396,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042139539038403535,
      "loss": 5.0593,
      "mean_token_accuracy": 0.20704751908779145,
      "num_tokens": 62460584.0,
      "step": 36005
    },
    {
      "entropy": 5.797648859024048,
      "epoch": 2.8017117292355573,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000421374342581937,
      "loss": 5.0833,
      "mean_token_accuracy": 0.19140104949474335,
      "num_tokens": 62468615.0,
      "step": 36010
    },
    {
      "entropy": 5.825901699066162,
      "epoch": 2.802100758607275,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042135329255884105,
      "loss": 5.1373,
      "mean_token_accuracy": 0.19498066008090972,
      "num_tokens": 62477630.0,
      "step": 36015
    },
    {
      "entropy": 5.798676252365112,
      "epoch": 2.8024897879789923,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004213322403150668,
      "loss": 4.9401,
      "mean_token_accuracy": 0.21292182356119155,
      "num_tokens": 62486385.0,
      "step": 36020
    },
    {
      "entropy": 5.7717653751373295,
      "epoch": 2.80287881735071,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004213111858509337,
      "loss": 4.901,
      "mean_token_accuracy": 0.21483900845050813,
      "num_tokens": 62494502.0,
      "step": 36025
    },
    {
      "entropy": 5.705852699279785,
      "epoch": 2.8032678467224277,
      "grad_norm": 1.25,
      "learning_rate": 0.0004212901291667612,
      "loss": 4.9207,
      "mean_token_accuracy": 0.2152720108628273,
      "num_tokens": 62503365.0,
      "step": 36030
    },
    {
      "entropy": 5.75157675743103,
      "epoch": 2.803656876094145,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042126907026286887,
      "loss": 5.0116,
      "mean_token_accuracy": 0.20198571532964707,
      "num_tokens": 62511932.0,
      "step": 36035
    },
    {
      "entropy": 5.796093416213989,
      "epoch": 2.8040459054658626,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042124800913957624,
      "loss": 4.9481,
      "mean_token_accuracy": 0.2136947512626648,
      "num_tokens": 62520478.0,
      "step": 36040
    },
    {
      "entropy": 5.731170320510865,
      "epoch": 2.8044349348375803,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004212269457972028,
      "loss": 4.8603,
      "mean_token_accuracy": 0.21292612105607986,
      "num_tokens": 62529514.0,
      "step": 36045
    },
    {
      "entropy": 5.810685729980468,
      "epoch": 2.804823964209298,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004212058802360682,
      "loss": 5.0771,
      "mean_token_accuracy": 0.20833201110363006,
      "num_tokens": 62538735.0,
      "step": 36050
    },
    {
      "entropy": 5.786462640762329,
      "epoch": 2.8052129935810153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004211848124564922,
      "loss": 4.9923,
      "mean_token_accuracy": 0.2047697976231575,
      "num_tokens": 62548440.0,
      "step": 36055
    },
    {
      "entropy": 5.85390453338623,
      "epoch": 2.805602022952733,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004211637424587942,
      "loss": 5.1019,
      "mean_token_accuracy": 0.19686453640460969,
      "num_tokens": 62556611.0,
      "step": 36060
    },
    {
      "entropy": 5.751316070556641,
      "epoch": 2.8059910523244502,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004211426702432941,
      "loss": 4.8757,
      "mean_token_accuracy": 0.22591349184513093,
      "num_tokens": 62565672.0,
      "step": 36065
    },
    {
      "entropy": 5.723591327667236,
      "epoch": 2.806380081696168,
      "grad_norm": 1.25,
      "learning_rate": 0.0004211215958103116,
      "loss": 4.9007,
      "mean_token_accuracy": 0.21199931800365449,
      "num_tokens": 62574037.0,
      "step": 36070
    },
    {
      "entropy": 5.725055932998657,
      "epoch": 2.8067691110678856,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042110051916016645,
      "loss": 5.0267,
      "mean_token_accuracy": 0.2047896757721901,
      "num_tokens": 62582210.0,
      "step": 36075
    },
    {
      "entropy": 5.746925401687622,
      "epoch": 2.8071581404396033,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004210794402931785,
      "loss": 4.9542,
      "mean_token_accuracy": 0.2102360889315605,
      "num_tokens": 62591442.0,
      "step": 36080
    },
    {
      "entropy": 5.822292137145996,
      "epoch": 2.807547169811321,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004210583592096675,
      "loss": 5.0672,
      "mean_token_accuracy": 0.20025772601366043,
      "num_tokens": 62599520.0,
      "step": 36085
    },
    {
      "entropy": 5.812478828430176,
      "epoch": 2.8079361991830383,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004210372759099534,
      "loss": 5.016,
      "mean_token_accuracy": 0.20694944113492966,
      "num_tokens": 62606663.0,
      "step": 36090
    },
    {
      "entropy": 5.777843856811524,
      "epoch": 2.808325228554756,
      "grad_norm": 1.171875,
      "learning_rate": 0.000421016190394356,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20489445924758912,
      "num_tokens": 62615056.0,
      "step": 36095
    },
    {
      "entropy": 5.802595424652099,
      "epoch": 2.8087142579264732,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042099510266319545,
      "loss": 5.0636,
      "mean_token_accuracy": 0.20504354685544968,
      "num_tokens": 62623627.0,
      "step": 36100
    },
    {
      "entropy": 5.818694353103638,
      "epoch": 2.809103287298191,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004209740127167915,
      "loss": 5.0806,
      "mean_token_accuracy": 0.20687176585197448,
      "num_tokens": 62632980.0,
      "step": 36105
    },
    {
      "entropy": 5.792288446426392,
      "epoch": 2.8094923166699086,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004209529205554643,
      "loss": 4.8998,
      "mean_token_accuracy": 0.21436135470867157,
      "num_tokens": 62641259.0,
      "step": 36110
    },
    {
      "entropy": 5.801562738418579,
      "epoch": 2.8098813460416263,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004209318261795339,
      "loss": 5.0748,
      "mean_token_accuracy": 0.20395449548959732,
      "num_tokens": 62649540.0,
      "step": 36115
    },
    {
      "entropy": 5.7858411312103275,
      "epoch": 2.8102703754133436,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004209107295893202,
      "loss": 4.9966,
      "mean_token_accuracy": 0.21107018142938613,
      "num_tokens": 62657496.0,
      "step": 36120
    },
    {
      "entropy": 5.741422891616821,
      "epoch": 2.8106594047850613,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042088963078514344,
      "loss": 4.9012,
      "mean_token_accuracy": 0.21873079240322113,
      "num_tokens": 62665800.0,
      "step": 36125
    },
    {
      "entropy": 5.846872329711914,
      "epoch": 2.811048434156779,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004208685297673238,
      "loss": 5.1001,
      "mean_token_accuracy": 0.20105350762605667,
      "num_tokens": 62675061.0,
      "step": 36130
    },
    {
      "entropy": 5.717379808425903,
      "epoch": 2.811437463528496,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004208474265361813,
      "loss": 4.8435,
      "mean_token_accuracy": 0.2168696254491806,
      "num_tokens": 62683889.0,
      "step": 36135
    },
    {
      "entropy": 5.697771644592285,
      "epoch": 2.811826492900214,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042082632109203627,
      "loss": 4.9415,
      "mean_token_accuracy": 0.20878593772649764,
      "num_tokens": 62692711.0,
      "step": 36140
    },
    {
      "entropy": 5.787750768661499,
      "epoch": 2.8122155222719316,
      "grad_norm": 1.328125,
      "learning_rate": 0.000420805213435209,
      "loss": 4.9751,
      "mean_token_accuracy": 0.21258010119199752,
      "num_tokens": 62701048.0,
      "step": 36145
    },
    {
      "entropy": 5.833490324020386,
      "epoch": 2.8126045516436493,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004207841035660196,
      "loss": 5.0833,
      "mean_token_accuracy": 0.19493373930454255,
      "num_tokens": 62710440.0,
      "step": 36150
    },
    {
      "entropy": 5.842769718170166,
      "epoch": 2.8129935810153666,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004207629914847885,
      "loss": 5.0669,
      "mean_token_accuracy": 0.19919547736644744,
      "num_tokens": 62719853.0,
      "step": 36155
    },
    {
      "entropy": 5.771313810348511,
      "epoch": 2.8133826103870843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000420741877191836,
      "loss": 5.0375,
      "mean_token_accuracy": 0.198835651576519,
      "num_tokens": 62729784.0,
      "step": 36160
    },
    {
      "entropy": 5.780418634414673,
      "epoch": 2.8137716397588015,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004207207606874825,
      "loss": 4.9589,
      "mean_token_accuracy": 0.21452402770519258,
      "num_tokens": 62738414.0,
      "step": 36165
    },
    {
      "entropy": 5.7513936996459964,
      "epoch": 2.814160669130519,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004206996419720484,
      "loss": 4.9374,
      "mean_token_accuracy": 0.21025937497615815,
      "num_tokens": 62746878.0,
      "step": 36170
    },
    {
      "entropy": 5.815637302398682,
      "epoch": 2.814549698502237,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004206785210458541,
      "loss": 5.0621,
      "mean_token_accuracy": 0.20343313962221146,
      "num_tokens": 62756060.0,
      "step": 36175
    },
    {
      "entropy": 5.787099838256836,
      "epoch": 2.8149387278739546,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004206573979092202,
      "loss": 4.9494,
      "mean_token_accuracy": 0.21304106563329697,
      "num_tokens": 62764864.0,
      "step": 36180
    },
    {
      "entropy": 5.757422924041748,
      "epoch": 2.8153277572456723,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042063627256246706,
      "loss": 4.9426,
      "mean_token_accuracy": 0.20788237899541856,
      "num_tokens": 62772920.0,
      "step": 36185
    },
    {
      "entropy": 5.6963413715362545,
      "epoch": 2.8157167866173896,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004206151450059153,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2125793918967247,
      "num_tokens": 62781099.0,
      "step": 36190
    },
    {
      "entropy": 5.862668704986572,
      "epoch": 2.8161058159891073,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042059401523988546,
      "loss": 5.0938,
      "mean_token_accuracy": 0.1970043733716011,
      "num_tokens": 62789656.0,
      "step": 36195
    },
    {
      "entropy": 5.819086408615112,
      "epoch": 2.8164948453608245,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004205728832646982,
      "loss": 5.0593,
      "mean_token_accuracy": 0.20396489053964614,
      "num_tokens": 62798402.0,
      "step": 36200
    },
    {
      "entropy": 5.771092748641967,
      "epoch": 2.816883874732542,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042055174908067414,
      "loss": 5.0481,
      "mean_token_accuracy": 0.20190461426973344,
      "num_tokens": 62807260.0,
      "step": 36205
    },
    {
      "entropy": 5.740847206115722,
      "epoch": 2.81727290410426,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042053061268813397,
      "loss": 4.8891,
      "mean_token_accuracy": 0.2122210755944252,
      "num_tokens": 62816240.0,
      "step": 36210
    },
    {
      "entropy": 5.830113744735717,
      "epoch": 2.8176619334759776,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042050947408739836,
      "loss": 4.9638,
      "mean_token_accuracy": 0.2082709863781929,
      "num_tokens": 62824630.0,
      "step": 36215
    },
    {
      "entropy": 5.770707702636718,
      "epoch": 2.818050962847695,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004204883332787881,
      "loss": 5.0864,
      "mean_token_accuracy": 0.2077701508998871,
      "num_tokens": 62833193.0,
      "step": 36220
    },
    {
      "entropy": 5.728986835479736,
      "epoch": 2.8184399922194125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042046719026262396,
      "loss": 4.9282,
      "mean_token_accuracy": 0.2152310013771057,
      "num_tokens": 62841424.0,
      "step": 36225
    },
    {
      "entropy": 5.7881958961486815,
      "epoch": 2.8188290215911302,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042044604503922674,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20436773151159288,
      "num_tokens": 62849388.0,
      "step": 36230
    },
    {
      "entropy": 5.762889862060547,
      "epoch": 2.8192180509628475,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004204248976089173,
      "loss": 4.9157,
      "mean_token_accuracy": 0.2110852673649788,
      "num_tokens": 62857859.0,
      "step": 36235
    },
    {
      "entropy": 5.759855127334594,
      "epoch": 2.819607080334565,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004204037479720165,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20698399394750594,
      "num_tokens": 62866558.0,
      "step": 36240
    },
    {
      "entropy": 5.679414176940918,
      "epoch": 2.819996109706283,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042038259612884514,
      "loss": 4.8816,
      "mean_token_accuracy": 0.21217338740825653,
      "num_tokens": 62875663.0,
      "step": 36245
    },
    {
      "entropy": 5.7437938213348385,
      "epoch": 2.8203851390780006,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042036144207972445,
      "loss": 4.945,
      "mean_token_accuracy": 0.21236516237258912,
      "num_tokens": 62884581.0,
      "step": 36250
    },
    {
      "entropy": 5.754427146911621,
      "epoch": 2.820774168449718,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004203402858249752,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20685985833406448,
      "num_tokens": 62892710.0,
      "step": 36255
    },
    {
      "entropy": 5.759410762786866,
      "epoch": 2.8211631978214355,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004203191273649184,
      "loss": 4.9398,
      "mean_token_accuracy": 0.2139760971069336,
      "num_tokens": 62901933.0,
      "step": 36260
    },
    {
      "entropy": 5.7797528266906735,
      "epoch": 2.8215522271931532,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042029796669987513,
      "loss": 4.9589,
      "mean_token_accuracy": 0.2020659029483795,
      "num_tokens": 62910155.0,
      "step": 36265
    },
    {
      "entropy": 5.781597709655761,
      "epoch": 2.8219412565648705,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004202768038301665,
      "loss": 5.0184,
      "mean_token_accuracy": 0.2031245857477188,
      "num_tokens": 62919645.0,
      "step": 36270
    },
    {
      "entropy": 5.88561749458313,
      "epoch": 2.822330285936588,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042025563875611367,
      "loss": 5.0828,
      "mean_token_accuracy": 0.1937992811203003,
      "num_tokens": 62928824.0,
      "step": 36275
    },
    {
      "entropy": 5.778196001052857,
      "epoch": 2.822719315308306,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004202344714780376,
      "loss": 4.935,
      "mean_token_accuracy": 0.2106197729706764,
      "num_tokens": 62936497.0,
      "step": 36280
    },
    {
      "entropy": 5.751037120819092,
      "epoch": 2.8231083446800236,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042021330199625966,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20692837834358216,
      "num_tokens": 62944245.0,
      "step": 36285
    },
    {
      "entropy": 5.749965715408325,
      "epoch": 2.823497374051741,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042019213031110105,
      "loss": 5.0132,
      "mean_token_accuracy": 0.20873634219169618,
      "num_tokens": 62952980.0,
      "step": 36290
    },
    {
      "entropy": 5.828504610061645,
      "epoch": 2.8238864034234585,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004201709564228829,
      "loss": 5.0513,
      "mean_token_accuracy": 0.20598533749580383,
      "num_tokens": 62962082.0,
      "step": 36295
    },
    {
      "entropy": 5.8700204372406,
      "epoch": 2.824275432795176,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042014978033192655,
      "loss": 5.1104,
      "mean_token_accuracy": 0.20220391303300858,
      "num_tokens": 62971033.0,
      "step": 36300
    },
    {
      "entropy": 5.864890623092651,
      "epoch": 2.8246644621668935,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042012860203855337,
      "loss": 5.0083,
      "mean_token_accuracy": 0.2045157626271248,
      "num_tokens": 62979646.0,
      "step": 36305
    },
    {
      "entropy": 5.755398559570312,
      "epoch": 2.825053491538611,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042010742154308453,
      "loss": 5.0079,
      "mean_token_accuracy": 0.2024176761507988,
      "num_tokens": 62988690.0,
      "step": 36310
    },
    {
      "entropy": 5.8255726337432865,
      "epoch": 2.825442520910329,
      "grad_norm": 1.25,
      "learning_rate": 0.00042008623884584167,
      "loss": 5.0412,
      "mean_token_accuracy": 0.20381843596696853,
      "num_tokens": 62997186.0,
      "step": 36315
    },
    {
      "entropy": 5.777501344680786,
      "epoch": 2.825831550282046,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042006505394714597,
      "loss": 4.9167,
      "mean_token_accuracy": 0.2133474349975586,
      "num_tokens": 63006449.0,
      "step": 36320
    },
    {
      "entropy": 5.807368659973145,
      "epoch": 2.826220579653764,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042004386684731896,
      "loss": 4.9571,
      "mean_token_accuracy": 0.2064094439148903,
      "num_tokens": 63015862.0,
      "step": 36325
    },
    {
      "entropy": 5.787207078933716,
      "epoch": 2.8266096090254815,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042002267754668216,
      "loss": 5.0309,
      "mean_token_accuracy": 0.20088393837213517,
      "num_tokens": 63024919.0,
      "step": 36330
    },
    {
      "entropy": 5.792542028427124,
      "epoch": 2.826998638397199,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000420001486045557,
      "loss": 4.9734,
      "mean_token_accuracy": 0.211808542907238,
      "num_tokens": 63033513.0,
      "step": 36335
    },
    {
      "entropy": 5.804136610031128,
      "epoch": 2.8273876677689165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004199802923442652,
      "loss": 5.1718,
      "mean_token_accuracy": 0.20556291341781616,
      "num_tokens": 63043094.0,
      "step": 36340
    },
    {
      "entropy": 5.731440019607544,
      "epoch": 2.827776697140634,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004199590964431282,
      "loss": 4.8686,
      "mean_token_accuracy": 0.22186427265405656,
      "num_tokens": 63050714.0,
      "step": 36345
    },
    {
      "entropy": 5.789445257186889,
      "epoch": 2.828165726512352,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004199378983424676,
      "loss": 4.9684,
      "mean_token_accuracy": 0.20612358748912812,
      "num_tokens": 63059381.0,
      "step": 36350
    },
    {
      "entropy": 5.788238763809204,
      "epoch": 2.828554755884069,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041991669804260505,
      "loss": 5.0193,
      "mean_token_accuracy": 0.20729867219924927,
      "num_tokens": 63067597.0,
      "step": 36355
    },
    {
      "entropy": 5.731617641448975,
      "epoch": 2.828943785255787,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004198954955438623,
      "loss": 4.8687,
      "mean_token_accuracy": 0.21195537745952606,
      "num_tokens": 63075680.0,
      "step": 36360
    },
    {
      "entropy": 5.74955267906189,
      "epoch": 2.8293328146275045,
      "grad_norm": 1.34375,
      "learning_rate": 0.000419874290846561,
      "loss": 5.0554,
      "mean_token_accuracy": 0.2060772567987442,
      "num_tokens": 63084067.0,
      "step": 36365
    },
    {
      "entropy": 5.826923036575318,
      "epoch": 2.8297218439992218,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041985308395102303,
      "loss": 5.0075,
      "mean_token_accuracy": 0.2042214021086693,
      "num_tokens": 63093694.0,
      "step": 36370
    },
    {
      "entropy": 5.784402370452881,
      "epoch": 2.8301108733709395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041983187485756997,
      "loss": 4.9044,
      "mean_token_accuracy": 0.21533576399087906,
      "num_tokens": 63102496.0,
      "step": 36375
    },
    {
      "entropy": 5.8296544551849365,
      "epoch": 2.830499902742657,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041981066356652373,
      "loss": 5.0414,
      "mean_token_accuracy": 0.20290652215480803,
      "num_tokens": 63111246.0,
      "step": 36380
    },
    {
      "entropy": 5.82724175453186,
      "epoch": 2.830888932114375,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041978945007820626,
      "loss": 5.0098,
      "mean_token_accuracy": 0.20927574932575227,
      "num_tokens": 63119953.0,
      "step": 36385
    },
    {
      "entropy": 5.860178470611572,
      "epoch": 2.831277961486092,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041976823439293925,
      "loss": 5.1035,
      "mean_token_accuracy": 0.20531425923109053,
      "num_tokens": 63128904.0,
      "step": 36390
    },
    {
      "entropy": 5.826114749908447,
      "epoch": 2.83166699085781,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004197470165110447,
      "loss": 5.0235,
      "mean_token_accuracy": 0.2056563064455986,
      "num_tokens": 63137049.0,
      "step": 36395
    },
    {
      "entropy": 5.728459644317627,
      "epoch": 2.832056020229527,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004197257964328446,
      "loss": 4.9902,
      "mean_token_accuracy": 0.20579117983579637,
      "num_tokens": 63145772.0,
      "step": 36400
    },
    {
      "entropy": 5.752023077011108,
      "epoch": 2.8324450496012448,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004197045741586609,
      "loss": 5.012,
      "mean_token_accuracy": 0.20646674185991287,
      "num_tokens": 63154913.0,
      "step": 36405
    },
    {
      "entropy": 5.754434394836426,
      "epoch": 2.8328340789729625,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004196833496888156,
      "loss": 4.9958,
      "mean_token_accuracy": 0.20476053953170775,
      "num_tokens": 63163671.0,
      "step": 36410
    },
    {
      "entropy": 5.781648731231689,
      "epoch": 2.83322310834468,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004196621230236308,
      "loss": 5.0614,
      "mean_token_accuracy": 0.20343200862407684,
      "num_tokens": 63172276.0,
      "step": 36415
    },
    {
      "entropy": 5.893756675720215,
      "epoch": 2.833612137716398,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041964089416342845,
      "loss": 5.0855,
      "mean_token_accuracy": 0.20354282706975937,
      "num_tokens": 63180987.0,
      "step": 36420
    },
    {
      "entropy": 5.86677885055542,
      "epoch": 2.834001167088115,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004196196631085308,
      "loss": 5.0155,
      "mean_token_accuracy": 0.20671121329069136,
      "num_tokens": 63189668.0,
      "step": 36425
    },
    {
      "entropy": 5.776651382446289,
      "epoch": 2.834390196459833,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041959842985925987,
      "loss": 4.9408,
      "mean_token_accuracy": 0.20556287616491317,
      "num_tokens": 63198757.0,
      "step": 36430
    },
    {
      "entropy": 5.811961126327515,
      "epoch": 2.83477922583155,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000419577194415938,
      "loss": 5.0555,
      "mean_token_accuracy": 0.20030248165130615,
      "num_tokens": 63207695.0,
      "step": 36435
    },
    {
      "entropy": 5.869120073318482,
      "epoch": 2.8351682552032678,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041955595677888735,
      "loss": 5.0855,
      "mean_token_accuracy": 0.2056986004114151,
      "num_tokens": 63216789.0,
      "step": 36440
    },
    {
      "entropy": 5.73993878364563,
      "epoch": 2.8355572845749855,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004195347169484301,
      "loss": 4.9708,
      "mean_token_accuracy": 0.20637117475271224,
      "num_tokens": 63225359.0,
      "step": 36445
    },
    {
      "entropy": 5.846277189254761,
      "epoch": 2.835946313946703,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004195134749248886,
      "loss": 5.121,
      "mean_token_accuracy": 0.1918686717748642,
      "num_tokens": 63234405.0,
      "step": 36450
    },
    {
      "entropy": 5.778949975967407,
      "epoch": 2.8363353433184204,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004194922307085851,
      "loss": 4.9653,
      "mean_token_accuracy": 0.21328257769346237,
      "num_tokens": 63243558.0,
      "step": 36455
    },
    {
      "entropy": 5.800359439849854,
      "epoch": 2.836724372690138,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004194709842998419,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20830978751182555,
      "num_tokens": 63251997.0,
      "step": 36460
    },
    {
      "entropy": 5.772463655471801,
      "epoch": 2.837113402061856,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041944973569898157,
      "loss": 4.9428,
      "mean_token_accuracy": 0.21174442619085312,
      "num_tokens": 63260439.0,
      "step": 36465
    },
    {
      "entropy": 5.801935243606567,
      "epoch": 2.837502431433573,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004194284849063265,
      "loss": 4.9589,
      "mean_token_accuracy": 0.21026259362697602,
      "num_tokens": 63268613.0,
      "step": 36470
    },
    {
      "entropy": 5.798560094833374,
      "epoch": 2.8378914608052908,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004194072319221989,
      "loss": 5.0349,
      "mean_token_accuracy": 0.20393248200416564,
      "num_tokens": 63277255.0,
      "step": 36475
    },
    {
      "entropy": 5.763137102127075,
      "epoch": 2.8382804901770085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004193859767469214,
      "loss": 5.0103,
      "mean_token_accuracy": 0.21093186736106873,
      "num_tokens": 63285596.0,
      "step": 36480
    },
    {
      "entropy": 5.778661775588989,
      "epoch": 2.838669519548726,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041936471938081667,
      "loss": 4.939,
      "mean_token_accuracy": 0.20971887707710266,
      "num_tokens": 63293752.0,
      "step": 36485
    },
    {
      "entropy": 5.783925771713257,
      "epoch": 2.8390585489204434,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041934345982420695,
      "loss": 4.9975,
      "mean_token_accuracy": 0.20397304445505143,
      "num_tokens": 63302766.0,
      "step": 36490
    },
    {
      "entropy": 5.816025733947754,
      "epoch": 2.839447578292161,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000419322198077415,
      "loss": 5.025,
      "mean_token_accuracy": 0.2054214060306549,
      "num_tokens": 63311166.0,
      "step": 36495
    },
    {
      "entropy": 5.809567403793335,
      "epoch": 2.8398366076638784,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041930093414076344,
      "loss": 4.9865,
      "mean_token_accuracy": 0.20785675346851348,
      "num_tokens": 63319355.0,
      "step": 36500
    },
    {
      "entropy": 5.782669305801392,
      "epoch": 2.840225637035596,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00041927966801457486,
      "loss": 5.0676,
      "mean_token_accuracy": 0.19753622710704805,
      "num_tokens": 63328548.0,
      "step": 36505
    },
    {
      "entropy": 5.817567920684814,
      "epoch": 2.8406146664073137,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000419258399699172,
      "loss": 5.0799,
      "mean_token_accuracy": 0.1966158613562584,
      "num_tokens": 63337743.0,
      "step": 36510
    },
    {
      "entropy": 5.852380895614624,
      "epoch": 2.8410036957790314,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004192371291948775,
      "loss": 4.9632,
      "mean_token_accuracy": 0.2090526506304741,
      "num_tokens": 63346064.0,
      "step": 36515
    },
    {
      "entropy": 5.794937992095948,
      "epoch": 2.841392725150749,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041921585650201413,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20618218630552293,
      "num_tokens": 63354957.0,
      "step": 36520
    },
    {
      "entropy": 5.762783336639404,
      "epoch": 2.8417817545224664,
      "grad_norm": 1.25,
      "learning_rate": 0.0004191945816209047,
      "loss": 4.9826,
      "mean_token_accuracy": 0.20589951276779175,
      "num_tokens": 63363918.0,
      "step": 36525
    },
    {
      "entropy": 5.808455562591552,
      "epoch": 2.842170783894184,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041917330455187195,
      "loss": 5.0306,
      "mean_token_accuracy": 0.20440523326396942,
      "num_tokens": 63372540.0,
      "step": 36530
    },
    {
      "entropy": 5.895424175262451,
      "epoch": 2.8425598132659013,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041915202529523894,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19563667923212053,
      "num_tokens": 63381215.0,
      "step": 36535
    },
    {
      "entropy": 5.789294910430908,
      "epoch": 2.842948842637619,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004191307438513283,
      "loss": 4.9782,
      "mean_token_accuracy": 0.21502976566553117,
      "num_tokens": 63390738.0,
      "step": 36540
    },
    {
      "entropy": 5.791314888000488,
      "epoch": 2.8433378720093367,
      "grad_norm": 1.328125,
      "learning_rate": 0.000419109460220463,
      "loss": 5.0934,
      "mean_token_accuracy": 0.19584109783172607,
      "num_tokens": 63398836.0,
      "step": 36545
    },
    {
      "entropy": 5.77953143119812,
      "epoch": 2.8437269013810544,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041908817440296605,
      "loss": 4.947,
      "mean_token_accuracy": 0.1985804632306099,
      "num_tokens": 63407211.0,
      "step": 36550
    },
    {
      "entropy": 5.828835773468017,
      "epoch": 2.8441159307527717,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041906688639916035,
      "loss": 5.0414,
      "mean_token_accuracy": 0.20090197026729584,
      "num_tokens": 63415415.0,
      "step": 36555
    },
    {
      "entropy": 5.662143802642822,
      "epoch": 2.8445049601244894,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004190455962093689,
      "loss": 4.841,
      "mean_token_accuracy": 0.21736225336790085,
      "num_tokens": 63422907.0,
      "step": 36560
    },
    {
      "entropy": 5.709348869323731,
      "epoch": 2.844893989496207,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041902430383391494,
      "loss": 4.914,
      "mean_token_accuracy": 0.20893315821886063,
      "num_tokens": 63431348.0,
      "step": 36565
    },
    {
      "entropy": 5.830709075927734,
      "epoch": 2.8452830188679243,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004190030092731214,
      "loss": 4.9892,
      "mean_token_accuracy": 0.21213430017232895,
      "num_tokens": 63440189.0,
      "step": 36570
    },
    {
      "entropy": 5.824370336532593,
      "epoch": 2.845672048239642,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004189817125273113,
      "loss": 5.0479,
      "mean_token_accuracy": 0.20580596327781678,
      "num_tokens": 63449082.0,
      "step": 36575
    },
    {
      "entropy": 5.814467573165894,
      "epoch": 2.8460610776113597,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041896041359680797,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20590615570545195,
      "num_tokens": 63458204.0,
      "step": 36580
    },
    {
      "entropy": 5.784512758255005,
      "epoch": 2.8464501069830774,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041893911248193437,
      "loss": 4.9328,
      "mean_token_accuracy": 0.20886454731225967,
      "num_tokens": 63466623.0,
      "step": 36585
    },
    {
      "entropy": 5.790114164352417,
      "epoch": 2.8468391363547947,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000418917809183014,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20256955325603485,
      "num_tokens": 63475903.0,
      "step": 36590
    },
    {
      "entropy": 5.869958114624024,
      "epoch": 2.8472281657265124,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041889650370036986,
      "loss": 5.0467,
      "mean_token_accuracy": 0.19589791744947432,
      "num_tokens": 63484413.0,
      "step": 36595
    },
    {
      "entropy": 5.792706537246704,
      "epoch": 2.84761719509823,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004188751960343253,
      "loss": 5.0218,
      "mean_token_accuracy": 0.2041142240166664,
      "num_tokens": 63494059.0,
      "step": 36600
    },
    {
      "entropy": 5.795566892623901,
      "epoch": 2.8480062244699473,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004188538861852037,
      "loss": 4.9068,
      "mean_token_accuracy": 0.20851334035396576,
      "num_tokens": 63502673.0,
      "step": 36605
    },
    {
      "entropy": 5.809505605697632,
      "epoch": 2.848395253841665,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004188325741533283,
      "loss": 5.0152,
      "mean_token_accuracy": 0.20344417095184325,
      "num_tokens": 63511006.0,
      "step": 36610
    },
    {
      "entropy": 5.765609455108643,
      "epoch": 2.8487842832133827,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004188112599390225,
      "loss": 5.015,
      "mean_token_accuracy": 0.21326516568660736,
      "num_tokens": 63519381.0,
      "step": 36615
    },
    {
      "entropy": 5.735032558441162,
      "epoch": 2.8491733125851004,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004187899435426098,
      "loss": 5.0059,
      "mean_token_accuracy": 0.2089775547385216,
      "num_tokens": 63528111.0,
      "step": 36620
    },
    {
      "entropy": 5.798235607147217,
      "epoch": 2.8495623419568177,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041876862496441347,
      "loss": 4.9652,
      "mean_token_accuracy": 0.2107584595680237,
      "num_tokens": 63537028.0,
      "step": 36625
    },
    {
      "entropy": 5.772319078445435,
      "epoch": 2.8499513713285354,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041874730420475716,
      "loss": 4.9544,
      "mean_token_accuracy": 0.2070559337735176,
      "num_tokens": 63545666.0,
      "step": 36630
    },
    {
      "entropy": 5.737691831588745,
      "epoch": 2.8503404007002526,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041872598126396434,
      "loss": 4.9884,
      "mean_token_accuracy": 0.20281550288200378,
      "num_tokens": 63554122.0,
      "step": 36635
    },
    {
      "entropy": 5.834388780593872,
      "epoch": 2.8507294300719703,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041870465614235843,
      "loss": 5.0095,
      "mean_token_accuracy": 0.20606082081794738,
      "num_tokens": 63563108.0,
      "step": 36640
    },
    {
      "entropy": 5.81687560081482,
      "epoch": 2.851118459443688,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041868332884026317,
      "loss": 5.0133,
      "mean_token_accuracy": 0.20348789691925048,
      "num_tokens": 63572854.0,
      "step": 36645
    },
    {
      "entropy": 5.843459749221802,
      "epoch": 2.8515074888154057,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004186619993580021,
      "loss": 5.0144,
      "mean_token_accuracy": 0.20828895568847655,
      "num_tokens": 63580900.0,
      "step": 36650
    },
    {
      "entropy": 5.7443499088287355,
      "epoch": 2.851896518187123,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041864066769589875,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21285949796438217,
      "num_tokens": 63590093.0,
      "step": 36655
    },
    {
      "entropy": 5.759358310699463,
      "epoch": 2.8522855475588407,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041861933385427697,
      "loss": 5.0722,
      "mean_token_accuracy": 0.20687043368816377,
      "num_tokens": 63599519.0,
      "step": 36660
    },
    {
      "entropy": 5.782386207580567,
      "epoch": 2.8526745769305584,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004185979978334604,
      "loss": 5.0698,
      "mean_token_accuracy": 0.20296140462160112,
      "num_tokens": 63608088.0,
      "step": 36665
    },
    {
      "entropy": 5.8715204238891605,
      "epoch": 2.8530636063022756,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004185766596337727,
      "loss": 5.023,
      "mean_token_accuracy": 0.20663402676582338,
      "num_tokens": 63617247.0,
      "step": 36670
    },
    {
      "entropy": 5.769886779785156,
      "epoch": 2.8534526356739933,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041855531925553773,
      "loss": 5.0284,
      "mean_token_accuracy": 0.2032049313187599,
      "num_tokens": 63626218.0,
      "step": 36675
    },
    {
      "entropy": 5.800747632980347,
      "epoch": 2.853841665045711,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004185339766990794,
      "loss": 4.9765,
      "mean_token_accuracy": 0.2098732516169548,
      "num_tokens": 63634667.0,
      "step": 36680
    },
    {
      "entropy": 5.7765305519104,
      "epoch": 2.8542306944174287,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004185126319647213,
      "loss": 4.9457,
      "mean_token_accuracy": 0.2114175021648407,
      "num_tokens": 63643506.0,
      "step": 36685
    },
    {
      "entropy": 5.724131011962891,
      "epoch": 2.854619723789146,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004184912850527875,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21366899758577346,
      "num_tokens": 63651883.0,
      "step": 36690
    },
    {
      "entropy": 5.817035675048828,
      "epoch": 2.8550087531608637,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004184699359636018,
      "loss": 5.0435,
      "mean_token_accuracy": 0.19753620028495789,
      "num_tokens": 63660829.0,
      "step": 36695
    },
    {
      "entropy": 5.853626108169555,
      "epoch": 2.8553977825325814,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004184485846974882,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20952989906072617,
      "num_tokens": 63669730.0,
      "step": 36700
    },
    {
      "entropy": 5.810510921478271,
      "epoch": 2.8557868119042986,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004184272312547706,
      "loss": 4.9918,
      "mean_token_accuracy": 0.2018452525138855,
      "num_tokens": 63678921.0,
      "step": 36705
    },
    {
      "entropy": 5.810440969467163,
      "epoch": 2.8561758412760163,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041840587563577315,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19599928259849547,
      "num_tokens": 63687636.0,
      "step": 36710
    },
    {
      "entropy": 5.821452999114991,
      "epoch": 2.856564870647734,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004183845178408197,
      "loss": 5.038,
      "mean_token_accuracy": 0.20465950518846512,
      "num_tokens": 63695701.0,
      "step": 36715
    },
    {
      "entropy": 5.801389217376709,
      "epoch": 2.8569539000194517,
      "grad_norm": 1.25,
      "learning_rate": 0.00041836315787023456,
      "loss": 5.0253,
      "mean_token_accuracy": 0.1996399164199829,
      "num_tokens": 63704904.0,
      "step": 36720
    },
    {
      "entropy": 5.747359657287598,
      "epoch": 2.857342929391169,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041834179572434153,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20873996317386628,
      "num_tokens": 63714250.0,
      "step": 36725
    },
    {
      "entropy": 5.800381326675415,
      "epoch": 2.8577319587628867,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041832043140346495,
      "loss": 5.039,
      "mean_token_accuracy": 0.20130584985017777,
      "num_tokens": 63723472.0,
      "step": 36730
    },
    {
      "entropy": 5.745037317276001,
      "epoch": 2.858120988134604,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004182990649079289,
      "loss": 4.8238,
      "mean_token_accuracy": 0.213958440721035,
      "num_tokens": 63731510.0,
      "step": 36735
    },
    {
      "entropy": 5.856986427307129,
      "epoch": 2.8585100175063216,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004182776962380577,
      "loss": 5.1404,
      "mean_token_accuracy": 0.1988018959760666,
      "num_tokens": 63739809.0,
      "step": 36740
    },
    {
      "entropy": 5.7785923957824705,
      "epoch": 2.8588990468780393,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041825632539417546,
      "loss": 4.9526,
      "mean_token_accuracy": 0.21079703867435456,
      "num_tokens": 63748387.0,
      "step": 36745
    },
    {
      "entropy": 5.863193607330322,
      "epoch": 2.859288076249757,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004182349523766065,
      "loss": 5.0668,
      "mean_token_accuracy": 0.20468565374612807,
      "num_tokens": 63756119.0,
      "step": 36750
    },
    {
      "entropy": 5.748387432098388,
      "epoch": 2.8596771056214743,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041821357718567514,
      "loss": 5.0168,
      "mean_token_accuracy": 0.2093930199742317,
      "num_tokens": 63764797.0,
      "step": 36755
    },
    {
      "entropy": 5.739344739913941,
      "epoch": 2.860066134993192,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004181921998217057,
      "loss": 4.9111,
      "mean_token_accuracy": 0.208493734896183,
      "num_tokens": 63773114.0,
      "step": 36760
    },
    {
      "entropy": 5.862214088439941,
      "epoch": 2.8604551643649097,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004181708202850225,
      "loss": 5.0881,
      "mean_token_accuracy": 0.19925374537706375,
      "num_tokens": 63781345.0,
      "step": 36765
    },
    {
      "entropy": 5.819920825958252,
      "epoch": 2.860844193736627,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041814943857595,
      "loss": 5.007,
      "mean_token_accuracy": 0.2051706627011299,
      "num_tokens": 63789666.0,
      "step": 36770
    },
    {
      "entropy": 5.7988800525665285,
      "epoch": 2.8612332231083446,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004181280546948127,
      "loss": 5.0868,
      "mean_token_accuracy": 0.1943100646138191,
      "num_tokens": 63798121.0,
      "step": 36775
    },
    {
      "entropy": 5.768915128707886,
      "epoch": 2.8616222524800623,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004181066686419349,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2162947401404381,
      "num_tokens": 63806443.0,
      "step": 36780
    },
    {
      "entropy": 5.761303806304932,
      "epoch": 2.86201128185178,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041808528041764115,
      "loss": 4.997,
      "mean_token_accuracy": 0.20732723474502562,
      "num_tokens": 63815068.0,
      "step": 36785
    },
    {
      "entropy": 5.803327131271362,
      "epoch": 2.8624003112234973,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004180638900222561,
      "loss": 5.0084,
      "mean_token_accuracy": 0.20811428129673004,
      "num_tokens": 63823518.0,
      "step": 36790
    },
    {
      "entropy": 5.841814279556274,
      "epoch": 2.862789340595215,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004180424974561042,
      "loss": 5.1211,
      "mean_token_accuracy": 0.1980796664953232,
      "num_tokens": 63832435.0,
      "step": 36795
    },
    {
      "entropy": 5.835335206985474,
      "epoch": 2.8631783699669326,
      "grad_norm": 1.25,
      "learning_rate": 0.00041802110271951,
      "loss": 5.0653,
      "mean_token_accuracy": 0.19666991084814073,
      "num_tokens": 63841384.0,
      "step": 36800
    },
    {
      "entropy": 5.858704471588135,
      "epoch": 2.86356739933865,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004179997058127983,
      "loss": 4.9506,
      "mean_token_accuracy": 0.20997270494699477,
      "num_tokens": 63850563.0,
      "step": 36805
    },
    {
      "entropy": 5.791460227966309,
      "epoch": 2.8639564287103676,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004179783067362936,
      "loss": 5.0129,
      "mean_token_accuracy": 0.20092549920082092,
      "num_tokens": 63858657.0,
      "step": 36810
    },
    {
      "entropy": 5.821629381179809,
      "epoch": 2.8643454580820853,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004179569054903207,
      "loss": 5.0771,
      "mean_token_accuracy": 0.2062112495303154,
      "num_tokens": 63866970.0,
      "step": 36815
    },
    {
      "entropy": 5.727802848815918,
      "epoch": 2.864734487453803,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041793550207520436,
      "loss": 4.9393,
      "mean_token_accuracy": 0.21322398483753205,
      "num_tokens": 63874875.0,
      "step": 36820
    },
    {
      "entropy": 5.737900876998902,
      "epoch": 2.8651235168255202,
      "grad_norm": 1.25,
      "learning_rate": 0.0004179140964912692,
      "loss": 5.0235,
      "mean_token_accuracy": 0.203223118185997,
      "num_tokens": 63883479.0,
      "step": 36825
    },
    {
      "entropy": 5.797003746032715,
      "epoch": 2.865512546197238,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004178926887388401,
      "loss": 4.9847,
      "mean_token_accuracy": 0.20345984697341918,
      "num_tokens": 63891700.0,
      "step": 36830
    },
    {
      "entropy": 5.834503030776977,
      "epoch": 2.865901575568955,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041787127881824193,
      "loss": 5.0266,
      "mean_token_accuracy": 0.20017229169607162,
      "num_tokens": 63900516.0,
      "step": 36835
    },
    {
      "entropy": 5.794922351837158,
      "epoch": 2.866290604940673,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041784986672979945,
      "loss": 4.9373,
      "mean_token_accuracy": 0.22074492275714874,
      "num_tokens": 63908290.0,
      "step": 36840
    },
    {
      "entropy": 5.8042747497558596,
      "epoch": 2.8666796343123906,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041782845247383767,
      "loss": 5.0564,
      "mean_token_accuracy": 0.20089934915304183,
      "num_tokens": 63917019.0,
      "step": 36845
    },
    {
      "entropy": 5.753231382369995,
      "epoch": 2.8670686636841083,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041780703605068145,
      "loss": 4.9243,
      "mean_token_accuracy": 0.21182291507720946,
      "num_tokens": 63925155.0,
      "step": 36850
    },
    {
      "entropy": 5.835081291198731,
      "epoch": 2.867457693055826,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004177856174606558,
      "loss": 5.0312,
      "mean_token_accuracy": 0.1982508972287178,
      "num_tokens": 63932931.0,
      "step": 36855
    },
    {
      "entropy": 5.789063215255737,
      "epoch": 2.8678467224275432,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004177641967040856,
      "loss": 5.0329,
      "mean_token_accuracy": 0.20462592840194702,
      "num_tokens": 63942368.0,
      "step": 36860
    },
    {
      "entropy": 5.814330387115478,
      "epoch": 2.868235751799261,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041774277378129595,
      "loss": 4.986,
      "mean_token_accuracy": 0.21063803732395173,
      "num_tokens": 63950739.0,
      "step": 36865
    },
    {
      "entropy": 5.748255205154419,
      "epoch": 2.868624781170978,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041772134869261186,
      "loss": 4.9545,
      "mean_token_accuracy": 0.21328649073839187,
      "num_tokens": 63959735.0,
      "step": 36870
    },
    {
      "entropy": 5.790006971359253,
      "epoch": 2.869013810542696,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004176999214383585,
      "loss": 4.9793,
      "mean_token_accuracy": 0.2072807177901268,
      "num_tokens": 63968771.0,
      "step": 36875
    },
    {
      "entropy": 5.747616147994995,
      "epoch": 2.8694028399144136,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041767849201886103,
      "loss": 4.935,
      "mean_token_accuracy": 0.2161041244864464,
      "num_tokens": 63977595.0,
      "step": 36880
    },
    {
      "entropy": 5.772293424606323,
      "epoch": 2.8697918692861313,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004176570604344445,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20684693604707718,
      "num_tokens": 63985676.0,
      "step": 36885
    },
    {
      "entropy": 5.817913770675659,
      "epoch": 2.8701808986578485,
      "grad_norm": 1.25,
      "learning_rate": 0.00041763562668543415,
      "loss": 5.0938,
      "mean_token_accuracy": 0.20178003013134002,
      "num_tokens": 63994134.0,
      "step": 36890
    },
    {
      "entropy": 5.855589962005615,
      "epoch": 2.8705699280295662,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004176141907721552,
      "loss": 5.1181,
      "mean_token_accuracy": 0.2042589321732521,
      "num_tokens": 64002605.0,
      "step": 36895
    },
    {
      "entropy": 5.756672477722168,
      "epoch": 2.870958957401284,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004175927526949329,
      "loss": 5.0205,
      "mean_token_accuracy": 0.20548584014177323,
      "num_tokens": 64011232.0,
      "step": 36900
    },
    {
      "entropy": 5.7165374755859375,
      "epoch": 2.871347986773001,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004175713124540925,
      "loss": 4.927,
      "mean_token_accuracy": 0.2085779845714569,
      "num_tokens": 64019696.0,
      "step": 36905
    },
    {
      "entropy": 5.859394836425781,
      "epoch": 2.871737016144719,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041754987004995935,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20564687103033066,
      "num_tokens": 64028716.0,
      "step": 36910
    },
    {
      "entropy": 5.805535984039307,
      "epoch": 2.8721260455164366,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041752842548285887,
      "loss": 4.9759,
      "mean_token_accuracy": 0.2111139938235283,
      "num_tokens": 64036915.0,
      "step": 36915
    },
    {
      "entropy": 5.731641340255737,
      "epoch": 2.8725150748881543,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004175069787531163,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20971932709217073,
      "num_tokens": 64045857.0,
      "step": 36920
    },
    {
      "entropy": 5.813305854797363,
      "epoch": 2.8729041042598715,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041748552986105726,
      "loss": 5.044,
      "mean_token_accuracy": 0.20651862621307374,
      "num_tokens": 64054958.0,
      "step": 36925
    },
    {
      "entropy": 5.790309238433838,
      "epoch": 2.8732931336315892,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000417464078807007,
      "loss": 5.0183,
      "mean_token_accuracy": 0.203393816947937,
      "num_tokens": 64063745.0,
      "step": 36930
    },
    {
      "entropy": 5.8528989315032955,
      "epoch": 2.8736821630033065,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004174426255912912,
      "loss": 5.0007,
      "mean_token_accuracy": 0.2031003177165985,
      "num_tokens": 64072496.0,
      "step": 36935
    },
    {
      "entropy": 5.825888061523438,
      "epoch": 2.874071192375024,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004174211702142352,
      "loss": 4.9791,
      "mean_token_accuracy": 0.21759361028671265,
      "num_tokens": 64081112.0,
      "step": 36940
    },
    {
      "entropy": 5.744197368621826,
      "epoch": 2.874460221746742,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041739971267616466,
      "loss": 4.8835,
      "mean_token_accuracy": 0.21157485544681548,
      "num_tokens": 64089346.0,
      "step": 36945
    },
    {
      "entropy": 5.818365859985351,
      "epoch": 2.8748492511184596,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004173782529774051,
      "loss": 5.0363,
      "mean_token_accuracy": 0.204910446703434,
      "num_tokens": 64097590.0,
      "step": 36950
    },
    {
      "entropy": 5.813367080688477,
      "epoch": 2.8752382804901773,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041735679111828223,
      "loss": 4.9835,
      "mean_token_accuracy": 0.20443275570869446,
      "num_tokens": 64105870.0,
      "step": 36955
    },
    {
      "entropy": 5.839142942428589,
      "epoch": 2.8756273098618945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041733532709912157,
      "loss": 5.0369,
      "mean_token_accuracy": 0.20096689760684966,
      "num_tokens": 64114963.0,
      "step": 36960
    },
    {
      "entropy": 5.859133529663086,
      "epoch": 2.876016339233612,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041731386092024893,
      "loss": 5.1246,
      "mean_token_accuracy": 0.19250556230545043,
      "num_tokens": 64123543.0,
      "step": 36965
    },
    {
      "entropy": 5.808759880065918,
      "epoch": 2.8764053686053295,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041729239258198996,
      "loss": 5.0478,
      "mean_token_accuracy": 0.20474354773759842,
      "num_tokens": 64131962.0,
      "step": 36970
    },
    {
      "entropy": 5.788818073272705,
      "epoch": 2.876794397977047,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041727092208467036,
      "loss": 5.0098,
      "mean_token_accuracy": 0.21519155353307723,
      "num_tokens": 64140292.0,
      "step": 36975
    },
    {
      "entropy": 5.804125690460205,
      "epoch": 2.877183427348765,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041724944942861603,
      "loss": 4.9626,
      "mean_token_accuracy": 0.20909482836723328,
      "num_tokens": 64148936.0,
      "step": 36980
    },
    {
      "entropy": 5.8505859375,
      "epoch": 2.8775724567204826,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041722797461415267,
      "loss": 5.0339,
      "mean_token_accuracy": 0.2014971613883972,
      "num_tokens": 64157801.0,
      "step": 36985
    },
    {
      "entropy": 5.844186687469483,
      "epoch": 2.8779614860922,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004172064976416062,
      "loss": 5.0655,
      "mean_token_accuracy": 0.20048424899578093,
      "num_tokens": 64166049.0,
      "step": 36990
    },
    {
      "entropy": 5.82602596282959,
      "epoch": 2.8783505154639175,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004171850185113024,
      "loss": 5.0802,
      "mean_token_accuracy": 0.20038343220949173,
      "num_tokens": 64174663.0,
      "step": 36995
    },
    {
      "entropy": 5.86979866027832,
      "epoch": 2.878739544835635,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041716353722356735,
      "loss": 5.0908,
      "mean_token_accuracy": 0.2006244108080864,
      "num_tokens": 64183683.0,
      "step": 37000
    },
    {
      "entropy": 5.869486141204834,
      "epoch": 2.8791285742073525,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004171420537787269,
      "loss": 5.0571,
      "mean_token_accuracy": 0.20415510684251786,
      "num_tokens": 64192808.0,
      "step": 37005
    },
    {
      "entropy": 5.817481565475464,
      "epoch": 2.87951760357907,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041712056817710697,
      "loss": 4.9612,
      "mean_token_accuracy": 0.2056852951645851,
      "num_tokens": 64201703.0,
      "step": 37010
    },
    {
      "entropy": 5.80070481300354,
      "epoch": 2.879906632950788,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004170990804190337,
      "loss": 5.0453,
      "mean_token_accuracy": 0.19815419912338256,
      "num_tokens": 64210343.0,
      "step": 37015
    },
    {
      "entropy": 5.8725025177001955,
      "epoch": 2.8802956623225056,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041707759050483304,
      "loss": 5.063,
      "mean_token_accuracy": 0.20444852113723755,
      "num_tokens": 64218871.0,
      "step": 37020
    },
    {
      "entropy": 5.824731969833374,
      "epoch": 2.880684691694223,
      "grad_norm": 1.25,
      "learning_rate": 0.000417056098434831,
      "loss": 4.9096,
      "mean_token_accuracy": 0.21652227342128755,
      "num_tokens": 64226300.0,
      "step": 37025
    },
    {
      "entropy": 5.76281852722168,
      "epoch": 2.8810737210659405,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004170346042093538,
      "loss": 4.9767,
      "mean_token_accuracy": 0.2041445329785347,
      "num_tokens": 64235042.0,
      "step": 37030
    },
    {
      "entropy": 5.766773986816406,
      "epoch": 2.881462750437658,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004170131078287276,
      "loss": 4.9341,
      "mean_token_accuracy": 0.21288384199142457,
      "num_tokens": 64242942.0,
      "step": 37035
    },
    {
      "entropy": 5.859264612197876,
      "epoch": 2.8818517798093755,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004169916092932785,
      "loss": 5.0616,
      "mean_token_accuracy": 0.19874853938817977,
      "num_tokens": 64251187.0,
      "step": 37040
    },
    {
      "entropy": 5.863649320602417,
      "epoch": 2.882240809181093,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004169701086033327,
      "loss": 5.0846,
      "mean_token_accuracy": 0.20325787216424943,
      "num_tokens": 64260025.0,
      "step": 37045
    },
    {
      "entropy": 5.79508228302002,
      "epoch": 2.882629838552811,
      "grad_norm": 1.375,
      "learning_rate": 0.00041694860575921654,
      "loss": 5.0803,
      "mean_token_accuracy": 0.20503873378038406,
      "num_tokens": 64269940.0,
      "step": 37050
    },
    {
      "entropy": 5.847331237792969,
      "epoch": 2.8830188679245285,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041692710076125615,
      "loss": 5.0463,
      "mean_token_accuracy": 0.19746011197566987,
      "num_tokens": 64279054.0,
      "step": 37055
    },
    {
      "entropy": 5.819596529006958,
      "epoch": 2.883407897296246,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000416905593609778,
      "loss": 5.0488,
      "mean_token_accuracy": 0.2018558457493782,
      "num_tokens": 64287332.0,
      "step": 37060
    },
    {
      "entropy": 5.807564544677734,
      "epoch": 2.8837969266679635,
      "grad_norm": 1.25,
      "learning_rate": 0.00041688408430510827,
      "loss": 4.9848,
      "mean_token_accuracy": 0.20878450572490692,
      "num_tokens": 64295889.0,
      "step": 37065
    },
    {
      "entropy": 5.819533395767212,
      "epoch": 2.8841859560396808,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004168625728475734,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20370020866394042,
      "num_tokens": 64305138.0,
      "step": 37070
    },
    {
      "entropy": 5.7920952320098875,
      "epoch": 2.8845749854113985,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041684105923749977,
      "loss": 5.0431,
      "mean_token_accuracy": 0.20553875863552093,
      "num_tokens": 64314041.0,
      "step": 37075
    },
    {
      "entropy": 5.921089506149292,
      "epoch": 2.884964014783116,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004168195434752139,
      "loss": 5.0831,
      "mean_token_accuracy": 0.19332961440086366,
      "num_tokens": 64323686.0,
      "step": 37080
    },
    {
      "entropy": 5.812370300292969,
      "epoch": 2.885353044154834,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041679802556104207,
      "loss": 5.0174,
      "mean_token_accuracy": 0.20155178755521774,
      "num_tokens": 64332395.0,
      "step": 37085
    },
    {
      "entropy": 5.7963550090789795,
      "epoch": 2.885742073526551,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041677650549531097,
      "loss": 4.9902,
      "mean_token_accuracy": 0.21578869968652725,
      "num_tokens": 64340577.0,
      "step": 37090
    },
    {
      "entropy": 5.767085695266724,
      "epoch": 2.886131102898269,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041675498327834706,
      "loss": 5.0122,
      "mean_token_accuracy": 0.2005954697728157,
      "num_tokens": 64348903.0,
      "step": 37095
    },
    {
      "entropy": 5.881518745422364,
      "epoch": 2.8865201322699865,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041673345891047686,
      "loss": 5.1201,
      "mean_token_accuracy": 0.19618806391954421,
      "num_tokens": 64357725.0,
      "step": 37100
    },
    {
      "entropy": 5.845756101608276,
      "epoch": 2.8869091616417037,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004167119323920271,
      "loss": 5.0808,
      "mean_token_accuracy": 0.20114561915397644,
      "num_tokens": 64366871.0,
      "step": 37105
    },
    {
      "entropy": 5.782095241546631,
      "epoch": 2.8872981910134214,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004166904037233243,
      "loss": 5.0166,
      "mean_token_accuracy": 0.19824107140302658,
      "num_tokens": 64375155.0,
      "step": 37110
    },
    {
      "entropy": 5.7788074016571045,
      "epoch": 2.887687220385139,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041666887290469506,
      "loss": 4.9589,
      "mean_token_accuracy": 0.21133533120155334,
      "num_tokens": 64383956.0,
      "step": 37115
    },
    {
      "entropy": 5.843722534179688,
      "epoch": 2.888076249756857,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004166473399364663,
      "loss": 5.0573,
      "mean_token_accuracy": 0.20161016285419464,
      "num_tokens": 64392816.0,
      "step": 37120
    },
    {
      "entropy": 5.802755641937256,
      "epoch": 2.888465279128574,
      "grad_norm": 1.25,
      "learning_rate": 0.00041662580481896446,
      "loss": 5.0177,
      "mean_token_accuracy": 0.20772883296012878,
      "num_tokens": 64401676.0,
      "step": 37125
    },
    {
      "entropy": 5.870348262786865,
      "epoch": 2.888854308500292,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004166042675525165,
      "loss": 5.1637,
      "mean_token_accuracy": 0.199164018034935,
      "num_tokens": 64410248.0,
      "step": 37130
    },
    {
      "entropy": 5.814644813537598,
      "epoch": 2.8892433378720095,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041658272813744924,
      "loss": 5.0076,
      "mean_token_accuracy": 0.20659082531929016,
      "num_tokens": 64419023.0,
      "step": 37135
    },
    {
      "entropy": 5.804802083969117,
      "epoch": 2.8896323672437267,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004165611865740894,
      "loss": 4.9078,
      "mean_token_accuracy": 0.20471362322568892,
      "num_tokens": 64427775.0,
      "step": 37140
    },
    {
      "entropy": 5.768785429000855,
      "epoch": 2.8900213966154444,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004165396428627638,
      "loss": 4.9849,
      "mean_token_accuracy": 0.20719853341579436,
      "num_tokens": 64436113.0,
      "step": 37145
    },
    {
      "entropy": 5.78173599243164,
      "epoch": 2.890410425987162,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041651809700379947,
      "loss": 5.0207,
      "mean_token_accuracy": 0.21192073673009873,
      "num_tokens": 64444325.0,
      "step": 37150
    },
    {
      "entropy": 5.7634283065795895,
      "epoch": 2.89079945535888,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041649654899752326,
      "loss": 4.9626,
      "mean_token_accuracy": 0.2086004674434662,
      "num_tokens": 64452560.0,
      "step": 37155
    },
    {
      "entropy": 5.806577491760254,
      "epoch": 2.891188484730597,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004164749988442621,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20923152714967727,
      "num_tokens": 64460983.0,
      "step": 37160
    },
    {
      "entropy": 5.798651266098022,
      "epoch": 2.891577514102315,
      "grad_norm": 1.15625,
      "learning_rate": 0.000416453446544343,
      "loss": 5.0664,
      "mean_token_accuracy": 0.19992144256830216,
      "num_tokens": 64470103.0,
      "step": 37165
    },
    {
      "entropy": 5.8343236446380615,
      "epoch": 2.891966543474032,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004164318920980931,
      "loss": 4.9575,
      "mean_token_accuracy": 0.21058415174484252,
      "num_tokens": 64478642.0,
      "step": 37170
    },
    {
      "entropy": 5.775297355651856,
      "epoch": 2.8923555728457497,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004164103355058393,
      "loss": 4.9875,
      "mean_token_accuracy": 0.2042151838541031,
      "num_tokens": 64487625.0,
      "step": 37175
    },
    {
      "entropy": 5.78716630935669,
      "epoch": 2.8927446022174674,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004163887767679087,
      "loss": 4.9821,
      "mean_token_accuracy": 0.21438337862491608,
      "num_tokens": 64497400.0,
      "step": 37180
    },
    {
      "entropy": 5.877262926101684,
      "epoch": 2.893133631589185,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004163672158846285,
      "loss": 5.0445,
      "mean_token_accuracy": 0.2042357236146927,
      "num_tokens": 64506119.0,
      "step": 37185
    },
    {
      "entropy": 5.796861696243286,
      "epoch": 2.8935226609609024,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041634565285632573,
      "loss": 4.9144,
      "mean_token_accuracy": 0.20339482575654982,
      "num_tokens": 64514372.0,
      "step": 37190
    },
    {
      "entropy": 5.770101833343506,
      "epoch": 2.89391169033262,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041632408768332776,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20458300411701202,
      "num_tokens": 64522877.0,
      "step": 37195
    },
    {
      "entropy": 5.715446043014526,
      "epoch": 2.8943007197043378,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041630252036596165,
      "loss": 4.8292,
      "mean_token_accuracy": 0.22165483832359315,
      "num_tokens": 64531647.0,
      "step": 37200
    },
    {
      "entropy": 5.750990295410157,
      "epoch": 2.894689749076055,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004162809509045547,
      "loss": 4.9641,
      "mean_token_accuracy": 0.20730386823415756,
      "num_tokens": 64540323.0,
      "step": 37205
    },
    {
      "entropy": 5.824655055999756,
      "epoch": 2.8950787784477727,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041625937929943424,
      "loss": 5.0507,
      "mean_token_accuracy": 0.20332174748182297,
      "num_tokens": 64548955.0,
      "step": 37210
    },
    {
      "entropy": 5.816799879074097,
      "epoch": 2.8954678078194904,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041623780555092747,
      "loss": 5.0411,
      "mean_token_accuracy": 0.20245813131332396,
      "num_tokens": 64557744.0,
      "step": 37215
    },
    {
      "entropy": 5.758823537826538,
      "epoch": 2.895856837191208,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004162162296593618,
      "loss": 4.8842,
      "mean_token_accuracy": 0.2136201187968254,
      "num_tokens": 64566119.0,
      "step": 37220
    },
    {
      "entropy": 5.778063344955444,
      "epoch": 2.8962458665629254,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041619465162506466,
      "loss": 5.0089,
      "mean_token_accuracy": 0.21109870374202727,
      "num_tokens": 64574995.0,
      "step": 37225
    },
    {
      "entropy": 5.817973041534424,
      "epoch": 2.896634895934643,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004161730714483633,
      "loss": 5.1011,
      "mean_token_accuracy": 0.20219702273607254,
      "num_tokens": 64583667.0,
      "step": 37230
    },
    {
      "entropy": 5.820700025558471,
      "epoch": 2.8970239253063608,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004161514891295854,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21453402936458588,
      "num_tokens": 64591857.0,
      "step": 37235
    },
    {
      "entropy": 5.786137104034424,
      "epoch": 2.897412954678078,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041612990466905825,
      "loss": 5.0461,
      "mean_token_accuracy": 0.20348697602748872,
      "num_tokens": 64600635.0,
      "step": 37240
    },
    {
      "entropy": 5.798447990417481,
      "epoch": 2.8978019840497957,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00041610831806710954,
      "loss": 4.9856,
      "mean_token_accuracy": 0.20221203863620757,
      "num_tokens": 64609141.0,
      "step": 37245
    },
    {
      "entropy": 5.837018632888794,
      "epoch": 2.8981910134215134,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041608672932406666,
      "loss": 5.0166,
      "mean_token_accuracy": 0.20852839350700378,
      "num_tokens": 64617804.0,
      "step": 37250
    },
    {
      "entropy": 5.817765140533448,
      "epoch": 2.898580042793231,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041606513844025716,
      "loss": 4.878,
      "mean_token_accuracy": 0.21432213634252548,
      "num_tokens": 64626119.0,
      "step": 37255
    },
    {
      "entropy": 5.806481742858887,
      "epoch": 2.8989690721649484,
      "grad_norm": 1.375,
      "learning_rate": 0.0004160435454160087,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20382003635168075,
      "num_tokens": 64634483.0,
      "step": 37260
    },
    {
      "entropy": 5.750558662414551,
      "epoch": 2.899358101536666,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000416021950251649,
      "loss": 5.031,
      "mean_token_accuracy": 0.2036902442574501,
      "num_tokens": 64642500.0,
      "step": 37265
    },
    {
      "entropy": 5.765313243865966,
      "epoch": 2.8997471309083833,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004160003529475057,
      "loss": 5.0177,
      "mean_token_accuracy": 0.20162294656038285,
      "num_tokens": 64651808.0,
      "step": 37270
    },
    {
      "entropy": 5.816401195526123,
      "epoch": 2.900136160280101,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004159787535039064,
      "loss": 5.0256,
      "mean_token_accuracy": 0.20391747504472732,
      "num_tokens": 64660502.0,
      "step": 37275
    },
    {
      "entropy": 5.839527750015259,
      "epoch": 2.9005251896518187,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004159571519211789,
      "loss": 5.0827,
      "mean_token_accuracy": 0.20521850436925887,
      "num_tokens": 64669156.0,
      "step": 37280
    },
    {
      "entropy": 5.854022884368897,
      "epoch": 2.9009142190235364,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000415935548199651,
      "loss": 5.0621,
      "mean_token_accuracy": 0.20263974070549012,
      "num_tokens": 64678321.0,
      "step": 37285
    },
    {
      "entropy": 5.861771726608277,
      "epoch": 2.901303248395254,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004159139423396504,
      "loss": 5.0249,
      "mean_token_accuracy": 0.19914842247962952,
      "num_tokens": 64686768.0,
      "step": 37290
    },
    {
      "entropy": 5.897479724884033,
      "epoch": 2.9016922777669714,
      "grad_norm": 1.25,
      "learning_rate": 0.0004158923343415051,
      "loss": 5.0965,
      "mean_token_accuracy": 0.20513201355934144,
      "num_tokens": 64695357.0,
      "step": 37295
    },
    {
      "entropy": 5.7430730819702145,
      "epoch": 2.902081307138689,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004158707242055429,
      "loss": 5.0279,
      "mean_token_accuracy": 0.19847038388252258,
      "num_tokens": 64703912.0,
      "step": 37300
    },
    {
      "entropy": 5.733249807357788,
      "epoch": 2.9024703365104063,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041584911193209167,
      "loss": 4.9441,
      "mean_token_accuracy": 0.21311118751764296,
      "num_tokens": 64713088.0,
      "step": 37305
    },
    {
      "entropy": 5.830707693099976,
      "epoch": 2.902859365882124,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004158274975214794,
      "loss": 5.0241,
      "mean_token_accuracy": 0.19602841287851333,
      "num_tokens": 64721950.0,
      "step": 37310
    },
    {
      "entropy": 5.849611902236939,
      "epoch": 2.9032483952538417,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041580588097403384,
      "loss": 5.0018,
      "mean_token_accuracy": 0.19721185714006423,
      "num_tokens": 64731032.0,
      "step": 37315
    },
    {
      "entropy": 5.765616989135742,
      "epoch": 2.9036374246255594,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041578426229008325,
      "loss": 5.0352,
      "mean_token_accuracy": 0.20342419892549515,
      "num_tokens": 64739676.0,
      "step": 37320
    },
    {
      "entropy": 5.788762474060059,
      "epoch": 2.9040264539972767,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041576264146995557,
      "loss": 4.9854,
      "mean_token_accuracy": 0.21113596856594086,
      "num_tokens": 64748827.0,
      "step": 37325
    },
    {
      "entropy": 5.784723711013794,
      "epoch": 2.9044154833689944,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041574101851397883,
      "loss": 4.9035,
      "mean_token_accuracy": 0.21476234048604964,
      "num_tokens": 64757990.0,
      "step": 37330
    },
    {
      "entropy": 5.7669695854187015,
      "epoch": 2.904804512740712,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004157193934224811,
      "loss": 5.087,
      "mean_token_accuracy": 0.20181526690721513,
      "num_tokens": 64767351.0,
      "step": 37335
    },
    {
      "entropy": 5.812999057769775,
      "epoch": 2.9051935421124293,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004156977661957906,
      "loss": 5.0582,
      "mean_token_accuracy": 0.2091486155986786,
      "num_tokens": 64775294.0,
      "step": 37340
    },
    {
      "entropy": 5.810967636108399,
      "epoch": 2.905582571484147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004156761368342355,
      "loss": 5.0381,
      "mean_token_accuracy": 0.21480512470006943,
      "num_tokens": 64784933.0,
      "step": 37345
    },
    {
      "entropy": 5.837831830978393,
      "epoch": 2.9059716008558647,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041565450533814384,
      "loss": 4.9796,
      "mean_token_accuracy": 0.21399618983268737,
      "num_tokens": 64794137.0,
      "step": 37350
    },
    {
      "entropy": 5.820476341247558,
      "epoch": 2.9063606302275824,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041563287170784396,
      "loss": 5.0055,
      "mean_token_accuracy": 0.20169112086296082,
      "num_tokens": 64803100.0,
      "step": 37355
    },
    {
      "entropy": 5.710199785232544,
      "epoch": 2.9067496595992997,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004156112359436641,
      "loss": 4.9246,
      "mean_token_accuracy": 0.2127266749739647,
      "num_tokens": 64811980.0,
      "step": 37360
    },
    {
      "entropy": 5.781800079345703,
      "epoch": 2.9071386889710173,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041558959804593255,
      "loss": 5.002,
      "mean_token_accuracy": 0.20657373666763307,
      "num_tokens": 64820201.0,
      "step": 37365
    },
    {
      "entropy": 5.798106336593628,
      "epoch": 2.9075277183427346,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041556795801497766,
      "loss": 4.9686,
      "mean_token_accuracy": 0.21226968467235566,
      "num_tokens": 64829054.0,
      "step": 37370
    },
    {
      "entropy": 5.861152935028076,
      "epoch": 2.9079167477144523,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004155463158511277,
      "loss": 5.1296,
      "mean_token_accuracy": 0.19851900041103362,
      "num_tokens": 64838704.0,
      "step": 37375
    },
    {
      "entropy": 5.769803524017334,
      "epoch": 2.90830577708617,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041552467155471105,
      "loss": 4.8314,
      "mean_token_accuracy": 0.21827107965946196,
      "num_tokens": 64847508.0,
      "step": 37380
    },
    {
      "entropy": 5.806541872024536,
      "epoch": 2.9086948064578877,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004155030251260563,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21588921099901198,
      "num_tokens": 64856112.0,
      "step": 37385
    },
    {
      "entropy": 5.748947048187256,
      "epoch": 2.9090838358296054,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041548137656549167,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21236215382814408,
      "num_tokens": 64864807.0,
      "step": 37390
    },
    {
      "entropy": 5.798714923858642,
      "epoch": 2.9094728652013226,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004154597258733457,
      "loss": 4.9886,
      "mean_token_accuracy": 0.20708333551883698,
      "num_tokens": 64872744.0,
      "step": 37395
    },
    {
      "entropy": 5.799120569229126,
      "epoch": 2.9098618945730403,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004154380730499472,
      "loss": 4.9604,
      "mean_token_accuracy": 0.21208524852991104,
      "num_tokens": 64882581.0,
      "step": 37400
    },
    {
      "entropy": 5.75370078086853,
      "epoch": 2.9102509239447576,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041541641809562425,
      "loss": 4.9084,
      "mean_token_accuracy": 0.20981716215610505,
      "num_tokens": 64890935.0,
      "step": 37405
    },
    {
      "entropy": 5.826978302001953,
      "epoch": 2.9106399533164753,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004153947610107057,
      "loss": 5.0923,
      "mean_token_accuracy": 0.20015419274568558,
      "num_tokens": 64900132.0,
      "step": 37410
    },
    {
      "entropy": 5.818065214157104,
      "epoch": 2.911028982688193,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041537310179552017,
      "loss": 5.0191,
      "mean_token_accuracy": 0.2030436083674431,
      "num_tokens": 64908733.0,
      "step": 37415
    },
    {
      "entropy": 5.816516876220703,
      "epoch": 2.9114180120599107,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004153514404503962,
      "loss": 5.0125,
      "mean_token_accuracy": 0.2034083917737007,
      "num_tokens": 64916824.0,
      "step": 37420
    },
    {
      "entropy": 5.823220491409302,
      "epoch": 2.911807041431628,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004153297769756625,
      "loss": 5.008,
      "mean_token_accuracy": 0.20174604803323745,
      "num_tokens": 64926342.0,
      "step": 37425
    },
    {
      "entropy": 5.84201831817627,
      "epoch": 2.9121960708033456,
      "grad_norm": 1.25,
      "learning_rate": 0.0004153081113716478,
      "loss": 5.0174,
      "mean_token_accuracy": 0.20892584174871445,
      "num_tokens": 64935123.0,
      "step": 37430
    },
    {
      "entropy": 5.807203865051269,
      "epoch": 2.9125851001750633,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041528644363868087,
      "loss": 5.0422,
      "mean_token_accuracy": 0.20226745158433915,
      "num_tokens": 64944046.0,
      "step": 37435
    },
    {
      "entropy": 5.9178893089294435,
      "epoch": 2.9129741295467806,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004152647737770904,
      "loss": 5.1,
      "mean_token_accuracy": 0.19120790362358092,
      "num_tokens": 64952897.0,
      "step": 37440
    },
    {
      "entropy": 5.896544933319092,
      "epoch": 2.9133631589184983,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004152431017872053,
      "loss": 5.062,
      "mean_token_accuracy": 0.20506420731544495,
      "num_tokens": 64961432.0,
      "step": 37445
    },
    {
      "entropy": 5.837743043899536,
      "epoch": 2.913752188290216,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041522142766935426,
      "loss": 5.0937,
      "mean_token_accuracy": 0.19887920171022416,
      "num_tokens": 64970298.0,
      "step": 37450
    },
    {
      "entropy": 5.839676761627198,
      "epoch": 2.9141412176619337,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041519975142386627,
      "loss": 5.0761,
      "mean_token_accuracy": 0.20092874467372895,
      "num_tokens": 64978735.0,
      "step": 37455
    },
    {
      "entropy": 5.833218717575074,
      "epoch": 2.914530247033651,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004151780730510702,
      "loss": 5.0293,
      "mean_token_accuracy": 0.2011540338397026,
      "num_tokens": 64987222.0,
      "step": 37460
    },
    {
      "entropy": 5.737643241882324,
      "epoch": 2.9149192764053686,
      "grad_norm": 1.25,
      "learning_rate": 0.00041515639255129504,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21553818732500077,
      "num_tokens": 64995857.0,
      "step": 37465
    },
    {
      "entropy": 5.762152814865113,
      "epoch": 2.9153083057770863,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004151347099248697,
      "loss": 4.938,
      "mean_token_accuracy": 0.2085283324122429,
      "num_tokens": 65005583.0,
      "step": 37470
    },
    {
      "entropy": 5.818338251113891,
      "epoch": 2.9156973351488036,
      "grad_norm": 1.25,
      "learning_rate": 0.0004151130251721231,
      "loss": 4.9885,
      "mean_token_accuracy": 0.20802012830972672,
      "num_tokens": 65013533.0,
      "step": 37475
    },
    {
      "entropy": 5.796665334701538,
      "epoch": 2.9160863645205213,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041509133829338454,
      "loss": 4.9889,
      "mean_token_accuracy": 0.2163096308708191,
      "num_tokens": 65022674.0,
      "step": 37480
    },
    {
      "entropy": 5.763079833984375,
      "epoch": 2.916475393892239,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004150696492889827,
      "loss": 4.9211,
      "mean_token_accuracy": 0.21469176113605498,
      "num_tokens": 65030914.0,
      "step": 37485
    },
    {
      "entropy": 5.7955999851226805,
      "epoch": 2.9168644232639567,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041504795815924697,
      "loss": 5.0199,
      "mean_token_accuracy": 0.2059937059879303,
      "num_tokens": 65040036.0,
      "step": 37490
    },
    {
      "entropy": 5.826927661895752,
      "epoch": 2.917253452635674,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041502626490450635,
      "loss": 4.9681,
      "mean_token_accuracy": 0.21491282433271408,
      "num_tokens": 65048558.0,
      "step": 37495
    },
    {
      "entropy": 5.771075916290283,
      "epoch": 2.9176424820073916,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004150045695250901,
      "loss": 4.9752,
      "mean_token_accuracy": 0.2153475970029831,
      "num_tokens": 65057158.0,
      "step": 37500
    },
    {
      "entropy": 5.843840265274048,
      "epoch": 2.918031511379109,
      "grad_norm": 1.25,
      "learning_rate": 0.0004149828720213272,
      "loss": 5.0613,
      "mean_token_accuracy": 0.2095615014433861,
      "num_tokens": 65065845.0,
      "step": 37505
    },
    {
      "entropy": 5.8007568359375,
      "epoch": 2.9184205407508266,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004149611723935472,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20663176029920577,
      "num_tokens": 65075182.0,
      "step": 37510
    },
    {
      "entropy": 5.8136069774627686,
      "epoch": 2.9188095701225443,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004149394706420791,
      "loss": 4.9973,
      "mean_token_accuracy": 0.2092272460460663,
      "num_tokens": 65084616.0,
      "step": 37515
    },
    {
      "entropy": 5.694690132141114,
      "epoch": 2.919198599494262,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041491776676725223,
      "loss": 4.8818,
      "mean_token_accuracy": 0.2258787289261818,
      "num_tokens": 65092683.0,
      "step": 37520
    },
    {
      "entropy": 5.755844926834106,
      "epoch": 2.9195876288659792,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004148960607693961,
      "loss": 4.9535,
      "mean_token_accuracy": 0.20546974539756774,
      "num_tokens": 65100445.0,
      "step": 37525
    },
    {
      "entropy": 5.805068922042847,
      "epoch": 2.919976658237697,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041487435264883974,
      "loss": 5.0094,
      "mean_token_accuracy": 0.20912558138370513,
      "num_tokens": 65109512.0,
      "step": 37530
    },
    {
      "entropy": 5.7811380386352536,
      "epoch": 2.9203656876094146,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004148526424059128,
      "loss": 4.9893,
      "mean_token_accuracy": 0.20580063611268998,
      "num_tokens": 65117889.0,
      "step": 37535
    },
    {
      "entropy": 5.812647533416748,
      "epoch": 2.920754716981132,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004148309300409446,
      "loss": 5.0366,
      "mean_token_accuracy": 0.20423102378845215,
      "num_tokens": 65126226.0,
      "step": 37540
    },
    {
      "entropy": 5.78339033126831,
      "epoch": 2.9211437463528496,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004148092155542646,
      "loss": 4.9798,
      "mean_token_accuracy": 0.20625606179237366,
      "num_tokens": 65134474.0,
      "step": 37545
    },
    {
      "entropy": 5.830212211608886,
      "epoch": 2.9215327757245673,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004147874989462023,
      "loss": 5.077,
      "mean_token_accuracy": 0.2002380073070526,
      "num_tokens": 65143068.0,
      "step": 37550
    },
    {
      "entropy": 5.769702625274658,
      "epoch": 2.921921805096285,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004147657802170871,
      "loss": 4.985,
      "mean_token_accuracy": 0.2112153336405754,
      "num_tokens": 65151865.0,
      "step": 37555
    },
    {
      "entropy": 5.769666242599487,
      "epoch": 2.922310834468002,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041474405936724873,
      "loss": 4.9679,
      "mean_token_accuracy": 0.20770518481731415,
      "num_tokens": 65160684.0,
      "step": 37560
    },
    {
      "entropy": 5.836337327957153,
      "epoch": 2.92269986383972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004147223363970166,
      "loss": 5.0184,
      "mean_token_accuracy": 0.20567017942667007,
      "num_tokens": 65169084.0,
      "step": 37565
    },
    {
      "entropy": 5.723083019256592,
      "epoch": 2.9230888932114376,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004147006113067205,
      "loss": 4.8744,
      "mean_token_accuracy": 0.2138003408908844,
      "num_tokens": 65178428.0,
      "step": 37570
    },
    {
      "entropy": 5.787231206893921,
      "epoch": 2.923477922583155,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004146788840966901,
      "loss": 4.9637,
      "mean_token_accuracy": 0.20303383618593215,
      "num_tokens": 65187490.0,
      "step": 37575
    },
    {
      "entropy": 5.814029550552368,
      "epoch": 2.9238669519548726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004146571547672548,
      "loss": 5.0143,
      "mean_token_accuracy": 0.20838441401720048,
      "num_tokens": 65195987.0,
      "step": 37580
    },
    {
      "entropy": 5.750436878204345,
      "epoch": 2.9242559813265903,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041463542331874443,
      "loss": 4.9658,
      "mean_token_accuracy": 0.2072920098900795,
      "num_tokens": 65204466.0,
      "step": 37585
    },
    {
      "entropy": 5.782142639160156,
      "epoch": 2.924645010698308,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041461368975148876,
      "loss": 4.949,
      "mean_token_accuracy": 0.21029426753520966,
      "num_tokens": 65213023.0,
      "step": 37590
    },
    {
      "entropy": 5.860556888580322,
      "epoch": 2.925034040070025,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041459195406581763,
      "loss": 5.1259,
      "mean_token_accuracy": 0.2008754074573517,
      "num_tokens": 65221580.0,
      "step": 37595
    },
    {
      "entropy": 5.707093572616577,
      "epoch": 2.925423069441743,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004145702162620608,
      "loss": 4.984,
      "mean_token_accuracy": 0.20664133429527282,
      "num_tokens": 65230529.0,
      "step": 37600
    },
    {
      "entropy": 5.764512395858764,
      "epoch": 2.92581209881346,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041454847634054805,
      "loss": 4.99,
      "mean_token_accuracy": 0.20777445733547212,
      "num_tokens": 65239338.0,
      "step": 37605
    },
    {
      "entropy": 5.770494747161865,
      "epoch": 2.926201128185178,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041452673430160923,
      "loss": 4.9564,
      "mean_token_accuracy": 0.2069793552160263,
      "num_tokens": 65248523.0,
      "step": 37610
    },
    {
      "entropy": 5.8277137756347654,
      "epoch": 2.9265901575568956,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004145049901455744,
      "loss": 5.041,
      "mean_token_accuracy": 0.20219890028238297,
      "num_tokens": 65257740.0,
      "step": 37615
    },
    {
      "entropy": 5.827582931518554,
      "epoch": 2.9269791869286133,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041448324387277337,
      "loss": 4.9334,
      "mean_token_accuracy": 0.20945317447185516,
      "num_tokens": 65266728.0,
      "step": 37620
    },
    {
      "entropy": 5.844689178466797,
      "epoch": 2.927368216300331,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041446149548353606,
      "loss": 5.0922,
      "mean_token_accuracy": 0.1988937109708786,
      "num_tokens": 65275264.0,
      "step": 37625
    },
    {
      "entropy": 5.814268350601196,
      "epoch": 2.927757245672048,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004144397449781926,
      "loss": 5.0735,
      "mean_token_accuracy": 0.19979774057865143,
      "num_tokens": 65284247.0,
      "step": 37630
    },
    {
      "entropy": 5.812187004089355,
      "epoch": 2.928146275043766,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004144179923570729,
      "loss": 4.9641,
      "mean_token_accuracy": 0.20907962769269944,
      "num_tokens": 65293255.0,
      "step": 37635
    },
    {
      "entropy": 5.82471809387207,
      "epoch": 2.928535304415483,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004143962376205071,
      "loss": 5.0248,
      "mean_token_accuracy": 0.20541165471076966,
      "num_tokens": 65301945.0,
      "step": 37640
    },
    {
      "entropy": 5.80695652961731,
      "epoch": 2.928924333787201,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041437448076882526,
      "loss": 4.9578,
      "mean_token_accuracy": 0.21480377316474913,
      "num_tokens": 65310521.0,
      "step": 37645
    },
    {
      "entropy": 5.772918939590454,
      "epoch": 2.9293133631589185,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041435272180235757,
      "loss": 4.9975,
      "mean_token_accuracy": 0.20142122507095336,
      "num_tokens": 65318626.0,
      "step": 37650
    },
    {
      "entropy": 5.846328544616699,
      "epoch": 2.9297023925306362,
      "grad_norm": 1.296875,
      "learning_rate": 0.000414330960721434,
      "loss": 5.0184,
      "mean_token_accuracy": 0.202310049533844,
      "num_tokens": 65326098.0,
      "step": 37655
    },
    {
      "entropy": 5.842078161239624,
      "epoch": 2.9300914219023535,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041430919752638497,
      "loss": 5.0754,
      "mean_token_accuracy": 0.20232130140066146,
      "num_tokens": 65334705.0,
      "step": 37660
    },
    {
      "entropy": 5.829688215255738,
      "epoch": 2.930480451274071,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004142874322175406,
      "loss": 4.9592,
      "mean_token_accuracy": 0.20765394121408462,
      "num_tokens": 65343965.0,
      "step": 37665
    },
    {
      "entropy": 5.817558526992798,
      "epoch": 2.930869480645789,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041426566479523116,
      "loss": 4.9823,
      "mean_token_accuracy": 0.20613330900669097,
      "num_tokens": 65351467.0,
      "step": 37670
    },
    {
      "entropy": 5.774628305435181,
      "epoch": 2.931258510017506,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041424389525978683,
      "loss": 4.9688,
      "mean_token_accuracy": 0.20439230650663376,
      "num_tokens": 65359820.0,
      "step": 37675
    },
    {
      "entropy": 5.781310415267944,
      "epoch": 2.931647539389224,
      "grad_norm": 1.25,
      "learning_rate": 0.0004142221236115381,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21761462688446045,
      "num_tokens": 65368363.0,
      "step": 37680
    },
    {
      "entropy": 5.812486457824707,
      "epoch": 2.9320365687609415,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004142003498508152,
      "loss": 5.035,
      "mean_token_accuracy": 0.2088380828499794,
      "num_tokens": 65376771.0,
      "step": 37685
    },
    {
      "entropy": 5.8232825756072994,
      "epoch": 2.9324255981326592,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004141785739779486,
      "loss": 5.0774,
      "mean_token_accuracy": 0.2038669005036354,
      "num_tokens": 65385254.0,
      "step": 37690
    },
    {
      "entropy": 5.903758001327515,
      "epoch": 2.9328146275043765,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041415679599326853,
      "loss": 5.0211,
      "mean_token_accuracy": 0.19519663900136947,
      "num_tokens": 65393276.0,
      "step": 37695
    },
    {
      "entropy": 5.778813076019287,
      "epoch": 2.933203656876094,
      "grad_norm": 1.40625,
      "learning_rate": 0.00041413501589710567,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21075383573770523,
      "num_tokens": 65401745.0,
      "step": 37700
    },
    {
      "entropy": 5.844391298294068,
      "epoch": 2.9335926862478114,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004141132336897904,
      "loss": 5.1283,
      "mean_token_accuracy": 0.2011095628142357,
      "num_tokens": 65409982.0,
      "step": 37705
    },
    {
      "entropy": 5.784383726119995,
      "epoch": 2.933981715619529,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004140914493716533,
      "loss": 4.9794,
      "mean_token_accuracy": 0.21068640798330307,
      "num_tokens": 65419617.0,
      "step": 37710
    },
    {
      "entropy": 5.845173311233521,
      "epoch": 2.934370744991247,
      "grad_norm": 1.25,
      "learning_rate": 0.0004140696629430248,
      "loss": 5.0097,
      "mean_token_accuracy": 0.20728454738855362,
      "num_tokens": 65428230.0,
      "step": 37715
    },
    {
      "entropy": 5.8911927223205565,
      "epoch": 2.9347597743629645,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041404787440423534,
      "loss": 5.0417,
      "mean_token_accuracy": 0.2064600795507431,
      "num_tokens": 65437126.0,
      "step": 37720
    },
    {
      "entropy": 5.846333408355713,
      "epoch": 2.9351488037346822,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041402608375561595,
      "loss": 5.06,
      "mean_token_accuracy": 0.19700243771076204,
      "num_tokens": 65445910.0,
      "step": 37725
    },
    {
      "entropy": 5.792303800582886,
      "epoch": 2.9355378331063995,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041400429099749687,
      "loss": 5.0778,
      "mean_token_accuracy": 0.19839354157447814,
      "num_tokens": 65454371.0,
      "step": 37730
    },
    {
      "entropy": 5.828773021697998,
      "epoch": 2.935926862478117,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000413982496130209,
      "loss": 5.0318,
      "mean_token_accuracy": 0.20371669083833693,
      "num_tokens": 65463207.0,
      "step": 37735
    },
    {
      "entropy": 5.764865112304688,
      "epoch": 2.9363158918498344,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041396069915408294,
      "loss": 4.9905,
      "mean_token_accuracy": 0.20691478997468948,
      "num_tokens": 65471703.0,
      "step": 37740
    },
    {
      "entropy": 5.757013559341431,
      "epoch": 2.936704921221552,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041393890006944945,
      "loss": 4.9279,
      "mean_token_accuracy": 0.2161546006798744,
      "num_tokens": 65480516.0,
      "step": 37745
    },
    {
      "entropy": 5.787012004852295,
      "epoch": 2.93709395059327,
      "grad_norm": 1.25,
      "learning_rate": 0.0004139170988766393,
      "loss": 4.9835,
      "mean_token_accuracy": 0.20467787981033325,
      "num_tokens": 65489233.0,
      "step": 37750
    },
    {
      "entropy": 5.799408674240112,
      "epoch": 2.9374829799649875,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004138952955759833,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2064541757106781,
      "num_tokens": 65497827.0,
      "step": 37755
    },
    {
      "entropy": 5.7972640037536625,
      "epoch": 2.937872009336705,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041387349016781225,
      "loss": 5.0435,
      "mean_token_accuracy": 0.20737591087818147,
      "num_tokens": 65505629.0,
      "step": 37760
    },
    {
      "entropy": 5.8264671802520756,
      "epoch": 2.9382610387084225,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000413851682652457,
      "loss": 5.0997,
      "mean_token_accuracy": 0.19839289635419846,
      "num_tokens": 65514097.0,
      "step": 37765
    },
    {
      "entropy": 5.841024494171142,
      "epoch": 2.93865006808014,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004138298730302485,
      "loss": 5.0812,
      "mean_token_accuracy": 0.20563873797655105,
      "num_tokens": 65522868.0,
      "step": 37770
    },
    {
      "entropy": 5.831250333786011,
      "epoch": 2.9390390974518574,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004138080613015176,
      "loss": 4.972,
      "mean_token_accuracy": 0.20920588523149491,
      "num_tokens": 65531433.0,
      "step": 37775
    },
    {
      "entropy": 5.808960342407227,
      "epoch": 2.939428126823575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041378624746659536,
      "loss": 5.0113,
      "mean_token_accuracy": 0.2100109726190567,
      "num_tokens": 65541113.0,
      "step": 37780
    },
    {
      "entropy": 5.811487007141113,
      "epoch": 2.939817156195293,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041376443152581265,
      "loss": 4.9986,
      "mean_token_accuracy": 0.2041308179497719,
      "num_tokens": 65550103.0,
      "step": 37785
    },
    {
      "entropy": 5.799273443222046,
      "epoch": 2.9402061855670105,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041374261347950064,
      "loss": 5.0105,
      "mean_token_accuracy": 0.2080279991030693,
      "num_tokens": 65558618.0,
      "step": 37790
    },
    {
      "entropy": 5.833667993545532,
      "epoch": 2.9405952149387278,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004137207933279903,
      "loss": 4.9733,
      "mean_token_accuracy": 0.20529801696538924,
      "num_tokens": 65568025.0,
      "step": 37795
    },
    {
      "entropy": 5.7984785556793215,
      "epoch": 2.9409842443104455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041369897107161266,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20504961758852006,
      "num_tokens": 65576360.0,
      "step": 37800
    },
    {
      "entropy": 5.826273822784424,
      "epoch": 2.941373273682163,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041367714671069897,
      "loss": 5.0204,
      "mean_token_accuracy": 0.20957797765731812,
      "num_tokens": 65585764.0,
      "step": 37805
    },
    {
      "entropy": 5.791890621185303,
      "epoch": 2.9417623030538804,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041365532024558024,
      "loss": 5.032,
      "mean_token_accuracy": 0.20887646079063416,
      "num_tokens": 65594093.0,
      "step": 37810
    },
    {
      "entropy": 5.848724699020385,
      "epoch": 2.942151332425598,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004136334916765877,
      "loss": 4.976,
      "mean_token_accuracy": 0.20505814552307128,
      "num_tokens": 65602503.0,
      "step": 37815
    },
    {
      "entropy": 5.806312847137451,
      "epoch": 2.942540361797316,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041361166100405264,
      "loss": 4.9574,
      "mean_token_accuracy": 0.2143403947353363,
      "num_tokens": 65611107.0,
      "step": 37820
    },
    {
      "entropy": 5.777169895172119,
      "epoch": 2.9429293911690335,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004135898282283062,
      "loss": 5.0651,
      "mean_token_accuracy": 0.20864454209804534,
      "num_tokens": 65620008.0,
      "step": 37825
    },
    {
      "entropy": 5.7557793140411375,
      "epoch": 2.9433184205407508,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004135679933496797,
      "loss": 4.984,
      "mean_token_accuracy": 0.20655581653118132,
      "num_tokens": 65629206.0,
      "step": 37830
    },
    {
      "entropy": 5.792070960998535,
      "epoch": 2.9437074499124685,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004135461563685045,
      "loss": 5.0346,
      "mean_token_accuracy": 0.21161134093999862,
      "num_tokens": 65637403.0,
      "step": 37835
    },
    {
      "entropy": 5.8176158428192135,
      "epoch": 2.9440964792841857,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041352431728511194,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22102844715118408,
      "num_tokens": 65646123.0,
      "step": 37840
    },
    {
      "entropy": 5.83639931678772,
      "epoch": 2.9444855086559034,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004135024760998333,
      "loss": 5.0364,
      "mean_token_accuracy": 0.19793090969324112,
      "num_tokens": 65655019.0,
      "step": 37845
    },
    {
      "entropy": 5.813972520828247,
      "epoch": 2.944874538027621,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041348063281299995,
      "loss": 5.0831,
      "mean_token_accuracy": 0.20372422337532042,
      "num_tokens": 65663721.0,
      "step": 37850
    },
    {
      "entropy": 5.78133487701416,
      "epoch": 2.945263567399339,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041345878742494346,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2149618610739708,
      "num_tokens": 65671546.0,
      "step": 37855
    },
    {
      "entropy": 5.835765361785889,
      "epoch": 2.945652596771056,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041343693993599535,
      "loss": 5.0466,
      "mean_token_accuracy": 0.19850541204214095,
      "num_tokens": 65680630.0,
      "step": 37860
    },
    {
      "entropy": 5.817136478424072,
      "epoch": 2.9460416261427738,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004134150903464869,
      "loss": 5.0554,
      "mean_token_accuracy": 0.19645422995090484,
      "num_tokens": 65688579.0,
      "step": 37865
    },
    {
      "entropy": 5.759113836288452,
      "epoch": 2.9464306555144915,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004133932386567498,
      "loss": 4.953,
      "mean_token_accuracy": 0.19637044817209243,
      "num_tokens": 65696531.0,
      "step": 37870
    },
    {
      "entropy": 5.7489142417907715,
      "epoch": 2.9468196848862087,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004133713848671156,
      "loss": 4.9063,
      "mean_token_accuracy": 0.2204396516084671,
      "num_tokens": 65705271.0,
      "step": 37875
    },
    {
      "entropy": 5.762833118438721,
      "epoch": 2.9472087142579264,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004133495289779159,
      "loss": 5.0087,
      "mean_token_accuracy": 0.2050588145852089,
      "num_tokens": 65713638.0,
      "step": 37880
    },
    {
      "entropy": 5.864718198776245,
      "epoch": 2.947597743629644,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004133276709894823,
      "loss": 5.0792,
      "mean_token_accuracy": 0.20197757184505463,
      "num_tokens": 65722339.0,
      "step": 37885
    },
    {
      "entropy": 5.895796012878418,
      "epoch": 2.947986773001362,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004133058109021464,
      "loss": 5.142,
      "mean_token_accuracy": 0.20273630917072297,
      "num_tokens": 65730969.0,
      "step": 37890
    },
    {
      "entropy": 5.942363166809082,
      "epoch": 2.948375802373079,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041328394871624003,
      "loss": 5.0539,
      "mean_token_accuracy": 0.20344147682189942,
      "num_tokens": 65739196.0,
      "step": 37895
    },
    {
      "entropy": 5.9060554027557375,
      "epoch": 2.9487648317447968,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004132620844320948,
      "loss": 5.1157,
      "mean_token_accuracy": 0.19772998243570328,
      "num_tokens": 65748032.0,
      "step": 37900
    },
    {
      "entropy": 5.867882299423218,
      "epoch": 2.9491538611165145,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004132402180500425,
      "loss": 5.0247,
      "mean_token_accuracy": 0.19747343063354492,
      "num_tokens": 65756149.0,
      "step": 37905
    },
    {
      "entropy": 5.7582435131073,
      "epoch": 2.9495428904882317,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004132183495704149,
      "loss": 4.9328,
      "mean_token_accuracy": 0.2111212953925133,
      "num_tokens": 65765275.0,
      "step": 37910
    },
    {
      "entropy": 5.781107091903687,
      "epoch": 2.9499319198599494,
      "grad_norm": 1.25,
      "learning_rate": 0.0004131964789935439,
      "loss": 4.985,
      "mean_token_accuracy": 0.20437317490577697,
      "num_tokens": 65773517.0,
      "step": 37915
    },
    {
      "entropy": 5.803519010543823,
      "epoch": 2.950320949231667,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004131746063197612,
      "loss": 5.0588,
      "mean_token_accuracy": 0.20204575806856157,
      "num_tokens": 65782728.0,
      "step": 37920
    },
    {
      "entropy": 5.8230376720428465,
      "epoch": 2.950709978603385,
      "grad_norm": 1.25,
      "learning_rate": 0.0004131527315493989,
      "loss": 5.0154,
      "mean_token_accuracy": 0.20229255855083467,
      "num_tokens": 65791460.0,
      "step": 37925
    },
    {
      "entropy": 5.867028141021729,
      "epoch": 2.951099007975102,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041313085468278873,
      "loss": 5.0374,
      "mean_token_accuracy": 0.20351543724536897,
      "num_tokens": 65799760.0,
      "step": 37930
    },
    {
      "entropy": 5.879566383361817,
      "epoch": 2.9514880373468197,
      "grad_norm": 1.25,
      "learning_rate": 0.0004131089757202627,
      "loss": 5.0958,
      "mean_token_accuracy": 0.19996506422758104,
      "num_tokens": 65808675.0,
      "step": 37935
    },
    {
      "entropy": 5.881805896759033,
      "epoch": 2.951877066718537,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041308709466215276,
      "loss": 5.063,
      "mean_token_accuracy": 0.20561113506555556,
      "num_tokens": 65817372.0,
      "step": 37940
    },
    {
      "entropy": 5.771470546722412,
      "epoch": 2.9522660960902547,
      "grad_norm": 1.3125,
      "learning_rate": 0.000413065211508791,
      "loss": 5.0103,
      "mean_token_accuracy": 0.20785278379917144,
      "num_tokens": 65825899.0,
      "step": 37945
    },
    {
      "entropy": 5.750229597091675,
      "epoch": 2.9526551254619724,
      "grad_norm": 1.375,
      "learning_rate": 0.00041304332626050937,
      "loss": 4.9616,
      "mean_token_accuracy": 0.21219722628593446,
      "num_tokens": 65834357.0,
      "step": 37950
    },
    {
      "entropy": 5.766220951080323,
      "epoch": 2.95304415483369,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041302143891764,
      "loss": 4.9268,
      "mean_token_accuracy": 0.21969254016876222,
      "num_tokens": 65842869.0,
      "step": 37955
    },
    {
      "entropy": 5.749959802627563,
      "epoch": 2.9534331842054073,
      "grad_norm": 1.265625,
      "learning_rate": 0.000412999549480515,
      "loss": 4.8606,
      "mean_token_accuracy": 0.21479169130325318,
      "num_tokens": 65851378.0,
      "step": 37960
    },
    {
      "entropy": 5.812968921661377,
      "epoch": 2.953822213577125,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004129776579494665,
      "loss": 4.9445,
      "mean_token_accuracy": 0.20802685469388962,
      "num_tokens": 65860209.0,
      "step": 37965
    },
    {
      "entropy": 5.805966234207153,
      "epoch": 2.9542112429488427,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004129557643248265,
      "loss": 5.0459,
      "mean_token_accuracy": 0.2047601193189621,
      "num_tokens": 65868313.0,
      "step": 37970
    },
    {
      "entropy": 5.762943601608276,
      "epoch": 2.95460027232056,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004129338686069276,
      "loss": 4.942,
      "mean_token_accuracy": 0.21552699357271193,
      "num_tokens": 65876049.0,
      "step": 37975
    },
    {
      "entropy": 5.848810338973999,
      "epoch": 2.9549893016922777,
      "grad_norm": 1.375,
      "learning_rate": 0.0004129119707961016,
      "loss": 5.0595,
      "mean_token_accuracy": 0.20347446501255034,
      "num_tokens": 65884042.0,
      "step": 37980
    },
    {
      "entropy": 5.823304271697998,
      "epoch": 2.9553783310639954,
      "grad_norm": 1.296875,
      "learning_rate": 0.000412890070892681,
      "loss": 5.0243,
      "mean_token_accuracy": 0.20991072207689285,
      "num_tokens": 65892303.0,
      "step": 37985
    },
    {
      "entropy": 5.78401141166687,
      "epoch": 2.955767360435713,
      "grad_norm": 1.34375,
      "learning_rate": 0.000412868168896998,
      "loss": 5.0475,
      "mean_token_accuracy": 0.193398879468441,
      "num_tokens": 65900515.0,
      "step": 37990
    },
    {
      "entropy": 5.772557735443115,
      "epoch": 2.9561563898074303,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004128462648093851,
      "loss": 4.9913,
      "mean_token_accuracy": 0.20448630154132844,
      "num_tokens": 65908609.0,
      "step": 37995
    },
    {
      "entropy": 5.845780372619629,
      "epoch": 2.956545419179148,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004128243586301745,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21213323026895523,
      "num_tokens": 65917803.0,
      "step": 38000
    },
    {
      "entropy": 5.786803960800171,
      "epoch": 2.9569344485508657,
      "grad_norm": 1.25,
      "learning_rate": 0.0004128024503596986,
      "loss": 4.955,
      "mean_token_accuracy": 0.20539263188838958,
      "num_tokens": 65926269.0,
      "step": 38005
    },
    {
      "entropy": 5.756941890716552,
      "epoch": 2.957323477922583,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041278053999828986,
      "loss": 4.9418,
      "mean_token_accuracy": 0.2079220175743103,
      "num_tokens": 65934826.0,
      "step": 38010
    },
    {
      "entropy": 5.82415452003479,
      "epoch": 2.9577125072943007,
      "grad_norm": 1.25,
      "learning_rate": 0.0004127586275462806,
      "loss": 5.0052,
      "mean_token_accuracy": 0.20489151179790496,
      "num_tokens": 65943609.0,
      "step": 38015
    },
    {
      "entropy": 5.710185813903808,
      "epoch": 2.9581015366660184,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041273671300400363,
      "loss": 4.9111,
      "mean_token_accuracy": 0.21219113618135452,
      "num_tokens": 65952244.0,
      "step": 38020
    },
    {
      "entropy": 5.750405883789062,
      "epoch": 2.958490566037736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041271479637179116,
      "loss": 4.9395,
      "mean_token_accuracy": 0.2090214341878891,
      "num_tokens": 65960915.0,
      "step": 38025
    },
    {
      "entropy": 5.757425165176391,
      "epoch": 2.9588795954094533,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004126928776499759,
      "loss": 4.9372,
      "mean_token_accuracy": 0.21206235885620117,
      "num_tokens": 65969375.0,
      "step": 38030
    },
    {
      "entropy": 5.813358449935913,
      "epoch": 2.959268624781171,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004126709568388903,
      "loss": 4.995,
      "mean_token_accuracy": 0.21130070835351944,
      "num_tokens": 65978724.0,
      "step": 38035
    },
    {
      "entropy": 5.851937484741211,
      "epoch": 2.9596576541528883,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004126490339388672,
      "loss": 5.0389,
      "mean_token_accuracy": 0.20050172358751298,
      "num_tokens": 65987255.0,
      "step": 38040
    },
    {
      "entropy": 5.840566682815552,
      "epoch": 2.960046683524606,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000412627108950239,
      "loss": 5.1062,
      "mean_token_accuracy": 0.20407545566558838,
      "num_tokens": 65996193.0,
      "step": 38045
    },
    {
      "entropy": 5.858949279785156,
      "epoch": 2.9604357128963237,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004126051818733385,
      "loss": 5.0857,
      "mean_token_accuracy": 0.20551228821277617,
      "num_tokens": 66005081.0,
      "step": 38050
    },
    {
      "entropy": 5.811345338821411,
      "epoch": 2.9608247422680414,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004125832527084984,
      "loss": 5.0599,
      "mean_token_accuracy": 0.20373188108205795,
      "num_tokens": 66014389.0,
      "step": 38055
    },
    {
      "entropy": 5.8832982063293455,
      "epoch": 2.961213771639759,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041256132145605143,
      "loss": 5.1016,
      "mean_token_accuracy": 0.20053936839103698,
      "num_tokens": 66022486.0,
      "step": 38060
    },
    {
      "entropy": 5.805521774291992,
      "epoch": 2.9616028010114763,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004125393881163304,
      "loss": 4.9648,
      "mean_token_accuracy": 0.2047223448753357,
      "num_tokens": 66031058.0,
      "step": 38065
    },
    {
      "entropy": 5.756514596939087,
      "epoch": 2.961991830383194,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041251745268966797,
      "loss": 4.9243,
      "mean_token_accuracy": 0.20539276748895646,
      "num_tokens": 66038705.0,
      "step": 38070
    },
    {
      "entropy": 5.844672298431396,
      "epoch": 2.9623808597549113,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004124955151763971,
      "loss": 4.9969,
      "mean_token_accuracy": 0.20948389023542405,
      "num_tokens": 66047231.0,
      "step": 38075
    },
    {
      "entropy": 5.81997504234314,
      "epoch": 2.962769889126629,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004124735755768506,
      "loss": 5.0342,
      "mean_token_accuracy": 0.20159341990947724,
      "num_tokens": 66055536.0,
      "step": 38080
    },
    {
      "entropy": 5.8472301959991455,
      "epoch": 2.9631589184983467,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004124516338913615,
      "loss": 5.0622,
      "mean_token_accuracy": 0.19506220668554305,
      "num_tokens": 66065215.0,
      "step": 38085
    },
    {
      "entropy": 5.860423803329468,
      "epoch": 2.9635479478700644,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004124296901202626,
      "loss": 5.0586,
      "mean_token_accuracy": 0.2044026583433151,
      "num_tokens": 66074455.0,
      "step": 38090
    },
    {
      "entropy": 5.875318908691407,
      "epoch": 2.9639369772417816,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041240774426388685,
      "loss": 5.0208,
      "mean_token_accuracy": 0.19916004538536072,
      "num_tokens": 66082793.0,
      "step": 38095
    },
    {
      "entropy": 5.78303713798523,
      "epoch": 2.9643260066134993,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004123857963225673,
      "loss": 4.9712,
      "mean_token_accuracy": 0.21020321547985077,
      "num_tokens": 66091503.0,
      "step": 38100
    },
    {
      "entropy": 5.815596055984497,
      "epoch": 2.964715035985217,
      "grad_norm": 1.296875,
      "learning_rate": 0.000412363846296637,
      "loss": 5.0272,
      "mean_token_accuracy": 0.20916155576705933,
      "num_tokens": 66099210.0,
      "step": 38105
    },
    {
      "entropy": 5.754657030105591,
      "epoch": 2.9651040653569343,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000412341894186429,
      "loss": 4.9989,
      "mean_token_accuracy": 0.20170512795448303,
      "num_tokens": 66107338.0,
      "step": 38110
    },
    {
      "entropy": 5.784589672088623,
      "epoch": 2.965493094728652,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004123199399922763,
      "loss": 5.0409,
      "mean_token_accuracy": 0.20100805014371873,
      "num_tokens": 66115962.0,
      "step": 38115
    },
    {
      "entropy": 5.765013694763184,
      "epoch": 2.9658821241003697,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041229798371451203,
      "loss": 4.9931,
      "mean_token_accuracy": 0.2025275394320488,
      "num_tokens": 66123849.0,
      "step": 38120
    },
    {
      "entropy": 5.855650949478149,
      "epoch": 2.9662711534720874,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004122760253534695,
      "loss": 5.0304,
      "mean_token_accuracy": 0.2008976772427559,
      "num_tokens": 66132791.0,
      "step": 38125
    },
    {
      "entropy": 5.887872648239136,
      "epoch": 2.9666601828438046,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004122540649094817,
      "loss": 5.1165,
      "mean_token_accuracy": 0.20156443268060684,
      "num_tokens": 66141434.0,
      "step": 38130
    },
    {
      "entropy": 5.8039475440979,
      "epoch": 2.9670492122155223,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041223210238288196,
      "loss": 5.0245,
      "mean_token_accuracy": 0.2071179449558258,
      "num_tokens": 66149941.0,
      "step": 38135
    },
    {
      "entropy": 5.8704945087432865,
      "epoch": 2.9674382415872396,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004122101377740035,
      "loss": 5.0328,
      "mean_token_accuracy": 0.20251037925481796,
      "num_tokens": 66159411.0,
      "step": 38140
    },
    {
      "entropy": 5.815524482727051,
      "epoch": 2.9678272709589573,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004121881710831796,
      "loss": 4.974,
      "mean_token_accuracy": 0.21091575622558595,
      "num_tokens": 66168504.0,
      "step": 38145
    },
    {
      "entropy": 5.851507043838501,
      "epoch": 2.968216300330675,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004121662023107435,
      "loss": 5.0424,
      "mean_token_accuracy": 0.20674943625926973,
      "num_tokens": 66176777.0,
      "step": 38150
    },
    {
      "entropy": 5.806068563461304,
      "epoch": 2.9686053297023927,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004121442314570286,
      "loss": 4.9709,
      "mean_token_accuracy": 0.2041332095861435,
      "num_tokens": 66185028.0,
      "step": 38155
    },
    {
      "entropy": 5.830725765228271,
      "epoch": 2.9689943590741104,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041212225852236834,
      "loss": 4.974,
      "mean_token_accuracy": 0.21149863451719284,
      "num_tokens": 66193139.0,
      "step": 38160
    },
    {
      "entropy": 5.835852527618409,
      "epoch": 2.9693833884458276,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004121002835070961,
      "loss": 5.067,
      "mean_token_accuracy": 0.2000282034277916,
      "num_tokens": 66202196.0,
      "step": 38165
    },
    {
      "entropy": 5.808925914764404,
      "epoch": 2.9697724178175453,
      "grad_norm": 1.25,
      "learning_rate": 0.0004120783064115452,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20315721035003662,
      "num_tokens": 66209889.0,
      "step": 38170
    },
    {
      "entropy": 5.7886124610900875,
      "epoch": 2.9701614471892626,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004120563272360492,
      "loss": 4.9887,
      "mean_token_accuracy": 0.2004638135433197,
      "num_tokens": 66218889.0,
      "step": 38175
    },
    {
      "entropy": 5.870897579193115,
      "epoch": 2.9705504765609803,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004120343459809416,
      "loss": 5.0797,
      "mean_token_accuracy": 0.20324015319347383,
      "num_tokens": 66227293.0,
      "step": 38180
    },
    {
      "entropy": 5.876148748397827,
      "epoch": 2.970939505932698,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004120123626465559,
      "loss": 5.0273,
      "mean_token_accuracy": 0.2086865335702896,
      "num_tokens": 66235322.0,
      "step": 38185
    },
    {
      "entropy": 5.8473135471344,
      "epoch": 2.9713285353044157,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041199037723322566,
      "loss": 5.0474,
      "mean_token_accuracy": 0.1990977257490158,
      "num_tokens": 66244776.0,
      "step": 38190
    },
    {
      "entropy": 5.7889721393585205,
      "epoch": 2.971717564676133,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041196838974128446,
      "loss": 4.9932,
      "mean_token_accuracy": 0.2035525068640709,
      "num_tokens": 66254058.0,
      "step": 38195
    },
    {
      "entropy": 5.8379798412323,
      "epoch": 2.9721065940478506,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000411946400171066,
      "loss": 5.0969,
      "mean_token_accuracy": 0.20382908582687378,
      "num_tokens": 66262816.0,
      "step": 38200
    },
    {
      "entropy": 5.9071673393249515,
      "epoch": 2.9724956234195683,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041192440852290385,
      "loss": 5.0859,
      "mean_token_accuracy": 0.19741087257862092,
      "num_tokens": 66271808.0,
      "step": 38205
    },
    {
      "entropy": 5.859216165542603,
      "epoch": 2.9728846527912856,
      "grad_norm": 1.25,
      "learning_rate": 0.0004119024147971318,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20637826323509217,
      "num_tokens": 66280737.0,
      "step": 38210
    },
    {
      "entropy": 5.801422786712647,
      "epoch": 2.9732736821630033,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041188041899408345,
      "loss": 4.9939,
      "mean_token_accuracy": 0.2070898547768593,
      "num_tokens": 66289970.0,
      "step": 38215
    },
    {
      "entropy": 5.857291555404663,
      "epoch": 2.973662711534721,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004118584211140926,
      "loss": 5.061,
      "mean_token_accuracy": 0.20229993909597396,
      "num_tokens": 66297917.0,
      "step": 38220
    },
    {
      "entropy": 5.77235517501831,
      "epoch": 2.9740517409064386,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041183642115749316,
      "loss": 4.9906,
      "mean_token_accuracy": 0.20838579088449477,
      "num_tokens": 66306094.0,
      "step": 38225
    },
    {
      "entropy": 5.79345474243164,
      "epoch": 2.974440770278156,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041181441912461873,
      "loss": 4.9606,
      "mean_token_accuracy": 0.21303750574588776,
      "num_tokens": 66315002.0,
      "step": 38230
    },
    {
      "entropy": 5.851683044433594,
      "epoch": 2.9748297996498736,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004117924150158032,
      "loss": 4.976,
      "mean_token_accuracy": 0.20883897542953492,
      "num_tokens": 66322720.0,
      "step": 38235
    },
    {
      "entropy": 5.76937952041626,
      "epoch": 2.9752188290215913,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041177040883138064,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21242665499448776,
      "num_tokens": 66331478.0,
      "step": 38240
    },
    {
      "entropy": 5.882160758972168,
      "epoch": 2.9756078583933085,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041174840057168474,
      "loss": 5.0637,
      "mean_token_accuracy": 0.20337686091661453,
      "num_tokens": 66339683.0,
      "step": 38245
    },
    {
      "entropy": 5.844421768188477,
      "epoch": 2.9759968877650262,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004117263902370495,
      "loss": 5.0241,
      "mean_token_accuracy": 0.20622842162847518,
      "num_tokens": 66347944.0,
      "step": 38250
    },
    {
      "entropy": 5.829503774642944,
      "epoch": 2.976385917136744,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041170437782780896,
      "loss": 5.1066,
      "mean_token_accuracy": 0.20311209857463836,
      "num_tokens": 66356947.0,
      "step": 38255
    },
    {
      "entropy": 5.752525711059571,
      "epoch": 2.9767749465084616,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000411682363344297,
      "loss": 4.8722,
      "mean_token_accuracy": 0.20997137427330018,
      "num_tokens": 66365843.0,
      "step": 38260
    },
    {
      "entropy": 5.829859209060669,
      "epoch": 2.977163975880179,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041166034678684777,
      "loss": 5.116,
      "mean_token_accuracy": 0.19459507316350938,
      "num_tokens": 66375305.0,
      "step": 38265
    },
    {
      "entropy": 5.799845266342163,
      "epoch": 2.9775530052518966,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004116383281557953,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20652880370616913,
      "num_tokens": 66383758.0,
      "step": 38270
    },
    {
      "entropy": 5.772051429748535,
      "epoch": 2.977942034623614,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041161630745147374,
      "loss": 4.9885,
      "mean_token_accuracy": 0.2079613447189331,
      "num_tokens": 66393538.0,
      "step": 38275
    },
    {
      "entropy": 5.8443769931793215,
      "epoch": 2.9783310639953315,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004115942846742171,
      "loss": 5.0321,
      "mean_token_accuracy": 0.20438508242368697,
      "num_tokens": 66402395.0,
      "step": 38280
    },
    {
      "entropy": 5.82994589805603,
      "epoch": 2.9787200933670492,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004115722598243596,
      "loss": 4.9693,
      "mean_token_accuracy": 0.2011784255504608,
      "num_tokens": 66410671.0,
      "step": 38285
    },
    {
      "entropy": 5.852193307876587,
      "epoch": 2.979109122738767,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041155023290223537,
      "loss": 5.0507,
      "mean_token_accuracy": 0.20262469053268434,
      "num_tokens": 66419370.0,
      "step": 38290
    },
    {
      "entropy": 5.764391994476318,
      "epoch": 2.979498152110484,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004115282039081788,
      "loss": 4.9688,
      "mean_token_accuracy": 0.21316221058368684,
      "num_tokens": 66427816.0,
      "step": 38295
    },
    {
      "entropy": 5.791408252716065,
      "epoch": 2.979887181482202,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041150617284252405,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20855241864919663,
      "num_tokens": 66436371.0,
      "step": 38300
    },
    {
      "entropy": 5.86045355796814,
      "epoch": 2.9802762108539196,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004114841397056053,
      "loss": 5.0571,
      "mean_token_accuracy": 0.20817727893590926,
      "num_tokens": 66445158.0,
      "step": 38305
    },
    {
      "entropy": 5.860354995727539,
      "epoch": 2.980665240225637,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041146210449775704,
      "loss": 5.0234,
      "mean_token_accuracy": 0.20724112838506697,
      "num_tokens": 66454133.0,
      "step": 38310
    },
    {
      "entropy": 5.745707035064697,
      "epoch": 2.9810542695973545,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004114400672193136,
      "loss": 5.0296,
      "mean_token_accuracy": 0.20704961866140364,
      "num_tokens": 66462698.0,
      "step": 38315
    },
    {
      "entropy": 5.807286214828491,
      "epoch": 2.9814432989690722,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041141802787060923,
      "loss": 5.0322,
      "mean_token_accuracy": 0.205136039853096,
      "num_tokens": 66473145.0,
      "step": 38320
    },
    {
      "entropy": 5.860437679290771,
      "epoch": 2.98183232834079,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004113959864519784,
      "loss": 5.0105,
      "mean_token_accuracy": 0.2037022516131401,
      "num_tokens": 66482481.0,
      "step": 38325
    },
    {
      "entropy": 5.871683597564697,
      "epoch": 2.982221357712507,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004113739429637556,
      "loss": 4.9961,
      "mean_token_accuracy": 0.2225971907377243,
      "num_tokens": 66491338.0,
      "step": 38330
    },
    {
      "entropy": 5.804482269287109,
      "epoch": 2.982610387084225,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004113518974062753,
      "loss": 5.0772,
      "mean_token_accuracy": 0.21179619133472444,
      "num_tokens": 66500517.0,
      "step": 38335
    },
    {
      "entropy": 5.838915872573852,
      "epoch": 2.9829994164559426,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000411329849779872,
      "loss": 4.9862,
      "mean_token_accuracy": 0.20882016867399217,
      "num_tokens": 66509431.0,
      "step": 38340
    },
    {
      "entropy": 5.803640604019165,
      "epoch": 2.98338844582766,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041130780008488015,
      "loss": 5.0064,
      "mean_token_accuracy": 0.21008384376764297,
      "num_tokens": 66518111.0,
      "step": 38345
    },
    {
      "entropy": 5.773294830322266,
      "epoch": 2.9837774751993775,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004112857483216344,
      "loss": 4.9579,
      "mean_token_accuracy": 0.20589220374822617,
      "num_tokens": 66527325.0,
      "step": 38350
    },
    {
      "entropy": 5.807560014724731,
      "epoch": 2.9841665045710952,
      "grad_norm": 1.25,
      "learning_rate": 0.00041126369449046935,
      "loss": 4.9727,
      "mean_token_accuracy": 0.20548409968614578,
      "num_tokens": 66536542.0,
      "step": 38355
    },
    {
      "entropy": 5.851216459274292,
      "epoch": 2.984555533942813,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041124163859171967,
      "loss": 5.0064,
      "mean_token_accuracy": 0.20158574283123015,
      "num_tokens": 66544724.0,
      "step": 38360
    },
    {
      "entropy": 5.853804016113282,
      "epoch": 2.98494456331453,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004112195806257199,
      "loss": 5.0812,
      "mean_token_accuracy": 0.20403656959533692,
      "num_tokens": 66553679.0,
      "step": 38365
    },
    {
      "entropy": 5.772175598144531,
      "epoch": 2.985333592686248,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004111975205928049,
      "loss": 4.9382,
      "mean_token_accuracy": 0.20662160515785216,
      "num_tokens": 66562593.0,
      "step": 38370
    },
    {
      "entropy": 5.835263872146607,
      "epoch": 2.985722622057965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041117545849330924,
      "loss": 4.9775,
      "mean_token_accuracy": 0.20973959416151047,
      "num_tokens": 66570947.0,
      "step": 38375
    },
    {
      "entropy": 5.802307748794556,
      "epoch": 2.986111651429683,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041115339432756776,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2136720061302185,
      "num_tokens": 66579733.0,
      "step": 38380
    },
    {
      "entropy": 5.819054460525512,
      "epoch": 2.9865006808014005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004111313280959152,
      "loss": 4.994,
      "mean_token_accuracy": 0.20773123502731322,
      "num_tokens": 66589252.0,
      "step": 38385
    },
    {
      "entropy": 5.759513807296753,
      "epoch": 2.986889710173118,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041110925979868647,
      "loss": 4.9687,
      "mean_token_accuracy": 0.21514253169298173,
      "num_tokens": 66597817.0,
      "step": 38390
    },
    {
      "entropy": 5.766794300079345,
      "epoch": 2.9872787395448355,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004110871894362163,
      "loss": 4.9623,
      "mean_token_accuracy": 0.2122700557112694,
      "num_tokens": 66606154.0,
      "step": 38395
    },
    {
      "entropy": 5.837796926498413,
      "epoch": 2.987667768916553,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041106511700883975,
      "loss": 4.9885,
      "mean_token_accuracy": 0.21077273190021514,
      "num_tokens": 66614799.0,
      "step": 38400
    },
    {
      "entropy": 5.815525674819947,
      "epoch": 2.988056798288271,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041104304251689147,
      "loss": 4.9572,
      "mean_token_accuracy": 0.2093173399567604,
      "num_tokens": 66623250.0,
      "step": 38405
    },
    {
      "entropy": 5.722867202758789,
      "epoch": 2.988445827659988,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041102096596070666,
      "loss": 4.8922,
      "mean_token_accuracy": 0.21746176928281785,
      "num_tokens": 66631948.0,
      "step": 38410
    },
    {
      "entropy": 5.762844038009644,
      "epoch": 2.988834857031706,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041099888734062013,
      "loss": 4.9443,
      "mean_token_accuracy": 0.20811472982168197,
      "num_tokens": 66639864.0,
      "step": 38415
    },
    {
      "entropy": 5.8093348979949955,
      "epoch": 2.9892238864034235,
      "grad_norm": 1.34375,
      "learning_rate": 0.000410976806656967,
      "loss": 4.9569,
      "mean_token_accuracy": 0.208726267516613,
      "num_tokens": 66647758.0,
      "step": 38420
    },
    {
      "entropy": 5.847859668731689,
      "epoch": 2.989612915775141,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004109547239100822,
      "loss": 4.9671,
      "mean_token_accuracy": 0.21477610021829605,
      "num_tokens": 66655490.0,
      "step": 38425
    },
    {
      "entropy": 5.723924016952514,
      "epoch": 2.9900019451468585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004109326391003009,
      "loss": 4.949,
      "mean_token_accuracy": 0.21224759817123412,
      "num_tokens": 66664264.0,
      "step": 38430
    },
    {
      "entropy": 5.792990446090698,
      "epoch": 2.990390974518576,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041091055222795814,
      "loss": 4.9338,
      "mean_token_accuracy": 0.20295320600271224,
      "num_tokens": 66672595.0,
      "step": 38435
    },
    {
      "entropy": 5.829610538482666,
      "epoch": 2.990780003890294,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004108884632933891,
      "loss": 5.0146,
      "mean_token_accuracy": 0.20680088847875594,
      "num_tokens": 66680494.0,
      "step": 38440
    },
    {
      "entropy": 5.886165189743042,
      "epoch": 2.991169033262011,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041086637229692893,
      "loss": 5.0846,
      "mean_token_accuracy": 0.19897050708532332,
      "num_tokens": 66689462.0,
      "step": 38445
    },
    {
      "entropy": 5.858578491210937,
      "epoch": 2.991558062633729,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041084427923891275,
      "loss": 5.0002,
      "mean_token_accuracy": 0.20660516619682312,
      "num_tokens": 66697123.0,
      "step": 38450
    },
    {
      "entropy": 5.845050096511841,
      "epoch": 2.9919470920054465,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041082218411967586,
      "loss": 5.1091,
      "mean_token_accuracy": 0.19351996183395387,
      "num_tokens": 66705889.0,
      "step": 38455
    },
    {
      "entropy": 5.81690411567688,
      "epoch": 2.992336121377164,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041080008693955355,
      "loss": 5.0132,
      "mean_token_accuracy": 0.2114890143275261,
      "num_tokens": 66713832.0,
      "step": 38460
    },
    {
      "entropy": 5.769218635559082,
      "epoch": 2.9927251507488815,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004107779876988811,
      "loss": 4.9725,
      "mean_token_accuracy": 0.21637472361326218,
      "num_tokens": 66722594.0,
      "step": 38465
    },
    {
      "entropy": 5.83896837234497,
      "epoch": 2.993114180120599,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041075588639799375,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21047940105199814,
      "num_tokens": 66731125.0,
      "step": 38470
    },
    {
      "entropy": 5.803531837463379,
      "epoch": 2.9935032094923164,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004107337830372269,
      "loss": 5.1221,
      "mean_token_accuracy": 0.20163467675447463,
      "num_tokens": 66739234.0,
      "step": 38475
    },
    {
      "entropy": 5.790473461151123,
      "epoch": 2.993892238864034,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000410711677616916,
      "loss": 5.0056,
      "mean_token_accuracy": 0.2034662425518036,
      "num_tokens": 66747630.0,
      "step": 38480
    },
    {
      "entropy": 5.867070484161377,
      "epoch": 2.994281268235752,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004106895701373963,
      "loss": 5.0363,
      "mean_token_accuracy": 0.20204291194677354,
      "num_tokens": 66755545.0,
      "step": 38485
    },
    {
      "entropy": 5.844150257110596,
      "epoch": 2.9946702976074695,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041066746059900336,
      "loss": 5.021,
      "mean_token_accuracy": 0.2079653948545456,
      "num_tokens": 66764336.0,
      "step": 38490
    },
    {
      "entropy": 5.823588895797729,
      "epoch": 2.995059326979187,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004106453490020727,
      "loss": 5.0367,
      "mean_token_accuracy": 0.20857934802770614,
      "num_tokens": 66773045.0,
      "step": 38495
    },
    {
      "entropy": 5.807673120498658,
      "epoch": 2.9954483563509045,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00041062323534693973,
      "loss": 4.9703,
      "mean_token_accuracy": 0.21049587279558182,
      "num_tokens": 66781228.0,
      "step": 38500
    },
    {
      "entropy": 5.892860460281372,
      "epoch": 2.995837385722622,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041060111963394004,
      "loss": 5.0969,
      "mean_token_accuracy": 0.20815753936767578,
      "num_tokens": 66789829.0,
      "step": 38505
    },
    {
      "entropy": 5.860504007339477,
      "epoch": 2.9962264150943394,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041057900186340917,
      "loss": 4.9574,
      "mean_token_accuracy": 0.21139210611581802,
      "num_tokens": 66798023.0,
      "step": 38510
    },
    {
      "entropy": 5.831308031082154,
      "epoch": 2.996615444466057,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041055688203568274,
      "loss": 4.9775,
      "mean_token_accuracy": 0.21080498695373534,
      "num_tokens": 66806536.0,
      "step": 38515
    },
    {
      "entropy": 5.813809633255005,
      "epoch": 2.997004473837775,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004105347601510964,
      "loss": 4.9397,
      "mean_token_accuracy": 0.20484142303466796,
      "num_tokens": 66815698.0,
      "step": 38520
    },
    {
      "entropy": 5.794840478897095,
      "epoch": 2.9973935032094925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004105126362099859,
      "loss": 4.9797,
      "mean_token_accuracy": 0.2038421154022217,
      "num_tokens": 66824358.0,
      "step": 38525
    },
    {
      "entropy": 5.813934755325318,
      "epoch": 2.9977825325812097,
      "grad_norm": 1.25,
      "learning_rate": 0.0004104905102126867,
      "loss": 4.9827,
      "mean_token_accuracy": 0.2109039232134819,
      "num_tokens": 66832895.0,
      "step": 38530
    },
    {
      "entropy": 5.79021725654602,
      "epoch": 2.9981715619529274,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004104683821595347,
      "loss": 4.9361,
      "mean_token_accuracy": 0.20526452958583832,
      "num_tokens": 66842289.0,
      "step": 38535
    },
    {
      "entropy": 5.790495872497559,
      "epoch": 2.998560591324645,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041044625205086566,
      "loss": 4.9731,
      "mean_token_accuracy": 0.20876385271549225,
      "num_tokens": 66850656.0,
      "step": 38540
    },
    {
      "entropy": 5.77311897277832,
      "epoch": 2.9989496206963624,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041042411988701535,
      "loss": 4.9073,
      "mean_token_accuracy": 0.21063147634267806,
      "num_tokens": 66859518.0,
      "step": 38545
    },
    {
      "entropy": 5.742673683166504,
      "epoch": 2.99933865006808,
      "grad_norm": 1.25,
      "learning_rate": 0.00041040198566831944,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20628873705863954,
      "num_tokens": 66868431.0,
      "step": 38550
    },
    {
      "entropy": 5.813263177871704,
      "epoch": 2.999727679439798,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004103798493951141,
      "loss": 5.0071,
      "mean_token_accuracy": 0.2079155772924423,
      "num_tokens": 66877255.0,
      "step": 38555
    },
    {
      "entropy": 5.8732954661051435,
      "epoch": 3.0000778058743434,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041035771106773483,
      "loss": 4.9564,
      "mean_token_accuracy": 0.20959726803832585,
      "num_tokens": 66884881.0,
      "step": 38560
    },
    {
      "entropy": 5.820796775817871,
      "epoch": 3.000466835246061,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004103355706865179,
      "loss": 4.9698,
      "mean_token_accuracy": 0.21446372717618942,
      "num_tokens": 66893369.0,
      "step": 38565
    },
    {
      "entropy": 5.806730651855469,
      "epoch": 3.0008558646177788,
      "grad_norm": 1.296875,
      "learning_rate": 0.000410313428251799,
      "loss": 4.937,
      "mean_token_accuracy": 0.21228551864624023,
      "num_tokens": 66900994.0,
      "step": 38570
    },
    {
      "entropy": 5.770044374465942,
      "epoch": 3.001244893989496,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041029128376391424,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20944125652313234,
      "num_tokens": 66909906.0,
      "step": 38575
    },
    {
      "entropy": 5.80250768661499,
      "epoch": 3.0016339233612137,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041026913722319954,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21875356882810593,
      "num_tokens": 66918712.0,
      "step": 38580
    },
    {
      "entropy": 5.846242761611938,
      "epoch": 3.0020229527329314,
      "grad_norm": 1.265625,
      "learning_rate": 0.000410246988629991,
      "loss": 4.9543,
      "mean_token_accuracy": 0.20791738629341125,
      "num_tokens": 66927462.0,
      "step": 38585
    },
    {
      "entropy": 5.7267364978790285,
      "epoch": 3.002411982104649,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041022483798462466,
      "loss": 4.8682,
      "mean_token_accuracy": 0.21656473278999328,
      "num_tokens": 66935920.0,
      "step": 38590
    },
    {
      "entropy": 5.762704515457154,
      "epoch": 3.0028010114763664,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004102026852874367,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21821916550397874,
      "num_tokens": 66944595.0,
      "step": 38595
    },
    {
      "entropy": 5.861315679550171,
      "epoch": 3.003190040848084,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041018053053876316,
      "loss": 5.12,
      "mean_token_accuracy": 0.2037678211927414,
      "num_tokens": 66954025.0,
      "step": 38600
    },
    {
      "entropy": 5.82306694984436,
      "epoch": 3.0035790702198017,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004101583737389402,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21962472349405288,
      "num_tokens": 66963143.0,
      "step": 38605
    },
    {
      "entropy": 5.8677215576171875,
      "epoch": 3.003968099591519,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041013621488830406,
      "loss": 5.028,
      "mean_token_accuracy": 0.20667544305324553,
      "num_tokens": 66972619.0,
      "step": 38610
    },
    {
      "entropy": 5.78109827041626,
      "epoch": 3.0043571289632367,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041011405398719096,
      "loss": 4.8982,
      "mean_token_accuracy": 0.21780141741037368,
      "num_tokens": 66980749.0,
      "step": 38615
    },
    {
      "entropy": 5.8513952732086185,
      "epoch": 3.0047461583349544,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041009189103593716,
      "loss": 4.9593,
      "mean_token_accuracy": 0.2091781660914421,
      "num_tokens": 66989656.0,
      "step": 38620
    },
    {
      "entropy": 5.806668281555176,
      "epoch": 3.0051351877066717,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004100697260348789,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21583760380744935,
      "num_tokens": 66998308.0,
      "step": 38625
    },
    {
      "entropy": 5.870392227172852,
      "epoch": 3.0055242170783893,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041004755898435254,
      "loss": 4.9918,
      "mean_token_accuracy": 0.2033906638622284,
      "num_tokens": 67007596.0,
      "step": 38630
    },
    {
      "entropy": 5.718611574172973,
      "epoch": 3.005913246450107,
      "grad_norm": 1.25,
      "learning_rate": 0.0004100253898846945,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2201899468898773,
      "num_tokens": 67016137.0,
      "step": 38635
    },
    {
      "entropy": 5.835782814025879,
      "epoch": 3.0063022758218247,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000410003218736241,
      "loss": 5.0721,
      "mean_token_accuracy": 0.20686845481395721,
      "num_tokens": 67024839.0,
      "step": 38640
    },
    {
      "entropy": 5.848434114456177,
      "epoch": 3.006691305193542,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040998104553932856,
      "loss": 4.9515,
      "mean_token_accuracy": 0.2058714136481285,
      "num_tokens": 67033572.0,
      "step": 38645
    },
    {
      "entropy": 5.857256603240967,
      "epoch": 3.0070803345652597,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004099588702942936,
      "loss": 4.9472,
      "mean_token_accuracy": 0.2129300743341446,
      "num_tokens": 67042355.0,
      "step": 38650
    },
    {
      "entropy": 5.781149244308471,
      "epoch": 3.0074693639369774,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040993669300147257,
      "loss": 4.8967,
      "mean_token_accuracy": 0.209248349070549,
      "num_tokens": 67050418.0,
      "step": 38655
    },
    {
      "entropy": 5.81992998123169,
      "epoch": 3.0078583933086946,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040991451366120205,
      "loss": 5.0476,
      "mean_token_accuracy": 0.19984836876392365,
      "num_tokens": 67059846.0,
      "step": 38660
    },
    {
      "entropy": 5.867749738693237,
      "epoch": 3.0082474226804123,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004098923322738185,
      "loss": 4.9948,
      "mean_token_accuracy": 0.2110414147377014,
      "num_tokens": 67068680.0,
      "step": 38665
    },
    {
      "entropy": 5.776160335540771,
      "epoch": 3.00863645205213,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040987014883965843,
      "loss": 4.9027,
      "mean_token_accuracy": 0.21377013027668,
      "num_tokens": 67077607.0,
      "step": 38670
    },
    {
      "entropy": 5.809245538711548,
      "epoch": 3.0090254814238473,
      "grad_norm": 1.25,
      "learning_rate": 0.00040984796335905867,
      "loss": 4.9879,
      "mean_token_accuracy": 0.20853404253721236,
      "num_tokens": 67086553.0,
      "step": 38675
    },
    {
      "entropy": 5.895388746261597,
      "epoch": 3.009414510795565,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004098257758323556,
      "loss": 4.985,
      "mean_token_accuracy": 0.2156752660870552,
      "num_tokens": 67095543.0,
      "step": 38680
    },
    {
      "entropy": 5.808364582061768,
      "epoch": 3.0098035401672827,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004098035862598859,
      "loss": 5.0516,
      "mean_token_accuracy": 0.20129741430282594,
      "num_tokens": 67103831.0,
      "step": 38685
    },
    {
      "entropy": 5.799151420593262,
      "epoch": 3.0101925695390004,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004097813946419865,
      "loss": 4.9459,
      "mean_token_accuracy": 0.20996808111667634,
      "num_tokens": 67112289.0,
      "step": 38690
    },
    {
      "entropy": 5.801869583129883,
      "epoch": 3.0105815989107176,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004097592009789939,
      "loss": 5.0146,
      "mean_token_accuracy": 0.20646202564239502,
      "num_tokens": 67120376.0,
      "step": 38695
    },
    {
      "entropy": 5.863368654251099,
      "epoch": 3.0109706282824353,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040973700527124494,
      "loss": 5.0481,
      "mean_token_accuracy": 0.20137663036584855,
      "num_tokens": 67128412.0,
      "step": 38700
    },
    {
      "entropy": 5.775389862060547,
      "epoch": 3.011359657654153,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040971480751907633,
      "loss": 4.8353,
      "mean_token_accuracy": 0.21419100761413573,
      "num_tokens": 67136212.0,
      "step": 38705
    },
    {
      "entropy": 5.793587398529053,
      "epoch": 3.0117486870258703,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040969260772282494,
      "loss": 5.0356,
      "mean_token_accuracy": 0.2053624778985977,
      "num_tokens": 67145206.0,
      "step": 38710
    },
    {
      "entropy": 5.80336365699768,
      "epoch": 3.012137716397588,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004096704058828277,
      "loss": 4.935,
      "mean_token_accuracy": 0.20755007565021516,
      "num_tokens": 67153760.0,
      "step": 38715
    },
    {
      "entropy": 5.912252902984619,
      "epoch": 3.0125267457693057,
      "grad_norm": 1.375,
      "learning_rate": 0.00040964820199942135,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20961494743824005,
      "num_tokens": 67162574.0,
      "step": 38720
    },
    {
      "entropy": 5.854893636703491,
      "epoch": 3.012915775141023,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004096259960729428,
      "loss": 4.9802,
      "mean_token_accuracy": 0.20706947147846222,
      "num_tokens": 67171329.0,
      "step": 38725
    },
    {
      "entropy": 5.824176216125489,
      "epoch": 3.0133048045127406,
      "grad_norm": 1.25,
      "learning_rate": 0.00040960378810372906,
      "loss": 4.9576,
      "mean_token_accuracy": 0.20258052200078963,
      "num_tokens": 67180243.0,
      "step": 38730
    },
    {
      "entropy": 5.729430198669434,
      "epoch": 3.0136938338844583,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004095815780921171,
      "loss": 4.8215,
      "mean_token_accuracy": 0.22056725472211838,
      "num_tokens": 67189610.0,
      "step": 38735
    },
    {
      "entropy": 5.8282819271087645,
      "epoch": 3.014082863256176,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004095593660384439,
      "loss": 5.0346,
      "mean_token_accuracy": 0.20375740826129912,
      "num_tokens": 67199084.0,
      "step": 38740
    },
    {
      "entropy": 5.807944965362549,
      "epoch": 3.0144718926278933,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004095371519430465,
      "loss": 4.9573,
      "mean_token_accuracy": 0.20990421921014785,
      "num_tokens": 67207705.0,
      "step": 38745
    },
    {
      "entropy": 5.857130813598633,
      "epoch": 3.014860921999611,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040951493580626195,
      "loss": 4.9601,
      "mean_token_accuracy": 0.2063715785741806,
      "num_tokens": 67215576.0,
      "step": 38750
    },
    {
      "entropy": 5.76645073890686,
      "epoch": 3.0152499513713287,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004094927176284273,
      "loss": 4.8771,
      "mean_token_accuracy": 0.20913272947072983,
      "num_tokens": 67223700.0,
      "step": 38755
    },
    {
      "entropy": 5.825154876708984,
      "epoch": 3.015638980743046,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040947049740987977,
      "loss": 4.9835,
      "mean_token_accuracy": 0.21304094791412354,
      "num_tokens": 67231878.0,
      "step": 38760
    },
    {
      "entropy": 5.82283763885498,
      "epoch": 3.0160280101147636,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004094482751509565,
      "loss": 4.919,
      "mean_token_accuracy": 0.20861431509256362,
      "num_tokens": 67240174.0,
      "step": 38765
    },
    {
      "entropy": 5.859723234176636,
      "epoch": 3.0164170394864813,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004094260508519947,
      "loss": 5.0328,
      "mean_token_accuracy": 0.21041592061519623,
      "num_tokens": 67248617.0,
      "step": 38770
    },
    {
      "entropy": 5.878376626968384,
      "epoch": 3.0168060688581986,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040940382451333144,
      "loss": 5.0794,
      "mean_token_accuracy": 0.2061728999018669,
      "num_tokens": 67258581.0,
      "step": 38775
    },
    {
      "entropy": 5.9155456066131595,
      "epoch": 3.0171950982299163,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040938159613530405,
      "loss": 5.0413,
      "mean_token_accuracy": 0.2138422966003418,
      "num_tokens": 67267854.0,
      "step": 38780
    },
    {
      "entropy": 5.848206996917725,
      "epoch": 3.017584127601634,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004093593657182499,
      "loss": 4.9413,
      "mean_token_accuracy": 0.21178372204303741,
      "num_tokens": 67277391.0,
      "step": 38785
    },
    {
      "entropy": 5.791252756118775,
      "epoch": 3.0179731569733517,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004093371332625062,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20981811732053757,
      "num_tokens": 67285460.0,
      "step": 38790
    },
    {
      "entropy": 5.831331729888916,
      "epoch": 3.018362186345069,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004093148987684103,
      "loss": 4.9704,
      "mean_token_accuracy": 0.20694383680820466,
      "num_tokens": 67294023.0,
      "step": 38795
    },
    {
      "entropy": 5.8077614307403564,
      "epoch": 3.0187512157167866,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040929266223629966,
      "loss": 4.8776,
      "mean_token_accuracy": 0.20911927223205568,
      "num_tokens": 67302053.0,
      "step": 38800
    },
    {
      "entropy": 5.826122188568116,
      "epoch": 3.0191402450885043,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040927042366651164,
      "loss": 5.0653,
      "mean_token_accuracy": 0.2019258052110672,
      "num_tokens": 67310478.0,
      "step": 38805
    },
    {
      "entropy": 5.80563702583313,
      "epoch": 3.0195292744602216,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040924818305938354,
      "loss": 4.93,
      "mean_token_accuracy": 0.21595605164766313,
      "num_tokens": 67318096.0,
      "step": 38810
    },
    {
      "entropy": 5.810717153549194,
      "epoch": 3.0199183038319393,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040922594041525303,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2034677043557167,
      "num_tokens": 67326760.0,
      "step": 38815
    },
    {
      "entropy": 5.7640584945678714,
      "epoch": 3.020307333203657,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004092036957344574,
      "loss": 4.9069,
      "mean_token_accuracy": 0.21186792999505996,
      "num_tokens": 67335982.0,
      "step": 38820
    },
    {
      "entropy": 5.772571420669555,
      "epoch": 3.0206963625753747,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040918144901733443,
      "loss": 4.9515,
      "mean_token_accuracy": 0.2084176316857338,
      "num_tokens": 67344666.0,
      "step": 38825
    },
    {
      "entropy": 5.855813503265381,
      "epoch": 3.021085391947092,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004091592002642215,
      "loss": 5.0254,
      "mean_token_accuracy": 0.20823420733213424,
      "num_tokens": 67352607.0,
      "step": 38830
    },
    {
      "entropy": 5.881129264831543,
      "epoch": 3.0214744213188096,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004091369494754562,
      "loss": 5.0041,
      "mean_token_accuracy": 0.20866356641054154,
      "num_tokens": 67361097.0,
      "step": 38835
    },
    {
      "entropy": 5.7675717830657955,
      "epoch": 3.0218634506905273,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004091146966513763,
      "loss": 4.9558,
      "mean_token_accuracy": 0.20188004672527313,
      "num_tokens": 67370229.0,
      "step": 38840
    },
    {
      "entropy": 5.766801023483277,
      "epoch": 3.0222524800622446,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040909244179231926,
      "loss": 4.9636,
      "mean_token_accuracy": 0.21366574466228486,
      "num_tokens": 67379226.0,
      "step": 38845
    },
    {
      "entropy": 5.777174854278565,
      "epoch": 3.0226415094339623,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004090701848986229,
      "loss": 4.8809,
      "mean_token_accuracy": 0.21997925788164138,
      "num_tokens": 67388048.0,
      "step": 38850
    },
    {
      "entropy": 5.814140033721924,
      "epoch": 3.02303053880568,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004090479259706247,
      "loss": 4.829,
      "mean_token_accuracy": 0.21382132321596145,
      "num_tokens": 67396545.0,
      "step": 38855
    },
    {
      "entropy": 5.769040489196778,
      "epoch": 3.023419568177397,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004090256650086628,
      "loss": 4.901,
      "mean_token_accuracy": 0.21297713369131088,
      "num_tokens": 67405128.0,
      "step": 38860
    },
    {
      "entropy": 5.804767513275147,
      "epoch": 3.023808597549115,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040900340201307475,
      "loss": 5.0214,
      "mean_token_accuracy": 0.2125364676117897,
      "num_tokens": 67414231.0,
      "step": 38865
    },
    {
      "entropy": 5.795116281509399,
      "epoch": 3.0241976269208326,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040898113698419823,
      "loss": 4.9846,
      "mean_token_accuracy": 0.20727591663599015,
      "num_tokens": 67423023.0,
      "step": 38870
    },
    {
      "entropy": 5.847828340530396,
      "epoch": 3.0245866562925503,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004089588699223713,
      "loss": 5.0139,
      "mean_token_accuracy": 0.2003832772374153,
      "num_tokens": 67431377.0,
      "step": 38875
    },
    {
      "entropy": 5.8501341342926025,
      "epoch": 3.0249756856642676,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040893660082793167,
      "loss": 5.039,
      "mean_token_accuracy": 0.20102416723966599,
      "num_tokens": 67439888.0,
      "step": 38880
    },
    {
      "entropy": 5.89390230178833,
      "epoch": 3.0253647150359853,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040891432970121736,
      "loss": 4.9913,
      "mean_token_accuracy": 0.20649498850107192,
      "num_tokens": 67448174.0,
      "step": 38885
    },
    {
      "entropy": 5.843033504486084,
      "epoch": 3.025753744407703,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040889205654256615,
      "loss": 4.9893,
      "mean_token_accuracy": 0.20673566609621047,
      "num_tokens": 67456562.0,
      "step": 38890
    },
    {
      "entropy": 5.873806571960449,
      "epoch": 3.02614277377942,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040886978135231625,
      "loss": 5.0211,
      "mean_token_accuracy": 0.20730582773685455,
      "num_tokens": 67465518.0,
      "step": 38895
    },
    {
      "entropy": 5.864649820327759,
      "epoch": 3.026531803151138,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040884750413080535,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20669220238924027,
      "num_tokens": 67474988.0,
      "step": 38900
    },
    {
      "entropy": 5.739325523376465,
      "epoch": 3.0269208325228556,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040882522487837176,
      "loss": 4.8562,
      "mean_token_accuracy": 0.21792981624603272,
      "num_tokens": 67484097.0,
      "step": 38905
    },
    {
      "entropy": 5.8093891620635985,
      "epoch": 3.027309861894573,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040880294359535327,
      "loss": 4.9257,
      "mean_token_accuracy": 0.20927488058805466,
      "num_tokens": 67493348.0,
      "step": 38910
    },
    {
      "entropy": 5.834283447265625,
      "epoch": 3.0276988912662905,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040878066028208804,
      "loss": 4.9079,
      "mean_token_accuracy": 0.21435246765613555,
      "num_tokens": 67501586.0,
      "step": 38915
    },
    {
      "entropy": 5.881344032287598,
      "epoch": 3.0280879206380082,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040875837493891434,
      "loss": 5.0769,
      "mean_token_accuracy": 0.2027055189013481,
      "num_tokens": 67510416.0,
      "step": 38920
    },
    {
      "entropy": 5.757195520401001,
      "epoch": 3.028476950009726,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004087360875661701,
      "loss": 4.856,
      "mean_token_accuracy": 0.21772555112838746,
      "num_tokens": 67518974.0,
      "step": 38925
    },
    {
      "entropy": 5.734735012054443,
      "epoch": 3.028865979381443,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040871379816419376,
      "loss": 4.9527,
      "mean_token_accuracy": 0.2007041797041893,
      "num_tokens": 67528194.0,
      "step": 38930
    },
    {
      "entropy": 5.837574100494384,
      "epoch": 3.029255008753161,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004086915067333232,
      "loss": 4.996,
      "mean_token_accuracy": 0.21031927466392517,
      "num_tokens": 67536884.0,
      "step": 38935
    },
    {
      "entropy": 5.8759246349334715,
      "epoch": 3.0296440381248786,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004086692132738968,
      "loss": 4.9735,
      "mean_token_accuracy": 0.207813760638237,
      "num_tokens": 67544969.0,
      "step": 38940
    },
    {
      "entropy": 5.930097341537476,
      "epoch": 3.030033067496596,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000408646917786253,
      "loss": 5.1723,
      "mean_token_accuracy": 0.19234495759010314,
      "num_tokens": 67555681.0,
      "step": 38945
    },
    {
      "entropy": 5.860834121704102,
      "epoch": 3.0304220968683135,
      "grad_norm": 1.25,
      "learning_rate": 0.0004086246202707298,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21847247779369355,
      "num_tokens": 67564346.0,
      "step": 38950
    },
    {
      "entropy": 5.832881593704224,
      "epoch": 3.0308111262400312,
      "grad_norm": 1.53125,
      "learning_rate": 0.00040860232072766577,
      "loss": 5.0275,
      "mean_token_accuracy": 0.20158557295799256,
      "num_tokens": 67573497.0,
      "step": 38955
    },
    {
      "entropy": 5.828227472305298,
      "epoch": 3.0312001556117485,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004085800191573991,
      "loss": 4.9283,
      "mean_token_accuracy": 0.21165926605463029,
      "num_tokens": 67581842.0,
      "step": 38960
    },
    {
      "entropy": 5.835940742492676,
      "epoch": 3.031589184983466,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004085577155602683,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20788875073194504,
      "num_tokens": 67590763.0,
      "step": 38965
    },
    {
      "entropy": 5.82449779510498,
      "epoch": 3.031978214355184,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040853540993661175,
      "loss": 5.05,
      "mean_token_accuracy": 0.1979548454284668,
      "num_tokens": 67599232.0,
      "step": 38970
    },
    {
      "entropy": 5.827770519256592,
      "epoch": 3.0323672437269016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040851310228676787,
      "loss": 4.9788,
      "mean_token_accuracy": 0.21156167536973952,
      "num_tokens": 67608314.0,
      "step": 38975
    },
    {
      "entropy": 5.839980268478394,
      "epoch": 3.032756273098619,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004084907926110752,
      "loss": 4.9091,
      "mean_token_accuracy": 0.2161184161901474,
      "num_tokens": 67616902.0,
      "step": 38980
    },
    {
      "entropy": 5.789211940765381,
      "epoch": 3.0331453024703365,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004084684809098722,
      "loss": 4.9825,
      "mean_token_accuracy": 0.21209703236818314,
      "num_tokens": 67625163.0,
      "step": 38985
    },
    {
      "entropy": 5.738745260238647,
      "epoch": 3.0335343318420542,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040844616718349743,
      "loss": 4.9047,
      "mean_token_accuracy": 0.21769861727952958,
      "num_tokens": 67632997.0,
      "step": 38990
    },
    {
      "entropy": 5.817039823532104,
      "epoch": 3.0339233612137715,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040842385143228946,
      "loss": 4.9109,
      "mean_token_accuracy": 0.21328484863042832,
      "num_tokens": 67640813.0,
      "step": 38995
    },
    {
      "entropy": 5.784450721740723,
      "epoch": 3.034312390585489,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040840153365658693,
      "loss": 4.9078,
      "mean_token_accuracy": 0.2148588240146637,
      "num_tokens": 67649701.0,
      "step": 39000
    },
    {
      "epoch": 3.034312390585489,
      "eval_entropy": 5.529974334393704,
      "eval_loss": 5.0711822509765625,
      "eval_mean_token_accuracy": 0.21438223907905088,
      "eval_num_tokens": 67649701.0,
      "eval_runtime": 21.6194,
      "eval_samples_per_second": 1922.255,
      "eval_steps_per_second": 240.294,
      "step": 39000
    },
    {
      "entropy": 5.7857208251953125,
      "epoch": 3.034701419957207,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040837921385672855,
      "loss": 4.994,
      "mean_token_accuracy": 0.21321100741624832,
      "num_tokens": 67658663.0,
      "step": 39005
    },
    {
      "entropy": 5.84175214767456,
      "epoch": 3.035090449328924,
      "grad_norm": 1.25,
      "learning_rate": 0.00040835689203305275,
      "loss": 4.9695,
      "mean_token_accuracy": 0.2097387731075287,
      "num_tokens": 67667091.0,
      "step": 39010
    },
    {
      "entropy": 5.822213888168335,
      "epoch": 3.035479478700642,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040833456818589845,
      "loss": 4.9191,
      "mean_token_accuracy": 0.20974235534667968,
      "num_tokens": 67675413.0,
      "step": 39015
    },
    {
      "entropy": 5.809333658218383,
      "epoch": 3.0358685080723595,
      "grad_norm": 1.25,
      "learning_rate": 0.00040831224231560424,
      "loss": 4.973,
      "mean_token_accuracy": 0.21274153590202333,
      "num_tokens": 67684215.0,
      "step": 39020
    },
    {
      "entropy": 5.808569431304932,
      "epoch": 3.0362575374440772,
      "grad_norm": 1.609375,
      "learning_rate": 0.000408289914422509,
      "loss": 5.0449,
      "mean_token_accuracy": 0.20571707487106322,
      "num_tokens": 67692901.0,
      "step": 39025
    },
    {
      "entropy": 5.850010776519776,
      "epoch": 3.0366465668157945,
      "grad_norm": 1.375,
      "learning_rate": 0.0004082675845069515,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21396350115537643,
      "num_tokens": 67701248.0,
      "step": 39030
    },
    {
      "entropy": 5.895647764205933,
      "epoch": 3.037035596187512,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004082452525692704,
      "loss": 4.9961,
      "mean_token_accuracy": 0.20708965510129929,
      "num_tokens": 67709893.0,
      "step": 39035
    },
    {
      "entropy": 5.828452157974243,
      "epoch": 3.03742462555923,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040822291860980484,
      "loss": 4.9661,
      "mean_token_accuracy": 0.21064845472574234,
      "num_tokens": 67718139.0,
      "step": 39040
    },
    {
      "entropy": 5.817484283447266,
      "epoch": 3.037813654930947,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040820058262889343,
      "loss": 4.9742,
      "mean_token_accuracy": 0.20650716125965118,
      "num_tokens": 67726433.0,
      "step": 39045
    },
    {
      "entropy": 5.79857611656189,
      "epoch": 3.038202684302665,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040817824462687516,
      "loss": 4.9507,
      "mean_token_accuracy": 0.20646706968545914,
      "num_tokens": 67735473.0,
      "step": 39050
    },
    {
      "entropy": 5.808203029632568,
      "epoch": 3.0385917136743825,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040815590460408906,
      "loss": 4.818,
      "mean_token_accuracy": 0.2121747151017189,
      "num_tokens": 67743627.0,
      "step": 39055
    },
    {
      "entropy": 5.79123821258545,
      "epoch": 3.0389807430460998,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040813356256087395,
      "loss": 4.9003,
      "mean_token_accuracy": 0.21501027345657348,
      "num_tokens": 67752160.0,
      "step": 39060
    },
    {
      "entropy": 5.740716743469238,
      "epoch": 3.0393697724178175,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004081112184975691,
      "loss": 4.9001,
      "mean_token_accuracy": 0.2133759453892708,
      "num_tokens": 67760598.0,
      "step": 39065
    },
    {
      "entropy": 5.8257927894592285,
      "epoch": 3.039758801789535,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004080888724145132,
      "loss": 4.9731,
      "mean_token_accuracy": 0.20731221735477448,
      "num_tokens": 67769081.0,
      "step": 39070
    },
    {
      "entropy": 5.842122793197632,
      "epoch": 3.040147831161253,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004080665243120456,
      "loss": 4.9701,
      "mean_token_accuracy": 0.20987311601638795,
      "num_tokens": 67778268.0,
      "step": 39075
    },
    {
      "entropy": 5.843275880813598,
      "epoch": 3.04053686053297,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004080441741905053,
      "loss": 4.9744,
      "mean_token_accuracy": 0.2088385060429573,
      "num_tokens": 67787744.0,
      "step": 39080
    },
    {
      "entropy": 5.803491878509521,
      "epoch": 3.040925889904688,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004080218220502314,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21412625759840012,
      "num_tokens": 67796324.0,
      "step": 39085
    },
    {
      "entropy": 5.801105213165283,
      "epoch": 3.0413149192764055,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040799946789156304,
      "loss": 4.9067,
      "mean_token_accuracy": 0.21767957657575607,
      "num_tokens": 67805303.0,
      "step": 39090
    },
    {
      "entropy": 5.861781930923462,
      "epoch": 3.0417039486481228,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040797711171483954,
      "loss": 5.0783,
      "mean_token_accuracy": 0.19940189868211747,
      "num_tokens": 67814300.0,
      "step": 39095
    },
    {
      "entropy": 5.755396366119385,
      "epoch": 3.0420929780198405,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004079547535203999,
      "loss": 4.9204,
      "mean_token_accuracy": 0.21215637624263764,
      "num_tokens": 67822512.0,
      "step": 39100
    },
    {
      "entropy": 5.818038272857666,
      "epoch": 3.042482007391558,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004079323933085836,
      "loss": 5.0151,
      "mean_token_accuracy": 0.20936035811901094,
      "num_tokens": 67830618.0,
      "step": 39105
    },
    {
      "entropy": 5.8200901508331295,
      "epoch": 3.0428710367632754,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040791003107972976,
      "loss": 4.9212,
      "mean_token_accuracy": 0.21017327308654785,
      "num_tokens": 67838623.0,
      "step": 39110
    },
    {
      "entropy": 5.708788537979126,
      "epoch": 3.043260066134993,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004078876668341778,
      "loss": 4.8182,
      "mean_token_accuracy": 0.2211902692914009,
      "num_tokens": 67847676.0,
      "step": 39115
    },
    {
      "entropy": 5.78536434173584,
      "epoch": 3.043649095506711,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004078653005722669,
      "loss": 4.9364,
      "mean_token_accuracy": 0.2063515529036522,
      "num_tokens": 67855896.0,
      "step": 39120
    },
    {
      "entropy": 5.839370203018189,
      "epoch": 3.0440381248784285,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004078429322943366,
      "loss": 4.9808,
      "mean_token_accuracy": 0.20460814535617827,
      "num_tokens": 67864028.0,
      "step": 39125
    },
    {
      "entropy": 5.772445011138916,
      "epoch": 3.0444271542501458,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040782056200072627,
      "loss": 4.93,
      "mean_token_accuracy": 0.2093192458152771,
      "num_tokens": 67873518.0,
      "step": 39130
    },
    {
      "entropy": 5.898752880096436,
      "epoch": 3.0448161836218635,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040779818969177527,
      "loss": 5.0437,
      "mean_token_accuracy": 0.20342600643634795,
      "num_tokens": 67882432.0,
      "step": 39135
    },
    {
      "entropy": 5.857030153274536,
      "epoch": 3.045205212993581,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040777581536782315,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20856084674596786,
      "num_tokens": 67891448.0,
      "step": 39140
    },
    {
      "entropy": 5.919822311401367,
      "epoch": 3.0455942423652984,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004077534390292093,
      "loss": 5.0942,
      "mean_token_accuracy": 0.2053716629743576,
      "num_tokens": 67900460.0,
      "step": 39145
    },
    {
      "entropy": 5.861373472213745,
      "epoch": 3.045983271737016,
      "grad_norm": 1.375,
      "learning_rate": 0.00040773106067627335,
      "loss": 4.9693,
      "mean_token_accuracy": 0.2069399982690811,
      "num_tokens": 67908515.0,
      "step": 39150
    },
    {
      "entropy": 5.775004100799561,
      "epoch": 3.046372301108734,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040770868030935486,
      "loss": 4.8771,
      "mean_token_accuracy": 0.21028319299221038,
      "num_tokens": 67917467.0,
      "step": 39155
    },
    {
      "entropy": 5.791110229492188,
      "epoch": 3.046761330480451,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040768629792879344,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20362433791160583,
      "num_tokens": 67926322.0,
      "step": 39160
    },
    {
      "entropy": 5.805644559860229,
      "epoch": 3.0471503598521688,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004076639135349284,
      "loss": 5.0069,
      "mean_token_accuracy": 0.2024402290582657,
      "num_tokens": 67935022.0,
      "step": 39165
    },
    {
      "entropy": 5.863823270797729,
      "epoch": 3.0475393892238865,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040764152712809983,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21176383942365645,
      "num_tokens": 67943536.0,
      "step": 39170
    },
    {
      "entropy": 5.787704181671143,
      "epoch": 3.047928418595604,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004076191387086472,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21085478216409684,
      "num_tokens": 67952412.0,
      "step": 39175
    },
    {
      "entropy": 5.801259088516235,
      "epoch": 3.0483174479673214,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040759674827691014,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20871176719665527,
      "num_tokens": 67961318.0,
      "step": 39180
    },
    {
      "entropy": 5.887573909759522,
      "epoch": 3.048706477339039,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004075743558332285,
      "loss": 4.9735,
      "mean_token_accuracy": 0.20969752669334413,
      "num_tokens": 67969190.0,
      "step": 39185
    },
    {
      "entropy": 5.820132875442505,
      "epoch": 3.049095506710757,
      "grad_norm": 1.328125,
      "learning_rate": 0.000407551961377942,
      "loss": 4.8747,
      "mean_token_accuracy": 0.2154114231467247,
      "num_tokens": 67978339.0,
      "step": 39190
    },
    {
      "entropy": 5.694484519958496,
      "epoch": 3.049484536082474,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004075295649113905,
      "loss": 4.8598,
      "mean_token_accuracy": 0.2103315308690071,
      "num_tokens": 67987806.0,
      "step": 39195
    },
    {
      "entropy": 5.852518558502197,
      "epoch": 3.0498735654541917,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040750716643391384,
      "loss": 5.0145,
      "mean_token_accuracy": 0.2103867158293724,
      "num_tokens": 67997017.0,
      "step": 39200
    },
    {
      "entropy": 5.843323183059693,
      "epoch": 3.0502625948259094,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004074847659458518,
      "loss": 5.0001,
      "mean_token_accuracy": 0.20470632165670394,
      "num_tokens": 68006032.0,
      "step": 39205
    },
    {
      "entropy": 5.8357017040252686,
      "epoch": 3.050651624197627,
      "grad_norm": 1.390625,
      "learning_rate": 0.00040746236344754423,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20628325194120406,
      "num_tokens": 68014644.0,
      "step": 39210
    },
    {
      "entropy": 5.840534639358521,
      "epoch": 3.0510406535693444,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004074399589393312,
      "loss": 4.9821,
      "mean_token_accuracy": 0.2004847928881645,
      "num_tokens": 68023302.0,
      "step": 39215
    },
    {
      "entropy": 5.785561466217041,
      "epoch": 3.051429682941062,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004074175524215526,
      "loss": 4.9052,
      "mean_token_accuracy": 0.21334371715784073,
      "num_tokens": 68031644.0,
      "step": 39220
    },
    {
      "entropy": 5.791380882263184,
      "epoch": 3.05181871231278,
      "grad_norm": 1.25,
      "learning_rate": 0.0004073951438945484,
      "loss": 4.935,
      "mean_token_accuracy": 0.20596230328083037,
      "num_tokens": 68039789.0,
      "step": 39225
    },
    {
      "entropy": 5.828952980041504,
      "epoch": 3.052207741684497,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040737273335865866,
      "loss": 5.0099,
      "mean_token_accuracy": 0.208457051217556,
      "num_tokens": 68048131.0,
      "step": 39230
    },
    {
      "entropy": 5.841406154632568,
      "epoch": 3.0525967710562147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040735032081422334,
      "loss": 5.062,
      "mean_token_accuracy": 0.2044255644083023,
      "num_tokens": 68057009.0,
      "step": 39235
    },
    {
      "entropy": 5.848847770690918,
      "epoch": 3.0529858004279324,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040732790626158257,
      "loss": 4.9306,
      "mean_token_accuracy": 0.21125787049531936,
      "num_tokens": 68065611.0,
      "step": 39240
    },
    {
      "entropy": 5.837436008453369,
      "epoch": 3.0533748297996497,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00040730548970107646,
      "loss": 5.0033,
      "mean_token_accuracy": 0.20553603172302246,
      "num_tokens": 68073892.0,
      "step": 39245
    },
    {
      "entropy": 5.809541940689087,
      "epoch": 3.0537638591713674,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004072830711330451,
      "loss": 4.9729,
      "mean_token_accuracy": 0.2095554992556572,
      "num_tokens": 68081406.0,
      "step": 39250
    },
    {
      "entropy": 5.737035655975342,
      "epoch": 3.054152888543085,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004072606505578287,
      "loss": 4.8333,
      "mean_token_accuracy": 0.22030352652072907,
      "num_tokens": 68089883.0,
      "step": 39255
    },
    {
      "entropy": 5.763055610656738,
      "epoch": 3.054541917914803,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040723822797576737,
      "loss": 4.929,
      "mean_token_accuracy": 0.2127308130264282,
      "num_tokens": 68098587.0,
      "step": 39260
    },
    {
      "entropy": 5.871400165557861,
      "epoch": 3.05493094728652,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004072158033872015,
      "loss": 5.0534,
      "mean_token_accuracy": 0.20424974262714385,
      "num_tokens": 68107802.0,
      "step": 39265
    },
    {
      "entropy": 5.887045001983642,
      "epoch": 3.0553199766582377,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004071933767924713,
      "loss": 5.0835,
      "mean_token_accuracy": 0.199178446829319,
      "num_tokens": 68117777.0,
      "step": 39270
    },
    {
      "entropy": 5.841463804244995,
      "epoch": 3.0557090060299554,
      "grad_norm": 1.25,
      "learning_rate": 0.00040717094819191685,
      "loss": 4.9131,
      "mean_token_accuracy": 0.2105742335319519,
      "num_tokens": 68126964.0,
      "step": 39275
    },
    {
      "entropy": 5.791716003417969,
      "epoch": 3.0560980354016727,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040714851758587875,
      "loss": 4.9599,
      "mean_token_accuracy": 0.20925536304712294,
      "num_tokens": 68135801.0,
      "step": 39280
    },
    {
      "entropy": 5.802290678024292,
      "epoch": 3.0564870647733904,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040712608497469715,
      "loss": 5.0026,
      "mean_token_accuracy": 0.20765635073184968,
      "num_tokens": 68144930.0,
      "step": 39285
    },
    {
      "entropy": 5.760097980499268,
      "epoch": 3.056876094145108,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004071036503587126,
      "loss": 4.884,
      "mean_token_accuracy": 0.22195216566324233,
      "num_tokens": 68153000.0,
      "step": 39290
    },
    {
      "entropy": 5.838367652893067,
      "epoch": 3.0572651235168253,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004070812137382654,
      "loss": 5.0467,
      "mean_token_accuracy": 0.19612378478050232,
      "num_tokens": 68161510.0,
      "step": 39295
    },
    {
      "entropy": 5.900693130493164,
      "epoch": 3.057654152888543,
      "grad_norm": 1.3125,
      "learning_rate": 0.000407058775113696,
      "loss": 4.9795,
      "mean_token_accuracy": 0.21024883091449736,
      "num_tokens": 68170355.0,
      "step": 39300
    },
    {
      "entropy": 5.79980502128601,
      "epoch": 3.0580431822602607,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004070363344853449,
      "loss": 4.938,
      "mean_token_accuracy": 0.20916044116020202,
      "num_tokens": 68178611.0,
      "step": 39305
    },
    {
      "entropy": 5.803950262069702,
      "epoch": 3.0584322116319784,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040701389185355256,
      "loss": 4.9049,
      "mean_token_accuracy": 0.2097593605518341,
      "num_tokens": 68186799.0,
      "step": 39310
    },
    {
      "entropy": 5.825037574768066,
      "epoch": 3.0588212410036957,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004069914472186595,
      "loss": 4.9623,
      "mean_token_accuracy": 0.2136516496539116,
      "num_tokens": 68196261.0,
      "step": 39315
    },
    {
      "entropy": 5.831733369827271,
      "epoch": 3.0592102703754134,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004069690005810063,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21453803032636642,
      "num_tokens": 68204373.0,
      "step": 39320
    },
    {
      "entropy": 5.799441862106323,
      "epoch": 3.059599299747131,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040694655194093365,
      "loss": 4.9447,
      "mean_token_accuracy": 0.2057499721646309,
      "num_tokens": 68212972.0,
      "step": 39325
    },
    {
      "entropy": 5.84109902381897,
      "epoch": 3.0599883291188483,
      "grad_norm": 1.25,
      "learning_rate": 0.00040692410129878205,
      "loss": 5.0498,
      "mean_token_accuracy": 0.20423805415630342,
      "num_tokens": 68221317.0,
      "step": 39330
    },
    {
      "entropy": 5.851670742034912,
      "epoch": 3.060377358490566,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004069016486548922,
      "loss": 4.9854,
      "mean_token_accuracy": 0.2063519462943077,
      "num_tokens": 68229961.0,
      "step": 39335
    },
    {
      "entropy": 5.8360865116119385,
      "epoch": 3.0607663878622837,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004068791940096048,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21365064829587938,
      "num_tokens": 68238527.0,
      "step": 39340
    },
    {
      "entropy": 5.80261263847351,
      "epoch": 3.061155417234001,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040685673736326044,
      "loss": 4.9629,
      "mean_token_accuracy": 0.2109831839799881,
      "num_tokens": 68246935.0,
      "step": 39345
    },
    {
      "entropy": 5.862231969833374,
      "epoch": 3.0615444466057187,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004068342787162001,
      "loss": 4.9432,
      "mean_token_accuracy": 0.21289208084344863,
      "num_tokens": 68255370.0,
      "step": 39350
    },
    {
      "entropy": 5.8587945938110355,
      "epoch": 3.0619334759774364,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040681181806876435,
      "loss": 4.9652,
      "mean_token_accuracy": 0.20835043042898177,
      "num_tokens": 68264970.0,
      "step": 39355
    },
    {
      "entropy": 5.835612344741821,
      "epoch": 3.062322505349154,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000406789355421294,
      "loss": 5.0285,
      "mean_token_accuracy": 0.2052568942308426,
      "num_tokens": 68274158.0,
      "step": 39360
    },
    {
      "entropy": 5.840562868118286,
      "epoch": 3.0627115347208713,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004067668907741301,
      "loss": 5.0445,
      "mean_token_accuracy": 0.20047302842140197,
      "num_tokens": 68282939.0,
      "step": 39365
    },
    {
      "entropy": 5.865649700164795,
      "epoch": 3.063100564092589,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004067444241276132,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21067041903734207,
      "num_tokens": 68291235.0,
      "step": 39370
    },
    {
      "entropy": 5.848631763458252,
      "epoch": 3.0634895934643067,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040672195548208443,
      "loss": 4.9624,
      "mean_token_accuracy": 0.21415946930646895,
      "num_tokens": 68300498.0,
      "step": 39375
    },
    {
      "entropy": 5.887886953353882,
      "epoch": 3.063878622836024,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004066994848378846,
      "loss": 5.0683,
      "mean_token_accuracy": 0.20751016438007355,
      "num_tokens": 68309320.0,
      "step": 39380
    },
    {
      "entropy": 5.804331207275391,
      "epoch": 3.0642676522077417,
      "grad_norm": 1.25,
      "learning_rate": 0.00040667701219535473,
      "loss": 4.9386,
      "mean_token_accuracy": 0.20843724757432938,
      "num_tokens": 68318462.0,
      "step": 39385
    },
    {
      "entropy": 5.78813362121582,
      "epoch": 3.0646566815794594,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004066545375548358,
      "loss": 4.9814,
      "mean_token_accuracy": 0.20863882452249527,
      "num_tokens": 68327559.0,
      "step": 39390
    },
    {
      "entropy": 5.82707257270813,
      "epoch": 3.0650457109511766,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004066320609166689,
      "loss": 4.9805,
      "mean_token_accuracy": 0.20834171921014785,
      "num_tokens": 68336371.0,
      "step": 39395
    },
    {
      "entropy": 5.826039409637451,
      "epoch": 3.0654347403228943,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004066095822811949,
      "loss": 4.9667,
      "mean_token_accuracy": 0.2084718704223633,
      "num_tokens": 68345439.0,
      "step": 39400
    },
    {
      "entropy": 5.87598876953125,
      "epoch": 3.065823769694612,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000406587101648755,
      "loss": 5.0576,
      "mean_token_accuracy": 0.2105918288230896,
      "num_tokens": 68354228.0,
      "step": 39405
    },
    {
      "entropy": 5.803850746154785,
      "epoch": 3.0662127990663297,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004065646190196902,
      "loss": 4.9108,
      "mean_token_accuracy": 0.2077522471547127,
      "num_tokens": 68362264.0,
      "step": 39410
    },
    {
      "entropy": 5.794232797622681,
      "epoch": 3.066601828438047,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004065421343943417,
      "loss": 4.9122,
      "mean_token_accuracy": 0.20513679087162018,
      "num_tokens": 68370158.0,
      "step": 39415
    },
    {
      "entropy": 5.870185327529907,
      "epoch": 3.0669908578097647,
      "grad_norm": 1.40625,
      "learning_rate": 0.00040651964777305074,
      "loss": 4.9839,
      "mean_token_accuracy": 0.2094171941280365,
      "num_tokens": 68378193.0,
      "step": 39420
    },
    {
      "entropy": 5.787476634979248,
      "epoch": 3.0673798871814824,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004064971591561585,
      "loss": 4.9403,
      "mean_token_accuracy": 0.20669917464256288,
      "num_tokens": 68386840.0,
      "step": 39425
    },
    {
      "entropy": 5.733916616439819,
      "epoch": 3.0677689165531996,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040647466854400606,
      "loss": 4.9605,
      "mean_token_accuracy": 0.21660890132188798,
      "num_tokens": 68395587.0,
      "step": 39430
    },
    {
      "entropy": 5.724660348892212,
      "epoch": 3.0681579459249173,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004064521759369349,
      "loss": 4.8297,
      "mean_token_accuracy": 0.2217332437634468,
      "num_tokens": 68403921.0,
      "step": 39435
    },
    {
      "entropy": 5.820328950881958,
      "epoch": 3.068546975296635,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040642968133528613,
      "loss": 4.9732,
      "mean_token_accuracy": 0.2090347394347191,
      "num_tokens": 68412130.0,
      "step": 39440
    },
    {
      "entropy": 5.769091510772705,
      "epoch": 3.0689360046683523,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004064071847394012,
      "loss": 4.9087,
      "mean_token_accuracy": 0.21442843228578568,
      "num_tokens": 68420113.0,
      "step": 39445
    },
    {
      "entropy": 5.803831911087036,
      "epoch": 3.06932503404007,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004063846861496213,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20604742616415023,
      "num_tokens": 68428397.0,
      "step": 39450
    },
    {
      "entropy": 5.8736429691314695,
      "epoch": 3.0697140634117877,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000406362185566288,
      "loss": 5.0762,
      "mean_token_accuracy": 0.1993295818567276,
      "num_tokens": 68436607.0,
      "step": 39455
    },
    {
      "entropy": 5.870149374008179,
      "epoch": 3.0701030927835053,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004063396829897426,
      "loss": 4.999,
      "mean_token_accuracy": 0.21155939102172852,
      "num_tokens": 68445167.0,
      "step": 39460
    },
    {
      "entropy": 5.8394163131713865,
      "epoch": 3.0704921221552226,
      "grad_norm": 1.375,
      "learning_rate": 0.0004063171784203265,
      "loss": 5.0191,
      "mean_token_accuracy": 0.20901959240436555,
      "num_tokens": 68453758.0,
      "step": 39465
    },
    {
      "entropy": 5.922358703613281,
      "epoch": 3.0708811515269403,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004062946718583813,
      "loss": 5.1149,
      "mean_token_accuracy": 0.20606088638305664,
      "num_tokens": 68462536.0,
      "step": 39470
    },
    {
      "entropy": 5.850831794738769,
      "epoch": 3.071270180898658,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040627216330424844,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20335010141134263,
      "num_tokens": 68470862.0,
      "step": 39475
    },
    {
      "entropy": 5.8243536949157715,
      "epoch": 3.0716592102703753,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004062496527582694,
      "loss": 5.055,
      "mean_token_accuracy": 0.201138673722744,
      "num_tokens": 68480212.0,
      "step": 39480
    },
    {
      "entropy": 5.699077987670899,
      "epoch": 3.072048239642093,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004062271402207858,
      "loss": 4.7985,
      "mean_token_accuracy": 0.23066861927509308,
      "num_tokens": 68489883.0,
      "step": 39485
    },
    {
      "entropy": 5.839458227157593,
      "epoch": 3.0724372690138106,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004062046256921391,
      "loss": 5.0256,
      "mean_token_accuracy": 0.20650219321250915,
      "num_tokens": 68498875.0,
      "step": 39490
    },
    {
      "entropy": 5.839842033386231,
      "epoch": 3.072826298385528,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040618210917267117,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21424026638269425,
      "num_tokens": 68508212.0,
      "step": 39495
    },
    {
      "entropy": 5.802602863311767,
      "epoch": 3.0732153277572456,
      "grad_norm": 1.25,
      "learning_rate": 0.00040615959066272355,
      "loss": 4.9759,
      "mean_token_accuracy": 0.21247861087322234,
      "num_tokens": 68516990.0,
      "step": 39500
    },
    {
      "entropy": 5.866367626190185,
      "epoch": 3.0736043571289633,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004061370701626379,
      "loss": 5.0877,
      "mean_token_accuracy": 0.2035200998187065,
      "num_tokens": 68526807.0,
      "step": 39505
    },
    {
      "entropy": 5.8027928352355955,
      "epoch": 3.073993386500681,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004061145476727559,
      "loss": 5.0126,
      "mean_token_accuracy": 0.20905337035655974,
      "num_tokens": 68535327.0,
      "step": 39510
    },
    {
      "entropy": 5.914617300033569,
      "epoch": 3.0743824158723982,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004060920231934193,
      "loss": 5.0557,
      "mean_token_accuracy": 0.2015492394566536,
      "num_tokens": 68543389.0,
      "step": 39515
    },
    {
      "entropy": 5.8522467613220215,
      "epoch": 3.074771445244116,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004060694967249699,
      "loss": 4.9381,
      "mean_token_accuracy": 0.2134429022669792,
      "num_tokens": 68552271.0,
      "step": 39520
    },
    {
      "entropy": 5.833958005905151,
      "epoch": 3.0751604746158336,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004060469682677496,
      "loss": 4.9962,
      "mean_token_accuracy": 0.2096745640039444,
      "num_tokens": 68560931.0,
      "step": 39525
    },
    {
      "entropy": 5.79143443107605,
      "epoch": 3.075549503987551,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040602443782210003,
      "loss": 4.9347,
      "mean_token_accuracy": 0.21465360522270202,
      "num_tokens": 68569871.0,
      "step": 39530
    },
    {
      "entropy": 5.818266773223877,
      "epoch": 3.0759385333592686,
      "grad_norm": 1.375,
      "learning_rate": 0.0004060019053883632,
      "loss": 4.9225,
      "mean_token_accuracy": 0.2129931703209877,
      "num_tokens": 68578167.0,
      "step": 39535
    },
    {
      "entropy": 5.822011947631836,
      "epoch": 3.0763275627309863,
      "grad_norm": 1.25,
      "learning_rate": 0.000405979370966881,
      "loss": 4.956,
      "mean_token_accuracy": 0.20952725708484649,
      "num_tokens": 68587254.0,
      "step": 39540
    },
    {
      "entropy": 5.816270017623902,
      "epoch": 3.0767165921027035,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004059568345579953,
      "loss": 5.0162,
      "mean_token_accuracy": 0.2016645833849907,
      "num_tokens": 68595405.0,
      "step": 39545
    },
    {
      "entropy": 5.775061321258545,
      "epoch": 3.0771056214744212,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040593429616204806,
      "loss": 4.8878,
      "mean_token_accuracy": 0.21784939169883727,
      "num_tokens": 68604092.0,
      "step": 39550
    },
    {
      "entropy": 5.861136865615845,
      "epoch": 3.077494650846139,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040591175577938125,
      "loss": 4.914,
      "mean_token_accuracy": 0.22066772878170013,
      "num_tokens": 68612697.0,
      "step": 39555
    },
    {
      "entropy": 5.872352075576782,
      "epoch": 3.0778836802178566,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040588921341033697,
      "loss": 5.0595,
      "mean_token_accuracy": 0.2054194226861,
      "num_tokens": 68621770.0,
      "step": 39560
    },
    {
      "entropy": 5.805569410324097,
      "epoch": 3.078272709589574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004058666690552572,
      "loss": 4.9623,
      "mean_token_accuracy": 0.20760195404291154,
      "num_tokens": 68631789.0,
      "step": 39565
    },
    {
      "entropy": 5.8545427322387695,
      "epoch": 3.0786617389612916,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040584412271448407,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21655955910682678,
      "num_tokens": 68640033.0,
      "step": 39570
    },
    {
      "entropy": 5.813673210144043,
      "epoch": 3.0790507683330093,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004058215743883595,
      "loss": 5.0112,
      "mean_token_accuracy": 0.20706821978092194,
      "num_tokens": 68649082.0,
      "step": 39575
    },
    {
      "entropy": 5.817861080169678,
      "epoch": 3.0794397977047265,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004057990240772259,
      "loss": 4.9893,
      "mean_token_accuracy": 0.21172555088996886,
      "num_tokens": 68657965.0,
      "step": 39580
    },
    {
      "entropy": 5.852955675125122,
      "epoch": 3.0798288270764442,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00040577647178142523,
      "loss": 4.9404,
      "mean_token_accuracy": 0.21825944036245346,
      "num_tokens": 68666002.0,
      "step": 39585
    },
    {
      "entropy": 5.847254800796509,
      "epoch": 3.080217856448162,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004057539175012998,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2024742394685745,
      "num_tokens": 68675517.0,
      "step": 39590
    },
    {
      "entropy": 5.769647789001465,
      "epoch": 3.080606885819879,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040573136123719174,
      "loss": 4.9251,
      "mean_token_accuracy": 0.2065741539001465,
      "num_tokens": 68683943.0,
      "step": 39595
    },
    {
      "entropy": 5.751211881637573,
      "epoch": 3.080995915191597,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040570880298944334,
      "loss": 4.8944,
      "mean_token_accuracy": 0.20927375555038452,
      "num_tokens": 68692761.0,
      "step": 39600
    },
    {
      "entropy": 5.817933750152588,
      "epoch": 3.0813849445633146,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004056862427583969,
      "loss": 4.9649,
      "mean_token_accuracy": 0.2141027793288231,
      "num_tokens": 68701255.0,
      "step": 39605
    },
    {
      "entropy": 5.681625843048096,
      "epoch": 3.0817739739350323,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004056636805443948,
      "loss": 4.8295,
      "mean_token_accuracy": 0.22289353609085083,
      "num_tokens": 68709654.0,
      "step": 39610
    },
    {
      "entropy": 5.764019298553467,
      "epoch": 3.0821630033067495,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040564111634777927,
      "loss": 4.9693,
      "mean_token_accuracy": 0.20871956497430802,
      "num_tokens": 68719355.0,
      "step": 39615
    },
    {
      "entropy": 5.88094162940979,
      "epoch": 3.0825520326784672,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040561855016889276,
      "loss": 5.0156,
      "mean_token_accuracy": 0.20753912776708602,
      "num_tokens": 68728495.0,
      "step": 39620
    },
    {
      "entropy": 5.892963504791259,
      "epoch": 3.082941062050185,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040559598200807763,
      "loss": 4.9776,
      "mean_token_accuracy": 0.20800368189811708,
      "num_tokens": 68737098.0,
      "step": 39625
    },
    {
      "entropy": 5.8221831798553465,
      "epoch": 3.083330091421902,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004055734118656764,
      "loss": 4.9605,
      "mean_token_accuracy": 0.21473665833473204,
      "num_tokens": 68745563.0,
      "step": 39630
    },
    {
      "entropy": 5.754200744628906,
      "epoch": 3.08371912079362,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004055508397420314,
      "loss": 4.9054,
      "mean_token_accuracy": 0.21765429377555848,
      "num_tokens": 68753861.0,
      "step": 39635
    },
    {
      "entropy": 5.7909736156463625,
      "epoch": 3.0841081501653376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040552826563748526,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21376789063215257,
      "num_tokens": 68762623.0,
      "step": 39640
    },
    {
      "entropy": 5.801203298568725,
      "epoch": 3.0844971795370553,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004055056895523804,
      "loss": 4.9053,
      "mean_token_accuracy": 0.2101140484213829,
      "num_tokens": 68771235.0,
      "step": 39645
    },
    {
      "entropy": 5.814163589477539,
      "epoch": 3.0848862089087725,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040548311148705943,
      "loss": 5.0015,
      "mean_token_accuracy": 0.20949250012636184,
      "num_tokens": 68780264.0,
      "step": 39650
    },
    {
      "entropy": 5.7955036640167235,
      "epoch": 3.08527523828049,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004054605314418649,
      "loss": 4.97,
      "mean_token_accuracy": 0.21308656632900239,
      "num_tokens": 68789242.0,
      "step": 39655
    },
    {
      "entropy": 5.774907112121582,
      "epoch": 3.085664267652208,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040543794941713944,
      "loss": 4.9455,
      "mean_token_accuracy": 0.20601903945207595,
      "num_tokens": 68797441.0,
      "step": 39660
    },
    {
      "entropy": 5.8184737205505375,
      "epoch": 3.086053297023925,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004054153654132258,
      "loss": 5.0252,
      "mean_token_accuracy": 0.21047117859125136,
      "num_tokens": 68805884.0,
      "step": 39665
    },
    {
      "entropy": 5.790157508850098,
      "epoch": 3.086442326395643,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004053927794304665,
      "loss": 4.9219,
      "mean_token_accuracy": 0.21730638593435286,
      "num_tokens": 68813338.0,
      "step": 39670
    },
    {
      "entropy": 5.7888861179351805,
      "epoch": 3.0868313557673606,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004053701914692043,
      "loss": 4.9191,
      "mean_token_accuracy": 0.2131543204188347,
      "num_tokens": 68822646.0,
      "step": 39675
    },
    {
      "entropy": 5.888801383972168,
      "epoch": 3.087220385139078,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000405347601529782,
      "loss": 5.1217,
      "mean_token_accuracy": 0.20044277906417846,
      "num_tokens": 68832089.0,
      "step": 39680
    },
    {
      "entropy": 5.7748359680175785,
      "epoch": 3.0876094145107955,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004053250096125423,
      "loss": 4.9706,
      "mean_token_accuracy": 0.20926367193460466,
      "num_tokens": 68839694.0,
      "step": 39685
    },
    {
      "entropy": 5.802931451797486,
      "epoch": 3.087998443882513,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004053024157178281,
      "loss": 4.9641,
      "mean_token_accuracy": 0.20656691938638688,
      "num_tokens": 68848554.0,
      "step": 39690
    },
    {
      "entropy": 5.8489851474761965,
      "epoch": 3.088387473254231,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000405279819845982,
      "loss": 5.0755,
      "mean_token_accuracy": 0.20811651796102523,
      "num_tokens": 68857148.0,
      "step": 39695
    },
    {
      "entropy": 5.813358449935913,
      "epoch": 3.088776502625948,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040525722199734706,
      "loss": 4.9575,
      "mean_token_accuracy": 0.209647336602211,
      "num_tokens": 68865576.0,
      "step": 39700
    },
    {
      "entropy": 5.902824020385742,
      "epoch": 3.089165531997666,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004052346221722661,
      "loss": 5.1067,
      "mean_token_accuracy": 0.20678790807723998,
      "num_tokens": 68873228.0,
      "step": 39705
    },
    {
      "entropy": 5.911577463150024,
      "epoch": 3.0895545613693836,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040521202037108205,
      "loss": 5.0286,
      "mean_token_accuracy": 0.2123262971639633,
      "num_tokens": 68881480.0,
      "step": 39710
    },
    {
      "entropy": 5.854762697219849,
      "epoch": 3.089943590741101,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004051894165941378,
      "loss": 4.9928,
      "mean_token_accuracy": 0.2029785007238388,
      "num_tokens": 68890347.0,
      "step": 39715
    },
    {
      "entropy": 5.870352077484131,
      "epoch": 3.0903326201128185,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040516681084177636,
      "loss": 5.0246,
      "mean_token_accuracy": 0.20288817584514618,
      "num_tokens": 68898792.0,
      "step": 39720
    },
    {
      "entropy": 5.832594633102417,
      "epoch": 3.090721649484536,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040514420311434085,
      "loss": 5.0027,
      "mean_token_accuracy": 0.20829457938671112,
      "num_tokens": 68906947.0,
      "step": 39725
    },
    {
      "entropy": 5.762972974777222,
      "epoch": 3.0911106788562535,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040512159341217415,
      "loss": 4.8748,
      "mean_token_accuracy": 0.21641161441802978,
      "num_tokens": 68915357.0,
      "step": 39730
    },
    {
      "entropy": 5.834885120391846,
      "epoch": 3.091499708227971,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040509898173561934,
      "loss": 4.9985,
      "mean_token_accuracy": 0.20679308474063873,
      "num_tokens": 68923734.0,
      "step": 39735
    },
    {
      "entropy": 5.858195781707764,
      "epoch": 3.091888737599689,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004050763680850196,
      "loss": 5.0197,
      "mean_token_accuracy": 0.20343829989433287,
      "num_tokens": 68932225.0,
      "step": 39740
    },
    {
      "entropy": 5.884276580810547,
      "epoch": 3.0922777669714065,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004050537524607179,
      "loss": 5.0069,
      "mean_token_accuracy": 0.21300562620162963,
      "num_tokens": 68939886.0,
      "step": 39745
    },
    {
      "entropy": 5.770421743392944,
      "epoch": 3.092666796343124,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040503113486305766,
      "loss": 4.9754,
      "mean_token_accuracy": 0.21022918373346328,
      "num_tokens": 68949188.0,
      "step": 39750
    },
    {
      "entropy": 5.823473691940308,
      "epoch": 3.0930558257148415,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040500851529238184,
      "loss": 4.9974,
      "mean_token_accuracy": 0.2110453024506569,
      "num_tokens": 68957480.0,
      "step": 39755
    },
    {
      "entropy": 5.8669188022613525,
      "epoch": 3.093444855086559,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004049858937490337,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20377072989940642,
      "num_tokens": 68967319.0,
      "step": 39760
    },
    {
      "entropy": 5.902419090270996,
      "epoch": 3.0938338844582765,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004049632702333566,
      "loss": 5.0781,
      "mean_token_accuracy": 0.20742001086473466,
      "num_tokens": 68975659.0,
      "step": 39765
    },
    {
      "entropy": 5.810904884338379,
      "epoch": 3.094222913829994,
      "grad_norm": 1.375,
      "learning_rate": 0.00040494064474569375,
      "loss": 4.9096,
      "mean_token_accuracy": 0.21777776926755904,
      "num_tokens": 68984292.0,
      "step": 39770
    },
    {
      "entropy": 5.818319034576416,
      "epoch": 3.094611943201712,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040491801728638835,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21426538825035096,
      "num_tokens": 68993527.0,
      "step": 39775
    },
    {
      "entropy": 5.844941139221191,
      "epoch": 3.095000972573429,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040489538785578385,
      "loss": 4.9806,
      "mean_token_accuracy": 0.20998334288597106,
      "num_tokens": 69001382.0,
      "step": 39780
    },
    {
      "entropy": 5.812329626083374,
      "epoch": 3.095390001945147,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040487275645422356,
      "loss": 4.9629,
      "mean_token_accuracy": 0.19717445075511933,
      "num_tokens": 69010930.0,
      "step": 39785
    },
    {
      "entropy": 5.807558536529541,
      "epoch": 3.0957790313168645,
      "grad_norm": 1.21875,
      "learning_rate": 0.000404850123082051,
      "loss": 4.8867,
      "mean_token_accuracy": 0.21644314378499985,
      "num_tokens": 69020107.0,
      "step": 39790
    },
    {
      "entropy": 5.8125909805297855,
      "epoch": 3.096168060688582,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004048274877396094,
      "loss": 4.9268,
      "mean_token_accuracy": 0.21209898740053176,
      "num_tokens": 69028895.0,
      "step": 39795
    },
    {
      "entropy": 5.8018796920776365,
      "epoch": 3.0965570900602994,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004048048504272424,
      "loss": 4.9891,
      "mean_token_accuracy": 0.21167099922895433,
      "num_tokens": 69037660.0,
      "step": 39800
    },
    {
      "entropy": 5.819730234146118,
      "epoch": 3.096946119432017,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004047822111452933,
      "loss": 4.9818,
      "mean_token_accuracy": 0.214534755051136,
      "num_tokens": 69045681.0,
      "step": 39805
    },
    {
      "entropy": 5.815293979644776,
      "epoch": 3.097335148803735,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040475956989410584,
      "loss": 4.8743,
      "mean_token_accuracy": 0.2164829909801483,
      "num_tokens": 69054577.0,
      "step": 39810
    },
    {
      "entropy": 5.81224422454834,
      "epoch": 3.097724178175452,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004047369266740233,
      "loss": 4.9433,
      "mean_token_accuracy": 0.21623441874980925,
      "num_tokens": 69062887.0,
      "step": 39815
    },
    {
      "entropy": 5.847972774505616,
      "epoch": 3.09811320754717,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040471428148538944,
      "loss": 5.0658,
      "mean_token_accuracy": 0.20613196790218352,
      "num_tokens": 69071209.0,
      "step": 39820
    },
    {
      "entropy": 5.7591043472290036,
      "epoch": 3.0985022369188875,
      "grad_norm": 1.25,
      "learning_rate": 0.0004046916343285479,
      "loss": 4.8931,
      "mean_token_accuracy": 0.21789177507162094,
      "num_tokens": 69080026.0,
      "step": 39825
    },
    {
      "entropy": 5.864245939254761,
      "epoch": 3.0988912662906047,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040466898520384213,
      "loss": 5.0379,
      "mean_token_accuracy": 0.19988496154546737,
      "num_tokens": 69088283.0,
      "step": 39830
    },
    {
      "entropy": 5.7730776309967045,
      "epoch": 3.0992802956623224,
      "grad_norm": 1.34375,
      "learning_rate": 0.000404646334111616,
      "loss": 4.8324,
      "mean_token_accuracy": 0.21712164878845214,
      "num_tokens": 69097204.0,
      "step": 39835
    },
    {
      "entropy": 5.798954772949219,
      "epoch": 3.09966932503404,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004046236810522131,
      "loss": 5.0342,
      "mean_token_accuracy": 0.20535361915826797,
      "num_tokens": 69105586.0,
      "step": 39840
    },
    {
      "entropy": 5.744022560119629,
      "epoch": 3.100058354405758,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004046010260259771,
      "loss": 4.8966,
      "mean_token_accuracy": 0.21066858172416686,
      "num_tokens": 69115321.0,
      "step": 39845
    },
    {
      "entropy": 5.828773498535156,
      "epoch": 3.100447383777475,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040457836903325177,
      "loss": 4.947,
      "mean_token_accuracy": 0.210826176404953,
      "num_tokens": 69124472.0,
      "step": 39850
    },
    {
      "entropy": 5.818153953552246,
      "epoch": 3.100836413149193,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040455571007438104,
      "loss": 4.9516,
      "mean_token_accuracy": 0.21467797160148622,
      "num_tokens": 69132961.0,
      "step": 39855
    },
    {
      "entropy": 5.672987461090088,
      "epoch": 3.1012254425209105,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004045330491497085,
      "loss": 4.8365,
      "mean_token_accuracy": 0.21256709098815918,
      "num_tokens": 69142097.0,
      "step": 39860
    },
    {
      "entropy": 5.768345832824707,
      "epoch": 3.1016144718926277,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00040451038625957827,
      "loss": 4.944,
      "mean_token_accuracy": 0.21497393250465394,
      "num_tokens": 69150399.0,
      "step": 39865
    },
    {
      "entropy": 5.8519762516021725,
      "epoch": 3.1020035012643454,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004044877214043339,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20259847790002822,
      "num_tokens": 69158872.0,
      "step": 39870
    },
    {
      "entropy": 5.853964710235596,
      "epoch": 3.102392530636063,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004044650545843196,
      "loss": 5.0315,
      "mean_token_accuracy": 0.2086167514324188,
      "num_tokens": 69167852.0,
      "step": 39875
    },
    {
      "entropy": 5.76732177734375,
      "epoch": 3.1027815600077804,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000404442385799879,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21634586453437804,
      "num_tokens": 69176586.0,
      "step": 39880
    },
    {
      "entropy": 5.823845863342285,
      "epoch": 3.103170589379498,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004044197150513564,
      "loss": 5.0831,
      "mean_token_accuracy": 0.2007353574037552,
      "num_tokens": 69185512.0,
      "step": 39885
    },
    {
      "entropy": 5.876128768920898,
      "epoch": 3.1035596187512158,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004043970423390955,
      "loss": 5.0625,
      "mean_token_accuracy": 0.20131637901067734,
      "num_tokens": 69194696.0,
      "step": 39890
    },
    {
      "entropy": 5.86784029006958,
      "epoch": 3.1039486481229335,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004043743676634405,
      "loss": 4.9706,
      "mean_token_accuracy": 0.21102476865053177,
      "num_tokens": 69203533.0,
      "step": 39895
    },
    {
      "entropy": 5.865966558456421,
      "epoch": 3.1043376774946507,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004043516910247353,
      "loss": 5.0596,
      "mean_token_accuracy": 0.2029344156384468,
      "num_tokens": 69212864.0,
      "step": 39900
    },
    {
      "entropy": 5.800632333755493,
      "epoch": 3.1047267068663684,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004043290124233242,
      "loss": 4.9125,
      "mean_token_accuracy": 0.21313959807157518,
      "num_tokens": 69220598.0,
      "step": 39905
    },
    {
      "entropy": 5.795917654037476,
      "epoch": 3.105115736238086,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004043063318595511,
      "loss": 5.0053,
      "mean_token_accuracy": 0.20564965009689332,
      "num_tokens": 69230683.0,
      "step": 39910
    },
    {
      "entropy": 5.839149522781372,
      "epoch": 3.1055047656098034,
      "grad_norm": 1.25,
      "learning_rate": 0.0004042836493337603,
      "loss": 4.985,
      "mean_token_accuracy": 0.2078793004155159,
      "num_tokens": 69239274.0,
      "step": 39915
    },
    {
      "entropy": 5.8357256889343265,
      "epoch": 3.105893794981521,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004042609648462959,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21009508669376373,
      "num_tokens": 69248155.0,
      "step": 39920
    },
    {
      "entropy": 5.763851451873779,
      "epoch": 3.1062828243532388,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000404238278397502,
      "loss": 4.8461,
      "mean_token_accuracy": 0.2183850958943367,
      "num_tokens": 69256648.0,
      "step": 39925
    },
    {
      "entropy": 5.883499622344971,
      "epoch": 3.106671853724956,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040421558998772293,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21470011472702027,
      "num_tokens": 69265560.0,
      "step": 39930
    },
    {
      "entropy": 5.799188899993896,
      "epoch": 3.1070608830966737,
      "grad_norm": 1.328125,
      "learning_rate": 0.000404192899617303,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21160530745983125,
      "num_tokens": 69273826.0,
      "step": 39935
    },
    {
      "entropy": 5.900342273712158,
      "epoch": 3.1074499124683914,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004041702072865864,
      "loss": 5.1223,
      "mean_token_accuracy": 0.20296073257923125,
      "num_tokens": 69282734.0,
      "step": 39940
    },
    {
      "entropy": 5.851091098785401,
      "epoch": 3.107838941840109,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004041475129959176,
      "loss": 4.9858,
      "mean_token_accuracy": 0.20718630850315095,
      "num_tokens": 69291719.0,
      "step": 39945
    },
    {
      "entropy": 5.931534194946289,
      "epoch": 3.1082279712118264,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004041248167456408,
      "loss": 5.0741,
      "mean_token_accuracy": 0.19957407861948012,
      "num_tokens": 69300485.0,
      "step": 39950
    },
    {
      "entropy": 5.800767040252685,
      "epoch": 3.108617000583544,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004041021185361004,
      "loss": 5.0163,
      "mean_token_accuracy": 0.21054073423147202,
      "num_tokens": 69309565.0,
      "step": 39955
    },
    {
      "entropy": 5.889333868026734,
      "epoch": 3.1090060299552618,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004040794183676409,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20637501478195192,
      "num_tokens": 69317476.0,
      "step": 39960
    },
    {
      "entropy": 5.835377597808838,
      "epoch": 3.109395059326979,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004040567162406066,
      "loss": 4.9795,
      "mean_token_accuracy": 0.2079806849360466,
      "num_tokens": 69325639.0,
      "step": 39965
    },
    {
      "entropy": 5.877812242507934,
      "epoch": 3.1097840886986967,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004040340121553421,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20692819952964783,
      "num_tokens": 69334171.0,
      "step": 39970
    },
    {
      "entropy": 5.893484783172608,
      "epoch": 3.1101731180704144,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004040113061121918,
      "loss": 5.011,
      "mean_token_accuracy": 0.20641206949949265,
      "num_tokens": 69342246.0,
      "step": 39975
    },
    {
      "entropy": 5.883691263198853,
      "epoch": 3.110562147442132,
      "grad_norm": 1.453125,
      "learning_rate": 0.00040398859811150036,
      "loss": 5.0773,
      "mean_token_accuracy": 0.20161260664463043,
      "num_tokens": 69351153.0,
      "step": 39980
    },
    {
      "entropy": 5.915554857254028,
      "epoch": 3.1109511768138494,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004039658881536122,
      "loss": 4.9723,
      "mean_token_accuracy": 0.20937211364507674,
      "num_tokens": 69359367.0,
      "step": 39985
    },
    {
      "entropy": 5.889568519592285,
      "epoch": 3.111340206185567,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004039431762388719,
      "loss": 5.0734,
      "mean_token_accuracy": 0.2073922112584114,
      "num_tokens": 69367815.0,
      "step": 39990
    },
    {
      "entropy": 5.805947780609131,
      "epoch": 3.1117292355572848,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004039204623676242,
      "loss": 4.9173,
      "mean_token_accuracy": 0.21688165217638017,
      "num_tokens": 69376287.0,
      "step": 39995
    },
    {
      "entropy": 5.803370189666748,
      "epoch": 3.112118264929002,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040389774654021367,
      "loss": 4.9663,
      "mean_token_accuracy": 0.20872773975133896,
      "num_tokens": 69385153.0,
      "step": 40000
    },
    {
      "entropy": 5.806214761734009,
      "epoch": 3.1125072943007197,
      "grad_norm": 1.421875,
      "learning_rate": 0.000403875028756985,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2201690822839737,
      "num_tokens": 69393499.0,
      "step": 40005
    },
    {
      "entropy": 5.779525232315064,
      "epoch": 3.1128963236724374,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004038523090182829,
      "loss": 4.8692,
      "mean_token_accuracy": 0.21890356093645097,
      "num_tokens": 69401646.0,
      "step": 40010
    },
    {
      "entropy": 5.782726097106933,
      "epoch": 3.1132853530441547,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000403829587324452,
      "loss": 4.9218,
      "mean_token_accuracy": 0.20808209776878356,
      "num_tokens": 69410375.0,
      "step": 40015
    },
    {
      "entropy": 5.770151042938233,
      "epoch": 3.1136743824158724,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004038068636758373,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21252161711454393,
      "num_tokens": 69419092.0,
      "step": 40020
    },
    {
      "entropy": 5.863152980804443,
      "epoch": 3.11406341178759,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004037841380727834,
      "loss": 4.9914,
      "mean_token_accuracy": 0.2098824143409729,
      "num_tokens": 69427995.0,
      "step": 40025
    },
    {
      "entropy": 5.8368566036224365,
      "epoch": 3.1144524411593073,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004037614105156352,
      "loss": 4.9619,
      "mean_token_accuracy": 0.2097489833831787,
      "num_tokens": 69437598.0,
      "step": 40030
    },
    {
      "entropy": 5.790478181838989,
      "epoch": 3.114841470531025,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004037386810047376,
      "loss": 4.8481,
      "mean_token_accuracy": 0.21976030766963958,
      "num_tokens": 69445940.0,
      "step": 40035
    },
    {
      "entropy": 5.843204593658447,
      "epoch": 3.1152304999027427,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004037159495404354,
      "loss": 4.9866,
      "mean_token_accuracy": 0.2094413787126541,
      "num_tokens": 69454818.0,
      "step": 40040
    },
    {
      "entropy": 5.820031833648682,
      "epoch": 3.1156195292744604,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040369321612307343,
      "loss": 4.8711,
      "mean_token_accuracy": 0.2219756841659546,
      "num_tokens": 69463576.0,
      "step": 40045
    },
    {
      "entropy": 5.8234210968017575,
      "epoch": 3.1160085586461777,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040367048075299684,
      "loss": 4.9098,
      "mean_token_accuracy": 0.21445341259241105,
      "num_tokens": 69472178.0,
      "step": 40050
    },
    {
      "entropy": 5.801731729507447,
      "epoch": 3.1163975880178953,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004036477434305505,
      "loss": 4.929,
      "mean_token_accuracy": 0.21776547729969026,
      "num_tokens": 69480833.0,
      "step": 40055
    },
    {
      "entropy": 5.854745388031006,
      "epoch": 3.116786617389613,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004036250041560794,
      "loss": 5.0651,
      "mean_token_accuracy": 0.20434667617082597,
      "num_tokens": 69489803.0,
      "step": 40060
    },
    {
      "entropy": 5.836640357971191,
      "epoch": 3.1171756467613303,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004036022629299286,
      "loss": 4.9963,
      "mean_token_accuracy": 0.2055136740207672,
      "num_tokens": 69498718.0,
      "step": 40065
    },
    {
      "entropy": 5.787019157409668,
      "epoch": 3.117564676133048,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004035795197524431,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20869474560022355,
      "num_tokens": 69507817.0,
      "step": 40070
    },
    {
      "entropy": 5.761875486373901,
      "epoch": 3.1179537055047657,
      "grad_norm": 1.25,
      "learning_rate": 0.0004035567746239681,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2142111226916313,
      "num_tokens": 69516972.0,
      "step": 40075
    },
    {
      "entropy": 5.7798707485198975,
      "epoch": 3.1183427348764834,
      "grad_norm": 1.25,
      "learning_rate": 0.00040353402754484866,
      "loss": 4.9471,
      "mean_token_accuracy": 0.20742768347263335,
      "num_tokens": 69525454.0,
      "step": 40080
    },
    {
      "entropy": 5.812829923629761,
      "epoch": 3.1187317642482006,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004035112785154299,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2090970605611801,
      "num_tokens": 69534135.0,
      "step": 40085
    },
    {
      "entropy": 5.8310764789581295,
      "epoch": 3.1191207936199183,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00040348852753605707,
      "loss": 4.9976,
      "mean_token_accuracy": 0.20613928735256196,
      "num_tokens": 69542223.0,
      "step": 40090
    },
    {
      "entropy": 5.9026566505432125,
      "epoch": 3.119509822991636,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040346577460707535,
      "loss": 5.026,
      "mean_token_accuracy": 0.2035836935043335,
      "num_tokens": 69550708.0,
      "step": 40095
    },
    {
      "entropy": 5.8297196388244625,
      "epoch": 3.1198988523633533,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040344301972882997,
      "loss": 4.8858,
      "mean_token_accuracy": 0.22045668661594392,
      "num_tokens": 69559121.0,
      "step": 40100
    },
    {
      "entropy": 5.790639400482178,
      "epoch": 3.120287881735071,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004034202629016662,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2166861206293106,
      "num_tokens": 69567848.0,
      "step": 40105
    },
    {
      "entropy": 5.819609451293945,
      "epoch": 3.1206769111067887,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040339750412592937,
      "loss": 4.9774,
      "mean_token_accuracy": 0.21455069929361342,
      "num_tokens": 69577203.0,
      "step": 40110
    },
    {
      "entropy": 5.8763209819793705,
      "epoch": 3.121065940478506,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004033747434019647,
      "loss": 4.9737,
      "mean_token_accuracy": 0.20454567521810532,
      "num_tokens": 69585906.0,
      "step": 40115
    },
    {
      "entropy": 5.805637407302856,
      "epoch": 3.1214549698502236,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040335198073011765,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2043014094233513,
      "num_tokens": 69594089.0,
      "step": 40120
    },
    {
      "entropy": 5.814443159103393,
      "epoch": 3.1218439992219413,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004033292161107336,
      "loss": 5.0102,
      "mean_token_accuracy": 0.2048396185040474,
      "num_tokens": 69603592.0,
      "step": 40125
    },
    {
      "entropy": 5.778817844390869,
      "epoch": 3.122233028593659,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00040330644954415795,
      "loss": 4.8827,
      "mean_token_accuracy": 0.21359294205904006,
      "num_tokens": 69612159.0,
      "step": 40130
    },
    {
      "entropy": 5.824108552932739,
      "epoch": 3.1226220579653763,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004032836810307362,
      "loss": 4.9357,
      "mean_token_accuracy": 0.21542113572359084,
      "num_tokens": 69620459.0,
      "step": 40135
    },
    {
      "entropy": 5.881754207611084,
      "epoch": 3.123011087337094,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004032609105708137,
      "loss": 5.128,
      "mean_token_accuracy": 0.20173619091510772,
      "num_tokens": 69628629.0,
      "step": 40140
    },
    {
      "entropy": 5.776310014724731,
      "epoch": 3.1234001167088117,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040323813816473604,
      "loss": 4.9642,
      "mean_token_accuracy": 0.21074060052633287,
      "num_tokens": 69637932.0,
      "step": 40145
    },
    {
      "entropy": 5.821606397628784,
      "epoch": 3.123789146080529,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004032153638128488,
      "loss": 4.9006,
      "mean_token_accuracy": 0.21056067198514938,
      "num_tokens": 69646683.0,
      "step": 40150
    },
    {
      "entropy": 5.824298858642578,
      "epoch": 3.1241781754522466,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040319258751549743,
      "loss": 4.956,
      "mean_token_accuracy": 0.20730064958333969,
      "num_tokens": 69656030.0,
      "step": 40155
    },
    {
      "entropy": 5.818621730804443,
      "epoch": 3.1245672048239643,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040316980927302765,
      "loss": 4.9467,
      "mean_token_accuracy": 0.21067807525396348,
      "num_tokens": 69664862.0,
      "step": 40160
    },
    {
      "entropy": 5.858176279067993,
      "epoch": 3.1249562341956816,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00040314702908578496,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20013206154108049,
      "num_tokens": 69673149.0,
      "step": 40165
    },
    {
      "entropy": 5.822961235046387,
      "epoch": 3.1253452635673993,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004031242469541151,
      "loss": 4.9587,
      "mean_token_accuracy": 0.21044673323631286,
      "num_tokens": 69681734.0,
      "step": 40170
    },
    {
      "entropy": 5.779229354858399,
      "epoch": 3.125734292939117,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00040310146287836373,
      "loss": 4.939,
      "mean_token_accuracy": 0.2185847356915474,
      "num_tokens": 69690061.0,
      "step": 40175
    },
    {
      "entropy": 5.807046270370483,
      "epoch": 3.1261233223108347,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040307867685887653,
      "loss": 4.9297,
      "mean_token_accuracy": 0.2115536719560623,
      "num_tokens": 69697489.0,
      "step": 40180
    },
    {
      "entropy": 5.818188714981079,
      "epoch": 3.126512351682552,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040305588889599927,
      "loss": 4.917,
      "mean_token_accuracy": 0.21121968030929567,
      "num_tokens": 69705927.0,
      "step": 40185
    },
    {
      "entropy": 5.839116525650025,
      "epoch": 3.1269013810542696,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004030330989900777,
      "loss": 5.0659,
      "mean_token_accuracy": 0.2058378130197525,
      "num_tokens": 69714266.0,
      "step": 40190
    },
    {
      "entropy": 5.800013732910156,
      "epoch": 3.1272904104259873,
      "grad_norm": 1.375,
      "learning_rate": 0.00040301030714145764,
      "loss": 4.9727,
      "mean_token_accuracy": 0.2117864727973938,
      "num_tokens": 69723058.0,
      "step": 40195
    },
    {
      "entropy": 5.917342805862427,
      "epoch": 3.1276794397977046,
      "grad_norm": 1.453125,
      "learning_rate": 0.000402987513350485,
      "loss": 5.0766,
      "mean_token_accuracy": 0.20152245908975602,
      "num_tokens": 69732177.0,
      "step": 40200
    },
    {
      "entropy": 5.8905336380004885,
      "epoch": 3.1280684691694223,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004029647176175055,
      "loss": 5.005,
      "mean_token_accuracy": 0.20389846116304397,
      "num_tokens": 69740989.0,
      "step": 40205
    },
    {
      "entropy": 5.841110897064209,
      "epoch": 3.12845749854114,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040294191994286513,
      "loss": 4.9296,
      "mean_token_accuracy": 0.20669352263212204,
      "num_tokens": 69750167.0,
      "step": 40210
    },
    {
      "entropy": 5.8172996044158936,
      "epoch": 3.1288465279128572,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040291912032690963,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2121083542704582,
      "num_tokens": 69758893.0,
      "step": 40215
    },
    {
      "entropy": 5.822016763687134,
      "epoch": 3.129235557284575,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004028963187699852,
      "loss": 5.0064,
      "mean_token_accuracy": 0.20734311789274215,
      "num_tokens": 69767260.0,
      "step": 40220
    },
    {
      "entropy": 5.819591236114502,
      "epoch": 3.1296245866562926,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004028735152724377,
      "loss": 4.898,
      "mean_token_accuracy": 0.21299513578414916,
      "num_tokens": 69775445.0,
      "step": 40225
    },
    {
      "entropy": 5.810562658309936,
      "epoch": 3.1300136160280103,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040285070983461313,
      "loss": 4.9008,
      "mean_token_accuracy": 0.21200814545154573,
      "num_tokens": 69783349.0,
      "step": 40230
    },
    {
      "entropy": 5.850481605529785,
      "epoch": 3.1304026453997276,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040282790245685753,
      "loss": 5.0278,
      "mean_token_accuracy": 0.2077539950609207,
      "num_tokens": 69792133.0,
      "step": 40235
    },
    {
      "entropy": 5.8847472190856935,
      "epoch": 3.1307916747714453,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000402805093139517,
      "loss": 5.0607,
      "mean_token_accuracy": 0.20345744043588637,
      "num_tokens": 69800631.0,
      "step": 40240
    },
    {
      "entropy": 5.901767730712891,
      "epoch": 3.131180704143163,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004027822818829375,
      "loss": 5.0116,
      "mean_token_accuracy": 0.2094697266817093,
      "num_tokens": 69808587.0,
      "step": 40245
    },
    {
      "entropy": 5.8752203464508055,
      "epoch": 3.13156973351488,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004027594686874653,
      "loss": 4.9948,
      "mean_token_accuracy": 0.21248363703489304,
      "num_tokens": 69817623.0,
      "step": 40250
    },
    {
      "entropy": 5.795353698730469,
      "epoch": 3.131958762886598,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040273665355344655,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2137333258986473,
      "num_tokens": 69826917.0,
      "step": 40255
    },
    {
      "entropy": 5.803816270828247,
      "epoch": 3.1323477922583156,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004027138364812274,
      "loss": 4.9014,
      "mean_token_accuracy": 0.21781387478113173,
      "num_tokens": 69836117.0,
      "step": 40260
    },
    {
      "entropy": 5.785936975479126,
      "epoch": 3.132736821630033,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004026910174711541,
      "loss": 4.8887,
      "mean_token_accuracy": 0.21654711216688155,
      "num_tokens": 69843939.0,
      "step": 40265
    },
    {
      "entropy": 5.832336664199829,
      "epoch": 3.1331258510017506,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040266819652357266,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20243005603551864,
      "num_tokens": 69852295.0,
      "step": 40270
    },
    {
      "entropy": 5.82201042175293,
      "epoch": 3.1335148803734683,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040264537363882976,
      "loss": 5.051,
      "mean_token_accuracy": 0.20367805361747743,
      "num_tokens": 69860953.0,
      "step": 40275
    },
    {
      "entropy": 5.816664695739746,
      "epoch": 3.133903909745186,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040262254881727135,
      "loss": 4.9798,
      "mean_token_accuracy": 0.20768935829401017,
      "num_tokens": 69869894.0,
      "step": 40280
    },
    {
      "entropy": 5.774884748458862,
      "epoch": 3.134292939116903,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040259972205924395,
      "loss": 4.8761,
      "mean_token_accuracy": 0.21823182553052903,
      "num_tokens": 69878041.0,
      "step": 40285
    },
    {
      "entropy": 5.8078546047210695,
      "epoch": 3.134681968488621,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040257689336509385,
      "loss": 4.8628,
      "mean_token_accuracy": 0.2134632796049118,
      "num_tokens": 69886362.0,
      "step": 40290
    },
    {
      "entropy": 5.784028053283691,
      "epoch": 3.1350709978603386,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004025540627351675,
      "loss": 4.9311,
      "mean_token_accuracy": 0.21380906254053117,
      "num_tokens": 69895268.0,
      "step": 40295
    },
    {
      "entropy": 5.833011102676392,
      "epoch": 3.135460027232056,
      "grad_norm": 1.625,
      "learning_rate": 0.0004025312301698112,
      "loss": 4.9822,
      "mean_token_accuracy": 0.21301034986972808,
      "num_tokens": 69903947.0,
      "step": 40300
    },
    {
      "entropy": 5.791469669342041,
      "epoch": 3.1358490566037736,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004025083956693715,
      "loss": 4.9736,
      "mean_token_accuracy": 0.21207018941640854,
      "num_tokens": 69912979.0,
      "step": 40305
    },
    {
      "entropy": 5.8085626602172855,
      "epoch": 3.1362380859754913,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004024855592341948,
      "loss": 4.997,
      "mean_token_accuracy": 0.20559774935245514,
      "num_tokens": 69921586.0,
      "step": 40310
    },
    {
      "entropy": 5.853162431716919,
      "epoch": 3.136627115347209,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004024627208646277,
      "loss": 4.9913,
      "mean_token_accuracy": 0.21060082465410232,
      "num_tokens": 69930398.0,
      "step": 40315
    },
    {
      "entropy": 5.8178972721099855,
      "epoch": 3.137016144718926,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040243988056101667,
      "loss": 4.984,
      "mean_token_accuracy": 0.21199561059474945,
      "num_tokens": 69937910.0,
      "step": 40320
    },
    {
      "entropy": 5.787692546844482,
      "epoch": 3.137405174090644,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004024170383237083,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21200582534074783,
      "num_tokens": 69946740.0,
      "step": 40325
    },
    {
      "entropy": 5.775153589248657,
      "epoch": 3.1377942034623616,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004023941941530492,
      "loss": 4.9801,
      "mean_token_accuracy": 0.20861496925354003,
      "num_tokens": 69955998.0,
      "step": 40330
    },
    {
      "entropy": 5.837852954864502,
      "epoch": 3.138183232834079,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000402371348049386,
      "loss": 4.9974,
      "mean_token_accuracy": 0.20877424925565718,
      "num_tokens": 69964503.0,
      "step": 40335
    },
    {
      "entropy": 5.835185432434082,
      "epoch": 3.1385722622057965,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004023485000130653,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21645782440900802,
      "num_tokens": 69972814.0,
      "step": 40340
    },
    {
      "entropy": 5.765787124633789,
      "epoch": 3.1389612915775142,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040232565004443373,
      "loss": 4.8849,
      "mean_token_accuracy": 0.21246777474880219,
      "num_tokens": 69981891.0,
      "step": 40345
    },
    {
      "entropy": 5.821150875091552,
      "epoch": 3.1393503209492315,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004023027981438382,
      "loss": 5.0083,
      "mean_token_accuracy": 0.2026303917169571,
      "num_tokens": 69990596.0,
      "step": 40350
    },
    {
      "entropy": 5.804481840133667,
      "epoch": 3.139739350320949,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040227994431162517,
      "loss": 4.9416,
      "mean_token_accuracy": 0.2177239954471588,
      "num_tokens": 69999369.0,
      "step": 40355
    },
    {
      "entropy": 5.761489820480347,
      "epoch": 3.140128379692667,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040225708854814174,
      "loss": 4.9412,
      "mean_token_accuracy": 0.20815413892269136,
      "num_tokens": 70008290.0,
      "step": 40360
    },
    {
      "entropy": 5.869994497299194,
      "epoch": 3.140517409064384,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004022342308537345,
      "loss": 5.0514,
      "mean_token_accuracy": 0.2080064058303833,
      "num_tokens": 70016730.0,
      "step": 40365
    },
    {
      "entropy": 5.740296411514282,
      "epoch": 3.140906438436102,
      "grad_norm": 1.375,
      "learning_rate": 0.0004022113712287503,
      "loss": 4.9678,
      "mean_token_accuracy": 0.2100544676184654,
      "num_tokens": 70025115.0,
      "step": 40370
    },
    {
      "entropy": 5.821581125259399,
      "epoch": 3.1412954678078195,
      "grad_norm": 1.25,
      "learning_rate": 0.000402188509673536,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21034393906593324,
      "num_tokens": 70033399.0,
      "step": 40375
    },
    {
      "entropy": 5.856101703643799,
      "epoch": 3.1416844971795372,
      "grad_norm": 1.375,
      "learning_rate": 0.00040216564618843854,
      "loss": 4.9734,
      "mean_token_accuracy": 0.21026332974433898,
      "num_tokens": 70041756.0,
      "step": 40380
    },
    {
      "entropy": 5.816487884521484,
      "epoch": 3.1420735265512545,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004021427807738048,
      "loss": 4.8792,
      "mean_token_accuracy": 0.2189151644706726,
      "num_tokens": 70050457.0,
      "step": 40385
    },
    {
      "entropy": 5.844017362594604,
      "epoch": 3.142462555922972,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004021199134299817,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20541880279779434,
      "num_tokens": 70059176.0,
      "step": 40390
    },
    {
      "entropy": 5.793991994857788,
      "epoch": 3.14285158529469,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040209704415731623,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21523598730564117,
      "num_tokens": 70067882.0,
      "step": 40395
    },
    {
      "entropy": 5.785449600219726,
      "epoch": 3.143240614666407,
      "grad_norm": 1.375,
      "learning_rate": 0.00040207417295615546,
      "loss": 4.8998,
      "mean_token_accuracy": 0.21717049032449723,
      "num_tokens": 70076295.0,
      "step": 40400
    },
    {
      "entropy": 5.815502738952636,
      "epoch": 3.143629644038125,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040205129982684635,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21539311856031418,
      "num_tokens": 70085911.0,
      "step": 40405
    },
    {
      "entropy": 5.822138881683349,
      "epoch": 3.1440186734098425,
      "grad_norm": 1.453125,
      "learning_rate": 0.00040202842476973606,
      "loss": 5.0026,
      "mean_token_accuracy": 0.20450359135866164,
      "num_tokens": 70094518.0,
      "step": 40410
    },
    {
      "entropy": 5.847254705429077,
      "epoch": 3.1444077027815602,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040200554778517154,
      "loss": 5.0472,
      "mean_token_accuracy": 0.20179277658462524,
      "num_tokens": 70103628.0,
      "step": 40415
    },
    {
      "entropy": 5.875411939620972,
      "epoch": 3.1447967321532775,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040198266887349994,
      "loss": 4.9786,
      "mean_token_accuracy": 0.20918790698051454,
      "num_tokens": 70111974.0,
      "step": 40420
    },
    {
      "entropy": 5.865532970428466,
      "epoch": 3.145185761524995,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040195978803506853,
      "loss": 5.0119,
      "mean_token_accuracy": 0.20579386204481126,
      "num_tokens": 70121077.0,
      "step": 40425
    },
    {
      "entropy": 5.776202249526977,
      "epoch": 3.145574790896713,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040193690527022443,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2185719221830368,
      "num_tokens": 70129852.0,
      "step": 40430
    },
    {
      "entropy": 5.827106761932373,
      "epoch": 3.14596382026843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040191402057931474,
      "loss": 4.99,
      "mean_token_accuracy": 0.20987142473459244,
      "num_tokens": 70138285.0,
      "step": 40435
    },
    {
      "entropy": 5.925323390960694,
      "epoch": 3.146352849640148,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004018911339626869,
      "loss": 5.0166,
      "mean_token_accuracy": 0.2007739469408989,
      "num_tokens": 70147092.0,
      "step": 40440
    },
    {
      "entropy": 5.810503101348877,
      "epoch": 3.1467418790118655,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000401868245420688,
      "loss": 4.917,
      "mean_token_accuracy": 0.21432065963745117,
      "num_tokens": 70155663.0,
      "step": 40445
    },
    {
      "entropy": 5.770821142196655,
      "epoch": 3.147130908383583,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040184535495366543,
      "loss": 4.9216,
      "mean_token_accuracy": 0.21857451796531677,
      "num_tokens": 70163954.0,
      "step": 40450
    },
    {
      "entropy": 5.846753311157227,
      "epoch": 3.1475199377553005,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004018224625619664,
      "loss": 5.0275,
      "mean_token_accuracy": 0.20230946093797683,
      "num_tokens": 70173103.0,
      "step": 40455
    },
    {
      "entropy": 5.827665042877197,
      "epoch": 3.147908967127018,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040179956824593847,
      "loss": 5.0165,
      "mean_token_accuracy": 0.2041011556982994,
      "num_tokens": 70181821.0,
      "step": 40460
    },
    {
      "entropy": 5.8914872169494625,
      "epoch": 3.1482979964987354,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040177667200592894,
      "loss": 5.0156,
      "mean_token_accuracy": 0.2071906343102455,
      "num_tokens": 70190257.0,
      "step": 40465
    },
    {
      "entropy": 5.890468740463257,
      "epoch": 3.148687025870453,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004017537738422852,
      "loss": 5.0295,
      "mean_token_accuracy": 0.2050318956375122,
      "num_tokens": 70199487.0,
      "step": 40470
    },
    {
      "entropy": 5.741133689880371,
      "epoch": 3.149076055242171,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040173087375535457,
      "loss": 4.8825,
      "mean_token_accuracy": 0.21920528262853622,
      "num_tokens": 70207629.0,
      "step": 40475
    },
    {
      "entropy": 5.837794971466065,
      "epoch": 3.1494650846138885,
      "grad_norm": 1.421875,
      "learning_rate": 0.00040170797174548476,
      "loss": 4.9881,
      "mean_token_accuracy": 0.21289512366056443,
      "num_tokens": 70215768.0,
      "step": 40480
    },
    {
      "entropy": 5.8833390235900875,
      "epoch": 3.1498541139856058,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00040168506781302303,
      "loss": 4.9975,
      "mean_token_accuracy": 0.20677034705877304,
      "num_tokens": 70223570.0,
      "step": 40485
    },
    {
      "entropy": 5.7176148891448975,
      "epoch": 3.1502431433573235,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004016621619583171,
      "loss": 4.8545,
      "mean_token_accuracy": 0.21594338566064836,
      "num_tokens": 70233032.0,
      "step": 40490
    },
    {
      "entropy": 5.792489099502563,
      "epoch": 3.150632172729041,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040163925418171454,
      "loss": 4.8917,
      "mean_token_accuracy": 0.21792149245738984,
      "num_tokens": 70241091.0,
      "step": 40495
    },
    {
      "entropy": 5.8299891471862795,
      "epoch": 3.1510212021007584,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004016163444835628,
      "loss": 5.0735,
      "mean_token_accuracy": 0.19838575273752213,
      "num_tokens": 70249595.0,
      "step": 40500
    },
    {
      "entropy": 5.87692403793335,
      "epoch": 3.151410231472476,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004015934328642096,
      "loss": 5.0013,
      "mean_token_accuracy": 0.2075623616576195,
      "num_tokens": 70258376.0,
      "step": 40505
    },
    {
      "entropy": 5.849141407012939,
      "epoch": 3.151799260844194,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040157051932400244,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2094455197453499,
      "num_tokens": 70266501.0,
      "step": 40510
    },
    {
      "entropy": 5.834272050857544,
      "epoch": 3.1521882902159115,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004015476038632892,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20340104252099991,
      "num_tokens": 70274029.0,
      "step": 40515
    },
    {
      "entropy": 5.789268112182617,
      "epoch": 3.1525773195876288,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004015246864824175,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20911374241113662,
      "num_tokens": 70282608.0,
      "step": 40520
    },
    {
      "entropy": 5.770678329467773,
      "epoch": 3.1529663489593465,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040150176718173504,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20731696039438247,
      "num_tokens": 70291350.0,
      "step": 40525
    },
    {
      "entropy": 5.854278182983398,
      "epoch": 3.153355378331064,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004014788459615896,
      "loss": 4.9384,
      "mean_token_accuracy": 0.21519835740327836,
      "num_tokens": 70299883.0,
      "step": 40530
    },
    {
      "entropy": 5.845732688903809,
      "epoch": 3.1537444077027814,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004014559228223291,
      "loss": 4.9471,
      "mean_token_accuracy": 0.20714075416326522,
      "num_tokens": 70308803.0,
      "step": 40535
    },
    {
      "entropy": 5.876364803314209,
      "epoch": 3.154133437074499,
      "grad_norm": 1.375,
      "learning_rate": 0.00040143299776430115,
      "loss": 4.9832,
      "mean_token_accuracy": 0.21998822391033174,
      "num_tokens": 70317349.0,
      "step": 40540
    },
    {
      "entropy": 5.8057541847229,
      "epoch": 3.154522466446217,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004014100707878536,
      "loss": 4.9088,
      "mean_token_accuracy": 0.2139161556959152,
      "num_tokens": 70326199.0,
      "step": 40545
    },
    {
      "entropy": 5.813912200927734,
      "epoch": 3.154911495817934,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040138714189333455,
      "loss": 4.9899,
      "mean_token_accuracy": 0.20420707762241364,
      "num_tokens": 70335155.0,
      "step": 40550
    },
    {
      "entropy": 5.845198154449463,
      "epoch": 3.1553005251896518,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040136421108109176,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2106250450015068,
      "num_tokens": 70343127.0,
      "step": 40555
    },
    {
      "entropy": 5.856299638748169,
      "epoch": 3.1556895545613695,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004013412783514732,
      "loss": 5.0155,
      "mean_token_accuracy": 0.2048309862613678,
      "num_tokens": 70352684.0,
      "step": 40560
    },
    {
      "entropy": 5.824615001678467,
      "epoch": 3.156078583933087,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004013183437048268,
      "loss": 4.9102,
      "mean_token_accuracy": 0.21197241842746734,
      "num_tokens": 70361050.0,
      "step": 40565
    },
    {
      "entropy": 5.932951498031616,
      "epoch": 3.1564676133048044,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040129540714150066,
      "loss": 5.1325,
      "mean_token_accuracy": 0.19702691733837127,
      "num_tokens": 70371350.0,
      "step": 40570
    },
    {
      "entropy": 5.941982126235962,
      "epoch": 3.156856642676522,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004012724686618427,
      "loss": 5.0916,
      "mean_token_accuracy": 0.20024094432592393,
      "num_tokens": 70380806.0,
      "step": 40575
    },
    {
      "entropy": 5.874399948120117,
      "epoch": 3.15724567204824,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000401249528266201,
      "loss": 4.9969,
      "mean_token_accuracy": 0.20955812335014343,
      "num_tokens": 70390142.0,
      "step": 40580
    },
    {
      "entropy": 5.88062891960144,
      "epoch": 3.157634701419957,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040122658595492366,
      "loss": 5.0076,
      "mean_token_accuracy": 0.2118940085172653,
      "num_tokens": 70398345.0,
      "step": 40585
    },
    {
      "entropy": 5.81641206741333,
      "epoch": 3.1580237307916748,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004012036417283588,
      "loss": 4.9934,
      "mean_token_accuracy": 0.2095988377928734,
      "num_tokens": 70407107.0,
      "step": 40590
    },
    {
      "entropy": 5.806016540527343,
      "epoch": 3.1584127601633925,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040118069558685456,
      "loss": 4.9411,
      "mean_token_accuracy": 0.20756861567497253,
      "num_tokens": 70415053.0,
      "step": 40595
    },
    {
      "entropy": 5.798699235916137,
      "epoch": 3.1588017895351097,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040115774753075905,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21764702945947648,
      "num_tokens": 70423605.0,
      "step": 40600
    },
    {
      "entropy": 5.854818630218506,
      "epoch": 3.1591908189068274,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004011347975604206,
      "loss": 4.9954,
      "mean_token_accuracy": 0.20296927988529206,
      "num_tokens": 70432155.0,
      "step": 40605
    },
    {
      "entropy": 5.915959024429322,
      "epoch": 3.159579848278545,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004011118456761873,
      "loss": 5.0323,
      "mean_token_accuracy": 0.20209593772888185,
      "num_tokens": 70441349.0,
      "step": 40610
    },
    {
      "entropy": 5.794466209411621,
      "epoch": 3.159968877650263,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004010888918784075,
      "loss": 4.9881,
      "mean_token_accuracy": 0.20838527679443358,
      "num_tokens": 70449792.0,
      "step": 40615
    },
    {
      "entropy": 5.784317588806152,
      "epoch": 3.16035790702198,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004010659361674294,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21568038016557695,
      "num_tokens": 70457699.0,
      "step": 40620
    },
    {
      "entropy": 5.762719821929932,
      "epoch": 3.1607469363936977,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00040104297854360145,
      "loss": 4.9167,
      "mean_token_accuracy": 0.2138345345854759,
      "num_tokens": 70465704.0,
      "step": 40625
    },
    {
      "entropy": 5.763383913040161,
      "epoch": 3.1611359657654154,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004010200190072719,
      "loss": 4.9513,
      "mean_token_accuracy": 0.20991896688938141,
      "num_tokens": 70474462.0,
      "step": 40630
    },
    {
      "entropy": 5.861641931533813,
      "epoch": 3.1615249951371327,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004009970575587891,
      "loss": 4.9417,
      "mean_token_accuracy": 0.2092234268784523,
      "num_tokens": 70482957.0,
      "step": 40635
    },
    {
      "entropy": 5.947053527832031,
      "epoch": 3.1619140245088504,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004009740941985016,
      "loss": 5.085,
      "mean_token_accuracy": 0.20078970938920976,
      "num_tokens": 70492236.0,
      "step": 40640
    },
    {
      "entropy": 5.782573652267456,
      "epoch": 3.162303053880568,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004009511289267576,
      "loss": 4.9455,
      "mean_token_accuracy": 0.20875199586153032,
      "num_tokens": 70501457.0,
      "step": 40645
    },
    {
      "entropy": 5.808731985092163,
      "epoch": 3.1626920832522853,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004009281617439058,
      "loss": 4.9476,
      "mean_token_accuracy": 0.21594176441431046,
      "num_tokens": 70510444.0,
      "step": 40650
    },
    {
      "entropy": 5.817128658294678,
      "epoch": 3.163081112624003,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004009051926502945,
      "loss": 4.9522,
      "mean_token_accuracy": 0.20876814275979996,
      "num_tokens": 70518823.0,
      "step": 40655
    },
    {
      "entropy": 5.8698302745819095,
      "epoch": 3.1634701419957207,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040088222164627233,
      "loss": 4.9121,
      "mean_token_accuracy": 0.2229241505265236,
      "num_tokens": 70527177.0,
      "step": 40660
    },
    {
      "entropy": 5.897339487075806,
      "epoch": 3.1638591713674384,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040085924873218783,
      "loss": 5.0623,
      "mean_token_accuracy": 0.20917700082063675,
      "num_tokens": 70535749.0,
      "step": 40665
    },
    {
      "entropy": 5.7932007789611815,
      "epoch": 3.1642482007391557,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004008362739083895,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21103404015302657,
      "num_tokens": 70544612.0,
      "step": 40670
    },
    {
      "entropy": 5.834151792526245,
      "epoch": 3.1646372301108734,
      "grad_norm": 1.359375,
      "learning_rate": 0.000400813297175226,
      "loss": 4.9225,
      "mean_token_accuracy": 0.21026256680488586,
      "num_tokens": 70553079.0,
      "step": 40675
    },
    {
      "entropy": 5.842628526687622,
      "epoch": 3.165026259482591,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000400790318533046,
      "loss": 4.9781,
      "mean_token_accuracy": 0.21061391830444337,
      "num_tokens": 70561798.0,
      "step": 40680
    },
    {
      "entropy": 5.789656782150269,
      "epoch": 3.1654152888543083,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040076733798219817,
      "loss": 4.9892,
      "mean_token_accuracy": 0.20368519127368928,
      "num_tokens": 70570738.0,
      "step": 40685
    },
    {
      "entropy": 5.816572761535644,
      "epoch": 3.165804318226026,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004007443555230312,
      "loss": 4.9495,
      "mean_token_accuracy": 0.20662213414907454,
      "num_tokens": 70579324.0,
      "step": 40690
    },
    {
      "entropy": 5.871837711334228,
      "epoch": 3.1661933475977437,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040072137115589366,
      "loss": 4.9813,
      "mean_token_accuracy": 0.2075667753815651,
      "num_tokens": 70588462.0,
      "step": 40695
    },
    {
      "entropy": 5.786119651794434,
      "epoch": 3.166582376969461,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004006983848811345,
      "loss": 4.9124,
      "mean_token_accuracy": 0.21691821813583373,
      "num_tokens": 70596579.0,
      "step": 40700
    },
    {
      "entropy": 5.8506272315979,
      "epoch": 3.1669714063411787,
      "grad_norm": 1.40625,
      "learning_rate": 0.00040067539669910235,
      "loss": 4.9798,
      "mean_token_accuracy": 0.2085099160671234,
      "num_tokens": 70604516.0,
      "step": 40705
    },
    {
      "entropy": 5.847724342346192,
      "epoch": 3.1673604357128964,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004006524066101461,
      "loss": 5.0225,
      "mean_token_accuracy": 0.20602555423974991,
      "num_tokens": 70612263.0,
      "step": 40710
    },
    {
      "entropy": 5.714101982116699,
      "epoch": 3.167749465084614,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004006294146146146,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22389937490224837,
      "num_tokens": 70621246.0,
      "step": 40715
    },
    {
      "entropy": 5.80243992805481,
      "epoch": 3.1681384944563313,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004006064207128567,
      "loss": 4.9765,
      "mean_token_accuracy": 0.20752858072519303,
      "num_tokens": 70629935.0,
      "step": 40720
    },
    {
      "entropy": 5.907432126998901,
      "epoch": 3.168527523828049,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040058342490522134,
      "loss": 5.0339,
      "mean_token_accuracy": 0.19860286265611649,
      "num_tokens": 70638722.0,
      "step": 40725
    },
    {
      "entropy": 5.841650342941284,
      "epoch": 3.1689165531997667,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004005604271920573,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20609549880027772,
      "num_tokens": 70647835.0,
      "step": 40730
    },
    {
      "entropy": 5.8506042003631595,
      "epoch": 3.169305582571484,
      "grad_norm": 1.25,
      "learning_rate": 0.00040053742757371365,
      "loss": 4.9718,
      "mean_token_accuracy": 0.21097930371761323,
      "num_tokens": 70656668.0,
      "step": 40735
    },
    {
      "entropy": 5.888624382019043,
      "epoch": 3.1696946119432017,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004005144260505394,
      "loss": 4.9935,
      "mean_token_accuracy": 0.2065402314066887,
      "num_tokens": 70664921.0,
      "step": 40740
    },
    {
      "entropy": 5.86773829460144,
      "epoch": 3.1700836413149194,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004004914226228834,
      "loss": 4.9787,
      "mean_token_accuracy": 0.21195277571678162,
      "num_tokens": 70673435.0,
      "step": 40745
    },
    {
      "entropy": 5.766985750198364,
      "epoch": 3.170472670686637,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004004684172910949,
      "loss": 4.93,
      "mean_token_accuracy": 0.21306199580430984,
      "num_tokens": 70682102.0,
      "step": 40750
    },
    {
      "entropy": 5.7209718227386475,
      "epoch": 3.1708617000583543,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004004454100555229,
      "loss": 4.8836,
      "mean_token_accuracy": 0.21658333241939545,
      "num_tokens": 70690356.0,
      "step": 40755
    },
    {
      "entropy": 5.814652156829834,
      "epoch": 3.171250729430072,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040042240091651645,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21472711265087127,
      "num_tokens": 70698966.0,
      "step": 40760
    },
    {
      "entropy": 5.77323727607727,
      "epoch": 3.1716397588017897,
      "grad_norm": 1.421875,
      "learning_rate": 0.00040039938987442464,
      "loss": 4.9173,
      "mean_token_accuracy": 0.21694935411214827,
      "num_tokens": 70707144.0,
      "step": 40765
    },
    {
      "entropy": 5.801467037200927,
      "epoch": 3.172028788173507,
      "grad_norm": 1.25,
      "learning_rate": 0.00040037637692959667,
      "loss": 4.9629,
      "mean_token_accuracy": 0.21146585792303085,
      "num_tokens": 70716474.0,
      "step": 40770
    },
    {
      "entropy": 5.773419046401978,
      "epoch": 3.1724178175452247,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040035336208238176,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21132237315177918,
      "num_tokens": 70725458.0,
      "step": 40775
    },
    {
      "entropy": 5.861338186264038,
      "epoch": 3.1728068469169424,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040033034533312913,
      "loss": 5.0277,
      "mean_token_accuracy": 0.20323751866817474,
      "num_tokens": 70734828.0,
      "step": 40780
    },
    {
      "entropy": 5.907739305496216,
      "epoch": 3.1731958762886596,
      "grad_norm": 1.296875,
      "learning_rate": 0.000400307326682188,
      "loss": 5.0783,
      "mean_token_accuracy": 0.20688480585813523,
      "num_tokens": 70743229.0,
      "step": 40785
    },
    {
      "entropy": 5.871409273147583,
      "epoch": 3.1735849056603773,
      "grad_norm": 1.5625,
      "learning_rate": 0.00040028430612990755,
      "loss": 5.0286,
      "mean_token_accuracy": 0.2136978968977928,
      "num_tokens": 70751873.0,
      "step": 40790
    },
    {
      "entropy": 5.847549486160278,
      "epoch": 3.173973935032095,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040026128367663726,
      "loss": 5.0068,
      "mean_token_accuracy": 0.2006261706352234,
      "num_tokens": 70760617.0,
      "step": 40795
    },
    {
      "entropy": 5.847450304031372,
      "epoch": 3.1743629644038123,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040023825932272635,
      "loss": 4.9902,
      "mean_token_accuracy": 0.20568694919347763,
      "num_tokens": 70769167.0,
      "step": 40800
    },
    {
      "entropy": 5.756580495834351,
      "epoch": 3.17475199377553,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040021523306852416,
      "loss": 4.956,
      "mean_token_accuracy": 0.21337173730134965,
      "num_tokens": 70778890.0,
      "step": 40805
    },
    {
      "entropy": 5.856593799591065,
      "epoch": 3.1751410231472477,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040019220491438007,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20926121324300767,
      "num_tokens": 70787557.0,
      "step": 40810
    },
    {
      "entropy": 5.79703311920166,
      "epoch": 3.1755300525189654,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004001691748606436,
      "loss": 4.8802,
      "mean_token_accuracy": 0.2142474889755249,
      "num_tokens": 70795751.0,
      "step": 40815
    },
    {
      "entropy": 5.830465412139892,
      "epoch": 3.1759190818906826,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004001461429076641,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21754734516143798,
      "num_tokens": 70803938.0,
      "step": 40820
    },
    {
      "entropy": 5.845277404785156,
      "epoch": 3.1763081112624003,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004001231090557912,
      "loss": 4.9702,
      "mean_token_accuracy": 0.21298433542251588,
      "num_tokens": 70812196.0,
      "step": 40825
    },
    {
      "entropy": 5.832943248748779,
      "epoch": 3.176697140634118,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040010007330537405,
      "loss": 5.0108,
      "mean_token_accuracy": 0.20421263724565505,
      "num_tokens": 70821118.0,
      "step": 40830
    },
    {
      "entropy": 5.757676124572754,
      "epoch": 3.1770861700058353,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040007703565676263,
      "loss": 4.9051,
      "mean_token_accuracy": 0.20895315408706666,
      "num_tokens": 70829785.0,
      "step": 40835
    },
    {
      "entropy": 5.833765459060669,
      "epoch": 3.177475199377553,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040005399611030615,
      "loss": 4.93,
      "mean_token_accuracy": 0.21874927580356598,
      "num_tokens": 70838054.0,
      "step": 40840
    },
    {
      "entropy": 5.881453371047973,
      "epoch": 3.1778642287492707,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004000309546663543,
      "loss": 5.0425,
      "mean_token_accuracy": 0.2061159536242485,
      "num_tokens": 70846867.0,
      "step": 40845
    },
    {
      "entropy": 5.839722394943237,
      "epoch": 3.1782532581209884,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040000791132525676,
      "loss": 5.0042,
      "mean_token_accuracy": 0.2171002969145775,
      "num_tokens": 70855369.0,
      "step": 40850
    },
    {
      "entropy": 5.878963756561279,
      "epoch": 3.1786422874927056,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039998486608736305,
      "loss": 4.9794,
      "mean_token_accuracy": 0.20918461233377456,
      "num_tokens": 70863888.0,
      "step": 40855
    },
    {
      "entropy": 5.776246452331543,
      "epoch": 3.1790313168644233,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003999618189530231,
      "loss": 4.9788,
      "mean_token_accuracy": 0.2117355763912201,
      "num_tokens": 70872133.0,
      "step": 40860
    },
    {
      "entropy": 5.731360387802124,
      "epoch": 3.179420346236141,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003999387699225863,
      "loss": 4.8985,
      "mean_token_accuracy": 0.2181678369641304,
      "num_tokens": 70880691.0,
      "step": 40865
    },
    {
      "entropy": 5.893551158905029,
      "epoch": 3.1798093756078583,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003999157189964026,
      "loss": 5.0156,
      "mean_token_accuracy": 0.2124675914645195,
      "num_tokens": 70888552.0,
      "step": 40870
    },
    {
      "entropy": 5.855310392379761,
      "epoch": 3.180198404979576,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003998926661748217,
      "loss": 4.9501,
      "mean_token_accuracy": 0.21418712735176088,
      "num_tokens": 70897054.0,
      "step": 40875
    },
    {
      "entropy": 5.890038299560547,
      "epoch": 3.1805874343512937,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003998696114581933,
      "loss": 5.0965,
      "mean_token_accuracy": 0.20737634897232055,
      "num_tokens": 70905783.0,
      "step": 40880
    },
    {
      "entropy": 5.866379880905152,
      "epoch": 3.180976463723011,
      "grad_norm": 1.25,
      "learning_rate": 0.0003998465548468674,
      "loss": 5.0224,
      "mean_token_accuracy": 0.20334049463272094,
      "num_tokens": 70914424.0,
      "step": 40885
    },
    {
      "entropy": 5.852714729309082,
      "epoch": 3.1813654930947286,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003998234963411936,
      "loss": 4.9668,
      "mean_token_accuracy": 0.21362429857254028,
      "num_tokens": 70923571.0,
      "step": 40890
    },
    {
      "entropy": 5.807128667831421,
      "epoch": 3.1817545224664463,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003998004359415221,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21299878805875777,
      "num_tokens": 70931969.0,
      "step": 40895
    },
    {
      "entropy": 5.7305878639221195,
      "epoch": 3.1821435518381636,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039977737364820247,
      "loss": 4.8066,
      "mean_token_accuracy": 0.21640298664569854,
      "num_tokens": 70940064.0,
      "step": 40900
    },
    {
      "entropy": 5.7752539157867435,
      "epoch": 3.1825325812098813,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039975430946158493,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20028969198465346,
      "num_tokens": 70950429.0,
      "step": 40905
    },
    {
      "entropy": 5.797454166412353,
      "epoch": 3.182921610581599,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003997312433820192,
      "loss": 4.9403,
      "mean_token_accuracy": 0.21516282260417938,
      "num_tokens": 70959441.0,
      "step": 40910
    },
    {
      "entropy": 5.832349348068237,
      "epoch": 3.1833106399533166,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039970817540985545,
      "loss": 4.9487,
      "mean_token_accuracy": 0.21309252977371215,
      "num_tokens": 70967790.0,
      "step": 40915
    },
    {
      "entropy": 5.847241067886353,
      "epoch": 3.183699669325034,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003996851055454436,
      "loss": 4.9866,
      "mean_token_accuracy": 0.21228308826684952,
      "num_tokens": 70976851.0,
      "step": 40920
    },
    {
      "entropy": 5.920950412750244,
      "epoch": 3.1840886986967516,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003996620337891337,
      "loss": 5.1305,
      "mean_token_accuracy": 0.19785796403884887,
      "num_tokens": 70985452.0,
      "step": 40925
    },
    {
      "entropy": 5.881191682815552,
      "epoch": 3.1844777280684693,
      "grad_norm": 1.1875,
      "learning_rate": 0.000399638960141276,
      "loss": 4.9542,
      "mean_token_accuracy": 0.20491700917482375,
      "num_tokens": 70994597.0,
      "step": 40930
    },
    {
      "entropy": 5.867742109298706,
      "epoch": 3.1848667574401865,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039961588460222033,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20652139633893968,
      "num_tokens": 71003642.0,
      "step": 40935
    },
    {
      "entropy": 5.796478652954102,
      "epoch": 3.1852557868119042,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000399592807172317,
      "loss": 4.9486,
      "mean_token_accuracy": 0.21536874175071716,
      "num_tokens": 71012618.0,
      "step": 40940
    },
    {
      "entropy": 5.809241056442261,
      "epoch": 3.185644816183622,
      "grad_norm": 1.453125,
      "learning_rate": 0.00039956972785191606,
      "loss": 4.9622,
      "mean_token_accuracy": 0.20682704746723174,
      "num_tokens": 71021075.0,
      "step": 40945
    },
    {
      "entropy": 5.886703062057495,
      "epoch": 3.1860338455553396,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039954664664136787,
      "loss": 5.0117,
      "mean_token_accuracy": 0.20710003674030303,
      "num_tokens": 71029373.0,
      "step": 40950
    },
    {
      "entropy": 5.732402992248535,
      "epoch": 3.186422874927057,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003995235635410225,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21133046597242355,
      "num_tokens": 71037827.0,
      "step": 40955
    },
    {
      "entropy": 5.888251876831054,
      "epoch": 3.1868119042987746,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003995004785512302,
      "loss": 5.0272,
      "mean_token_accuracy": 0.20337484627962113,
      "num_tokens": 71046097.0,
      "step": 40960
    },
    {
      "entropy": 5.877272319793701,
      "epoch": 3.1872009336704923,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003994773916723414,
      "loss": 4.9877,
      "mean_token_accuracy": 0.2125300496816635,
      "num_tokens": 71055340.0,
      "step": 40965
    },
    {
      "entropy": 5.823552846908569,
      "epoch": 3.1875899630422095,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003994543029047063,
      "loss": 4.9345,
      "mean_token_accuracy": 0.2175980404019356,
      "num_tokens": 71063967.0,
      "step": 40970
    },
    {
      "entropy": 5.802626037597657,
      "epoch": 3.1879789924139272,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003994312122486752,
      "loss": 4.9748,
      "mean_token_accuracy": 0.21141192466020584,
      "num_tokens": 71072813.0,
      "step": 40975
    },
    {
      "entropy": 5.744529438018799,
      "epoch": 3.188368021785645,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003994081197045985,
      "loss": 4.8858,
      "mean_token_accuracy": 0.220436954498291,
      "num_tokens": 71081801.0,
      "step": 40980
    },
    {
      "entropy": 5.817735242843628,
      "epoch": 3.188757051157362,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003993850252728267,
      "loss": 4.8747,
      "mean_token_accuracy": 0.21554209142923356,
      "num_tokens": 71089784.0,
      "step": 40985
    },
    {
      "entropy": 5.828818607330322,
      "epoch": 3.18914608052908,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039936192895371006,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20684264749288558,
      "num_tokens": 71099415.0,
      "step": 40990
    },
    {
      "entropy": 5.8362205028533936,
      "epoch": 3.1895351099007976,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000399338830747599,
      "loss": 5.022,
      "mean_token_accuracy": 0.2033639669418335,
      "num_tokens": 71108073.0,
      "step": 40995
    },
    {
      "entropy": 5.93585433959961,
      "epoch": 3.1899241392725153,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003993157306548442,
      "loss": 5.0899,
      "mean_token_accuracy": 0.20103107988834382,
      "num_tokens": 71116755.0,
      "step": 41000
    },
    {
      "entropy": 5.852106046676636,
      "epoch": 3.1903131686442325,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039929262867579603,
      "loss": 5.0268,
      "mean_token_accuracy": 0.21021128594875335,
      "num_tokens": 71125278.0,
      "step": 41005
    },
    {
      "entropy": 5.85308952331543,
      "epoch": 3.1907021980159502,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039926952481080495,
      "loss": 4.9489,
      "mean_token_accuracy": 0.20986337959766388,
      "num_tokens": 71134111.0,
      "step": 41010
    },
    {
      "entropy": 5.870548057556152,
      "epoch": 3.191091227387668,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039924641906022176,
      "loss": 5.027,
      "mean_token_accuracy": 0.20331714749336244,
      "num_tokens": 71142966.0,
      "step": 41015
    },
    {
      "entropy": 5.830940914154053,
      "epoch": 3.191480256759385,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003992233114243969,
      "loss": 5.0133,
      "mean_token_accuracy": 0.20037883073091506,
      "num_tokens": 71151461.0,
      "step": 41020
    },
    {
      "entropy": 5.877682304382324,
      "epoch": 3.191869286131103,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000399200201903681,
      "loss": 5.055,
      "mean_token_accuracy": 0.2087266519665718,
      "num_tokens": 71160133.0,
      "step": 41025
    },
    {
      "entropy": 5.892140531539917,
      "epoch": 3.1922583155028206,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003991770904984247,
      "loss": 5.1099,
      "mean_token_accuracy": 0.2000988468527794,
      "num_tokens": 71168795.0,
      "step": 41030
    },
    {
      "entropy": 5.7832756519317625,
      "epoch": 3.192647344874538,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003991539772089787,
      "loss": 4.8988,
      "mean_token_accuracy": 0.21551756262779237,
      "num_tokens": 71177625.0,
      "step": 41035
    },
    {
      "entropy": 5.880005264282227,
      "epoch": 3.1930363742462555,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039913086203569373,
      "loss": 4.9784,
      "mean_token_accuracy": 0.20919807851314545,
      "num_tokens": 71186635.0,
      "step": 41040
    },
    {
      "entropy": 5.86978964805603,
      "epoch": 3.1934254036179732,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039910774497892046,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20387642085552216,
      "num_tokens": 71195407.0,
      "step": 41045
    },
    {
      "entropy": 5.837751340866089,
      "epoch": 3.193814432989691,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039908462603900977,
      "loss": 4.9651,
      "mean_token_accuracy": 0.20461779683828354,
      "num_tokens": 71203756.0,
      "step": 41050
    },
    {
      "entropy": 5.761446094512939,
      "epoch": 3.194203462361408,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003990615052163123,
      "loss": 4.9151,
      "mean_token_accuracy": 0.21475869566202163,
      "num_tokens": 71212077.0,
      "step": 41055
    },
    {
      "entropy": 5.828308343887329,
      "epoch": 3.194592491733126,
      "grad_norm": 1.328125,
      "learning_rate": 0.000399038382511179,
      "loss": 4.9273,
      "mean_token_accuracy": 0.21077749878168106,
      "num_tokens": 71220279.0,
      "step": 41060
    },
    {
      "entropy": 5.7490967273712155,
      "epoch": 3.1949815211048436,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003990152579239607,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2155602306127548,
      "num_tokens": 71228731.0,
      "step": 41065
    },
    {
      "entropy": 5.8000188827514645,
      "epoch": 3.195370550476561,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039899213145500815,
      "loss": 4.983,
      "mean_token_accuracy": 0.20990292280912398,
      "num_tokens": 71237520.0,
      "step": 41070
    },
    {
      "entropy": 5.706753158569336,
      "epoch": 3.1957595798482785,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039896900310467244,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22143174856901168,
      "num_tokens": 71245912.0,
      "step": 41075
    },
    {
      "entropy": 5.799352788925171,
      "epoch": 3.196148609219996,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039894587287330434,
      "loss": 4.9511,
      "mean_token_accuracy": 0.21596096456050873,
      "num_tokens": 71254815.0,
      "step": 41080
    },
    {
      "entropy": 5.874398899078369,
      "epoch": 3.1965376385917135,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00039892274076125503,
      "loss": 5.012,
      "mean_token_accuracy": 0.20449885725975037,
      "num_tokens": 71262983.0,
      "step": 41085
    },
    {
      "entropy": 5.806707811355591,
      "epoch": 3.196926667963431,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003988996067688753,
      "loss": 4.9318,
      "mean_token_accuracy": 0.2171550363302231,
      "num_tokens": 71270555.0,
      "step": 41090
    },
    {
      "entropy": 5.790062141418457,
      "epoch": 3.197315697335149,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003988764708965163,
      "loss": 4.9292,
      "mean_token_accuracy": 0.2130316197872162,
      "num_tokens": 71278593.0,
      "step": 41095
    },
    {
      "entropy": 5.816224861145019,
      "epoch": 3.1977047267068666,
      "grad_norm": 1.34375,
      "learning_rate": 0.000398853333144529,
      "loss": 4.8942,
      "mean_token_accuracy": 0.21932464987039565,
      "num_tokens": 71287048.0,
      "step": 41100
    },
    {
      "entropy": 5.851548862457276,
      "epoch": 3.198093756078584,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039883019351326447,
      "loss": 5.0718,
      "mean_token_accuracy": 0.19560532867908478,
      "num_tokens": 71295965.0,
      "step": 41105
    },
    {
      "entropy": 5.809825611114502,
      "epoch": 3.1984827854503015,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039880705200307377,
      "loss": 4.9607,
      "mean_token_accuracy": 0.20645642876625062,
      "num_tokens": 71305094.0,
      "step": 41110
    },
    {
      "entropy": 5.839958810806275,
      "epoch": 3.198871814822019,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003987839086143083,
      "loss": 4.9115,
      "mean_token_accuracy": 0.21504350155591964,
      "num_tokens": 71313466.0,
      "step": 41115
    },
    {
      "entropy": 5.842421531677246,
      "epoch": 3.1992608441937365,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039876076334731885,
      "loss": 5.018,
      "mean_token_accuracy": 0.2086254447698593,
      "num_tokens": 71322799.0,
      "step": 41120
    },
    {
      "entropy": 5.817697906494141,
      "epoch": 3.199649873565454,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000398737616202457,
      "loss": 4.973,
      "mean_token_accuracy": 0.21059157252311705,
      "num_tokens": 71331618.0,
      "step": 41125
    },
    {
      "entropy": 5.787054395675659,
      "epoch": 3.200038902937172,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039871446718007364,
      "loss": 4.9469,
      "mean_token_accuracy": 0.20836376547813415,
      "num_tokens": 71339784.0,
      "step": 41130
    },
    {
      "entropy": 5.859975814819336,
      "epoch": 3.200427932308889,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003986913162805201,
      "loss": 5.0778,
      "mean_token_accuracy": 0.19823651909828185,
      "num_tokens": 71347896.0,
      "step": 41135
    },
    {
      "entropy": 5.868724060058594,
      "epoch": 3.200816961680607,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039866816350414786,
      "loss": 5.0269,
      "mean_token_accuracy": 0.20309681296348572,
      "num_tokens": 71356888.0,
      "step": 41140
    },
    {
      "entropy": 5.807567834854126,
      "epoch": 3.2012059910523245,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000398645008851308,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21380703747272492,
      "num_tokens": 71365217.0,
      "step": 41145
    },
    {
      "entropy": 5.823787784576416,
      "epoch": 3.201595020424042,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039862185232235196,
      "loss": 4.9566,
      "mean_token_accuracy": 0.2020031541585922,
      "num_tokens": 71373975.0,
      "step": 41150
    },
    {
      "entropy": 5.741670989990235,
      "epoch": 3.2019840497957595,
      "grad_norm": 1.375,
      "learning_rate": 0.0003985986939176311,
      "loss": 4.9151,
      "mean_token_accuracy": 0.21347890496253968,
      "num_tokens": 71382733.0,
      "step": 41155
    },
    {
      "entropy": 5.809070730209351,
      "epoch": 3.202373079167477,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039857553363749674,
      "loss": 4.9556,
      "mean_token_accuracy": 0.2131333604454994,
      "num_tokens": 71392514.0,
      "step": 41160
    },
    {
      "entropy": 5.846995735168457,
      "epoch": 3.202762108539195,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039855237148230026,
      "loss": 4.9691,
      "mean_token_accuracy": 0.205020909011364,
      "num_tokens": 71401305.0,
      "step": 41165
    },
    {
      "entropy": 5.892605638504028,
      "epoch": 3.203151137910912,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039852920745239343,
      "loss": 4.9757,
      "mean_token_accuracy": 0.2060181751847267,
      "num_tokens": 71410449.0,
      "step": 41170
    },
    {
      "entropy": 5.8253387928009035,
      "epoch": 3.20354016728263,
      "grad_norm": 1.25,
      "learning_rate": 0.00039850604154812727,
      "loss": 4.9567,
      "mean_token_accuracy": 0.2064486026763916,
      "num_tokens": 71420284.0,
      "step": 41175
    },
    {
      "entropy": 5.791077518463135,
      "epoch": 3.2039291966543475,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003984828737698536,
      "loss": 5.0353,
      "mean_token_accuracy": 0.20231992602348328,
      "num_tokens": 71428890.0,
      "step": 41180
    },
    {
      "entropy": 5.861621952056884,
      "epoch": 3.204318226026065,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039845970411792384,
      "loss": 4.9834,
      "mean_token_accuracy": 0.20630451589822768,
      "num_tokens": 71437569.0,
      "step": 41185
    },
    {
      "entropy": 5.872475242614746,
      "epoch": 3.2047072553977825,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003984365325926896,
      "loss": 5.0948,
      "mean_token_accuracy": 0.20322130173444747,
      "num_tokens": 71446453.0,
      "step": 41190
    },
    {
      "entropy": 5.8338159084320065,
      "epoch": 3.2050962847695,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039841335919450245,
      "loss": 5.0347,
      "mean_token_accuracy": 0.20543756932020188,
      "num_tokens": 71455727.0,
      "step": 41195
    },
    {
      "entropy": 5.872816848754883,
      "epoch": 3.205485314141218,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039839018392371397,
      "loss": 5.0033,
      "mean_token_accuracy": 0.21344245821237565,
      "num_tokens": 71464528.0,
      "step": 41200
    },
    {
      "entropy": 5.849108171463013,
      "epoch": 3.205874343512935,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003983670067806759,
      "loss": 4.8674,
      "mean_token_accuracy": 0.21460047960281373,
      "num_tokens": 71472386.0,
      "step": 41205
    },
    {
      "entropy": 5.9271749496459964,
      "epoch": 3.206263372884653,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003983438277657398,
      "loss": 4.9481,
      "mean_token_accuracy": 0.21934157609939575,
      "num_tokens": 71480616.0,
      "step": 41210
    },
    {
      "entropy": 5.817424726486206,
      "epoch": 3.2066524022563705,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003983206468792575,
      "loss": 5.0187,
      "mean_token_accuracy": 0.20366560518741608,
      "num_tokens": 71490064.0,
      "step": 41215
    },
    {
      "entropy": 5.821478652954101,
      "epoch": 3.2070414316280877,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003982974641215806,
      "loss": 5.0146,
      "mean_token_accuracy": 0.2062872052192688,
      "num_tokens": 71499046.0,
      "step": 41220
    },
    {
      "entropy": 5.910622310638428,
      "epoch": 3.2074304609998054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000398274279493061,
      "loss": 5.0499,
      "mean_token_accuracy": 0.21046550571918488,
      "num_tokens": 71507775.0,
      "step": 41225
    },
    {
      "entropy": 5.824555540084839,
      "epoch": 3.207819490371523,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039825109299405036,
      "loss": 4.9389,
      "mean_token_accuracy": 0.22115633487701417,
      "num_tokens": 71517225.0,
      "step": 41230
    },
    {
      "entropy": 5.8287928104400635,
      "epoch": 3.2082085197432404,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003982279046249006,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20850649178028108,
      "num_tokens": 71526171.0,
      "step": 41235
    },
    {
      "entropy": 5.875583982467651,
      "epoch": 3.208597549114958,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039820471438596346,
      "loss": 5.006,
      "mean_token_accuracy": 0.20864122062921525,
      "num_tokens": 71534573.0,
      "step": 41240
    },
    {
      "entropy": 5.958389711380005,
      "epoch": 3.208986578486676,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039818152227759097,
      "loss": 5.0966,
      "mean_token_accuracy": 0.2026665210723877,
      "num_tokens": 71543993.0,
      "step": 41245
    },
    {
      "entropy": 5.868214511871338,
      "epoch": 3.2093756078583935,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003981583283001349,
      "loss": 5.0331,
      "mean_token_accuracy": 0.2008140578866005,
      "num_tokens": 71552547.0,
      "step": 41250
    },
    {
      "entropy": 5.792708063125611,
      "epoch": 3.2097646372301107,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003981351324539472,
      "loss": 4.8774,
      "mean_token_accuracy": 0.2201257437467575,
      "num_tokens": 71561523.0,
      "step": 41255
    },
    {
      "entropy": 5.845886182785034,
      "epoch": 3.2101536666018284,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003981119347393799,
      "loss": 4.9755,
      "mean_token_accuracy": 0.21426019072532654,
      "num_tokens": 71569505.0,
      "step": 41260
    },
    {
      "entropy": 5.881434154510498,
      "epoch": 3.210542695973546,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039808873515678495,
      "loss": 5.0609,
      "mean_token_accuracy": 0.2033266082406044,
      "num_tokens": 71577895.0,
      "step": 41265
    },
    {
      "entropy": 5.83377480506897,
      "epoch": 3.2109317253452634,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003980655337065144,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2082776039838791,
      "num_tokens": 71586454.0,
      "step": 41270
    },
    {
      "entropy": 5.826668310165405,
      "epoch": 3.211320754716981,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003980423303889201,
      "loss": 5.0214,
      "mean_token_accuracy": 0.20655311942100524,
      "num_tokens": 71595529.0,
      "step": 41275
    },
    {
      "entropy": 5.905229663848877,
      "epoch": 3.211709784088699,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039801912520435437,
      "loss": 5.08,
      "mean_token_accuracy": 0.2000127539038658,
      "num_tokens": 71604345.0,
      "step": 41280
    },
    {
      "entropy": 5.8134174823760985,
      "epoch": 3.2120988134604165,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003979959181531692,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21516081541776658,
      "num_tokens": 71613060.0,
      "step": 41285
    },
    {
      "entropy": 5.821517038345337,
      "epoch": 3.2124878428321337,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039797270923571683,
      "loss": 4.9541,
      "mean_token_accuracy": 0.2189641460776329,
      "num_tokens": 71622122.0,
      "step": 41290
    },
    {
      "entropy": 5.782667732238769,
      "epoch": 3.2128768722038514,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00039794949845234925,
      "loss": 4.9288,
      "mean_token_accuracy": 0.2190478801727295,
      "num_tokens": 71630812.0,
      "step": 41295
    },
    {
      "entropy": 5.8173774719238285,
      "epoch": 3.213265901575569,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039792628580341874,
      "loss": 5.0167,
      "mean_token_accuracy": 0.20952873080968856,
      "num_tokens": 71640341.0,
      "step": 41300
    },
    {
      "entropy": 5.816604232788086,
      "epoch": 3.2136549309472864,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039790307128927745,
      "loss": 4.8456,
      "mean_token_accuracy": 0.21552089005708694,
      "num_tokens": 71649383.0,
      "step": 41305
    },
    {
      "entropy": 5.842979478836059,
      "epoch": 3.214043960319004,
      "grad_norm": 1.25,
      "learning_rate": 0.00039787985491027773,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21119867861270905,
      "num_tokens": 71657989.0,
      "step": 41310
    },
    {
      "entropy": 5.743782043457031,
      "epoch": 3.2144329896907218,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00039785663666677175,
      "loss": 4.8997,
      "mean_token_accuracy": 0.21266344487667083,
      "num_tokens": 71666569.0,
      "step": 41315
    },
    {
      "entropy": 5.791001796722412,
      "epoch": 3.214822019062439,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00039783341655911185,
      "loss": 4.9248,
      "mean_token_accuracy": 0.2088156446814537,
      "num_tokens": 71675256.0,
      "step": 41320
    },
    {
      "entropy": 5.775504302978516,
      "epoch": 3.2152110484341567,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003978101945876504,
      "loss": 4.9039,
      "mean_token_accuracy": 0.20965370833873748,
      "num_tokens": 71684195.0,
      "step": 41325
    },
    {
      "entropy": 5.854995489120483,
      "epoch": 3.2156000778058744,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039778697075273977,
      "loss": 4.9029,
      "mean_token_accuracy": 0.21143727451562883,
      "num_tokens": 71692869.0,
      "step": 41330
    },
    {
      "entropy": 5.77829647064209,
      "epoch": 3.2159891071775917,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039776374505473224,
      "loss": 4.947,
      "mean_token_accuracy": 0.20896151661872864,
      "num_tokens": 71701560.0,
      "step": 41335
    },
    {
      "entropy": 5.885304355621338,
      "epoch": 3.2163781365493094,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003977405174939803,
      "loss": 5.0431,
      "mean_token_accuracy": 0.20134537816047668,
      "num_tokens": 71710282.0,
      "step": 41340
    },
    {
      "entropy": 5.8914295673370365,
      "epoch": 3.216767165921027,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039771728807083635,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2089582145214081,
      "num_tokens": 71718364.0,
      "step": 41345
    },
    {
      "entropy": 5.8717145919799805,
      "epoch": 3.2171561952927448,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039769405678565287,
      "loss": 5.0116,
      "mean_token_accuracy": 0.2029728874564171,
      "num_tokens": 71726712.0,
      "step": 41350
    },
    {
      "entropy": 5.822555446624756,
      "epoch": 3.217545224664462,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039767082363878237,
      "loss": 5.0155,
      "mean_token_accuracy": 0.20062066167593,
      "num_tokens": 71735954.0,
      "step": 41355
    },
    {
      "entropy": 5.903098917007446,
      "epoch": 3.2179342540361797,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003976475886305774,
      "loss": 5.1367,
      "mean_token_accuracy": 0.20088881701231004,
      "num_tokens": 71745025.0,
      "step": 41360
    },
    {
      "entropy": 5.931817245483399,
      "epoch": 3.2183232834078974,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003976243517613904,
      "loss": 5.0301,
      "mean_token_accuracy": 0.20712753683328627,
      "num_tokens": 71754722.0,
      "step": 41365
    },
    {
      "entropy": 6.012892198562622,
      "epoch": 3.2187123127796147,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039760111303157415,
      "loss": 5.1183,
      "mean_token_accuracy": 0.1994749441742897,
      "num_tokens": 71763957.0,
      "step": 41370
    },
    {
      "entropy": 5.850116729736328,
      "epoch": 3.2191013421513324,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000397577872441481,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20801551938056945,
      "num_tokens": 71772890.0,
      "step": 41375
    },
    {
      "entropy": 5.859717607498169,
      "epoch": 3.21949037152305,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003975546299914639,
      "loss": 5.018,
      "mean_token_accuracy": 0.20867818146944045,
      "num_tokens": 71781081.0,
      "step": 41380
    },
    {
      "entropy": 5.867979717254639,
      "epoch": 3.2198794008947678,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003975313856818753,
      "loss": 4.9718,
      "mean_token_accuracy": 0.20904824137687683,
      "num_tokens": 71788972.0,
      "step": 41385
    },
    {
      "entropy": 5.876728820800781,
      "epoch": 3.220268430266485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003975081395130679,
      "loss": 5.0669,
      "mean_token_accuracy": 0.20640120208263396,
      "num_tokens": 71797120.0,
      "step": 41390
    },
    {
      "entropy": 5.80188512802124,
      "epoch": 3.2206574596382027,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003974848914853946,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21137304306030275,
      "num_tokens": 71805558.0,
      "step": 41395
    },
    {
      "entropy": 5.866464567184448,
      "epoch": 3.2210464890099204,
      "grad_norm": 1.5625,
      "learning_rate": 0.00039746164159920795,
      "loss": 4.9756,
      "mean_token_accuracy": 0.2108137860894203,
      "num_tokens": 71814228.0,
      "step": 41400
    },
    {
      "entropy": 5.806661939620971,
      "epoch": 3.2214355183816377,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039743838985486075,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21098487824201584,
      "num_tokens": 71823436.0,
      "step": 41405
    },
    {
      "entropy": 5.910964393615723,
      "epoch": 3.2218245477533554,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003974151362527059,
      "loss": 5.0165,
      "mean_token_accuracy": 0.2088575229048729,
      "num_tokens": 71832135.0,
      "step": 41410
    },
    {
      "entropy": 5.8319660186767575,
      "epoch": 3.222213577125073,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039739188079309626,
      "loss": 4.9552,
      "mean_token_accuracy": 0.21015356481075287,
      "num_tokens": 71840780.0,
      "step": 41415
    },
    {
      "entropy": 5.816829347610474,
      "epoch": 3.2226026064967903,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003973686234763846,
      "loss": 5.0259,
      "mean_token_accuracy": 0.20309171825647354,
      "num_tokens": 71849135.0,
      "step": 41420
    },
    {
      "entropy": 5.90092716217041,
      "epoch": 3.222991635868508,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003973453643029239,
      "loss": 4.9841,
      "mean_token_accuracy": 0.21176661550998688,
      "num_tokens": 71858967.0,
      "step": 41425
    },
    {
      "entropy": 5.811150693893433,
      "epoch": 3.2233806652402257,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003973221032730669,
      "loss": 4.9109,
      "mean_token_accuracy": 0.2126663938164711,
      "num_tokens": 71867355.0,
      "step": 41430
    },
    {
      "entropy": 5.858457899093628,
      "epoch": 3.2237696946119434,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003972988403871668,
      "loss": 4.9175,
      "mean_token_accuracy": 0.20983185172080993,
      "num_tokens": 71875723.0,
      "step": 41435
    },
    {
      "entropy": 5.846921586990357,
      "epoch": 3.2241587239836607,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003972755756455764,
      "loss": 4.9564,
      "mean_token_accuracy": 0.21846983283758165,
      "num_tokens": 71884291.0,
      "step": 41440
    },
    {
      "entropy": 5.768748617172241,
      "epoch": 3.2245477533553784,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003972523090486487,
      "loss": 4.9537,
      "mean_token_accuracy": 0.21587966978549958,
      "num_tokens": 71893672.0,
      "step": 41445
    },
    {
      "entropy": 5.846840476989746,
      "epoch": 3.224936782727096,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003972290405967369,
      "loss": 5.0019,
      "mean_token_accuracy": 0.2067606896162033,
      "num_tokens": 71901964.0,
      "step": 41450
    },
    {
      "entropy": 5.852076578140259,
      "epoch": 3.2253258120988133,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003972057702901939,
      "loss": 5.0052,
      "mean_token_accuracy": 0.21014974713325502,
      "num_tokens": 71910801.0,
      "step": 41455
    },
    {
      "entropy": 5.882115650177002,
      "epoch": 3.225714841470531,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003971824981293729,
      "loss": 5.0133,
      "mean_token_accuracy": 0.20798835903406143,
      "num_tokens": 71919833.0,
      "step": 41460
    },
    {
      "entropy": 5.838105964660644,
      "epoch": 3.2261038708422487,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003971592241146269,
      "loss": 4.996,
      "mean_token_accuracy": 0.2031509056687355,
      "num_tokens": 71928117.0,
      "step": 41465
    },
    {
      "entropy": 5.732760810852051,
      "epoch": 3.226492900213966,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003971359482463092,
      "loss": 4.8157,
      "mean_token_accuracy": 0.22403545379638673,
      "num_tokens": 71937331.0,
      "step": 41470
    },
    {
      "entropy": 5.865112686157227,
      "epoch": 3.2268819295856837,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003971126705247727,
      "loss": 5.0283,
      "mean_token_accuracy": 0.2026947557926178,
      "num_tokens": 71945656.0,
      "step": 41475
    },
    {
      "entropy": 5.871706867218018,
      "epoch": 3.2272709589574013,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003970893909503709,
      "loss": 4.9637,
      "mean_token_accuracy": 0.2071118786931038,
      "num_tokens": 71954833.0,
      "step": 41480
    },
    {
      "entropy": 5.845706987380981,
      "epoch": 3.227659988329119,
      "grad_norm": 1.3125,
      "learning_rate": 0.000397066109523457,
      "loss": 4.8873,
      "mean_token_accuracy": 0.21931109577417374,
      "num_tokens": 71962798.0,
      "step": 41485
    },
    {
      "entropy": 5.869112873077393,
      "epoch": 3.2280490177008363,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003970428262443842,
      "loss": 5.0605,
      "mean_token_accuracy": 0.1988818407058716,
      "num_tokens": 71972000.0,
      "step": 41490
    },
    {
      "entropy": 5.8797636985778805,
      "epoch": 3.228438047072554,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039701954111350565,
      "loss": 5.0329,
      "mean_token_accuracy": 0.20906647741794587,
      "num_tokens": 71981950.0,
      "step": 41495
    },
    {
      "entropy": 5.839580965042114,
      "epoch": 3.2288270764442717,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003969962541311748,
      "loss": 4.9063,
      "mean_token_accuracy": 0.21319869458675383,
      "num_tokens": 71989662.0,
      "step": 41500
    },
    {
      "entropy": 5.754192924499511,
      "epoch": 3.229216105815989,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039697296529774494,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21008898168802262,
      "num_tokens": 71998140.0,
      "step": 41505
    },
    {
      "entropy": 5.793681621551514,
      "epoch": 3.2296051351877066,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039694967461356944,
      "loss": 4.9677,
      "mean_token_accuracy": 0.21143849492073058,
      "num_tokens": 72007365.0,
      "step": 41510
    },
    {
      "entropy": 5.834577751159668,
      "epoch": 3.2299941645594243,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039692638207900177,
      "loss": 4.9823,
      "mean_token_accuracy": 0.2113758370280266,
      "num_tokens": 72015380.0,
      "step": 41515
    },
    {
      "entropy": 5.845998334884643,
      "epoch": 3.2303831939311416,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003969030876943954,
      "loss": 4.9623,
      "mean_token_accuracy": 0.21121104955673217,
      "num_tokens": 72024322.0,
      "step": 41520
    },
    {
      "entropy": 5.947836542129517,
      "epoch": 3.2307722233028593,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00039687979146010367,
      "loss": 5.0697,
      "mean_token_accuracy": 0.20080548524856567,
      "num_tokens": 72032905.0,
      "step": 41525
    },
    {
      "entropy": 5.848176717758179,
      "epoch": 3.231161252674577,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039685649337648,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21673622131347656,
      "num_tokens": 72041738.0,
      "step": 41530
    },
    {
      "entropy": 5.793944263458252,
      "epoch": 3.2315502820462947,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039683319344387813,
      "loss": 4.902,
      "mean_token_accuracy": 0.21445939242839812,
      "num_tokens": 72050111.0,
      "step": 41535
    },
    {
      "entropy": 5.729693508148193,
      "epoch": 3.231939311418012,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039680989166265135,
      "loss": 4.843,
      "mean_token_accuracy": 0.21119237393140794,
      "num_tokens": 72058526.0,
      "step": 41540
    },
    {
      "entropy": 5.785159921646118,
      "epoch": 3.2323283407897296,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039678658803315336,
      "loss": 5.0359,
      "mean_token_accuracy": 0.21021993458271027,
      "num_tokens": 72067464.0,
      "step": 41545
    },
    {
      "entropy": 5.802345514297485,
      "epoch": 3.2327173701614473,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039676328255573777,
      "loss": 4.895,
      "mean_token_accuracy": 0.21891624480485916,
      "num_tokens": 72076857.0,
      "step": 41550
    },
    {
      "entropy": 5.82989010810852,
      "epoch": 3.2331063995331646,
      "grad_norm": 1.375,
      "learning_rate": 0.00039673997523075825,
      "loss": 4.9246,
      "mean_token_accuracy": 0.20682331919670105,
      "num_tokens": 72085327.0,
      "step": 41555
    },
    {
      "entropy": 5.81734414100647,
      "epoch": 3.2334954289048823,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039671666605856825,
      "loss": 4.8786,
      "mean_token_accuracy": 0.21542142033576966,
      "num_tokens": 72093336.0,
      "step": 41560
    },
    {
      "entropy": 5.856696367263794,
      "epoch": 3.2338844582766,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039669335503952165,
      "loss": 5.0043,
      "mean_token_accuracy": 0.20418593734502793,
      "num_tokens": 72102266.0,
      "step": 41565
    },
    {
      "entropy": 5.815034580230713,
      "epoch": 3.2342734876483172,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039667004217397206,
      "loss": 4.9663,
      "mean_token_accuracy": 0.2159833312034607,
      "num_tokens": 72111364.0,
      "step": 41570
    },
    {
      "entropy": 5.823343420028687,
      "epoch": 3.234662517020035,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039664672746227326,
      "loss": 4.9151,
      "mean_token_accuracy": 0.2111508145928383,
      "num_tokens": 72120038.0,
      "step": 41575
    },
    {
      "entropy": 5.785775184631348,
      "epoch": 3.2350515463917526,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003966234109047789,
      "loss": 4.9823,
      "mean_token_accuracy": 0.21344225108623505,
      "num_tokens": 72129033.0,
      "step": 41580
    },
    {
      "entropy": 5.935666465759278,
      "epoch": 3.2354405757634703,
      "grad_norm": 1.5,
      "learning_rate": 0.000396600092501843,
      "loss": 5.1283,
      "mean_token_accuracy": 0.19842617064714432,
      "num_tokens": 72138806.0,
      "step": 41585
    },
    {
      "entropy": 5.927566623687744,
      "epoch": 3.2358296051351876,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039657677225381915,
      "loss": 5.0136,
      "mean_token_accuracy": 0.21346484422683715,
      "num_tokens": 72147723.0,
      "step": 41590
    },
    {
      "entropy": 5.89341368675232,
      "epoch": 3.2362186345069053,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003965534501610613,
      "loss": 5.0758,
      "mean_token_accuracy": 0.19774211198091507,
      "num_tokens": 72156724.0,
      "step": 41595
    },
    {
      "entropy": 5.8553609371185305,
      "epoch": 3.236607663878623,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003965301262239234,
      "loss": 4.9526,
      "mean_token_accuracy": 0.21073439866304397,
      "num_tokens": 72164803.0,
      "step": 41600
    },
    {
      "entropy": 5.823162412643432,
      "epoch": 3.2369966932503402,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003965068004427591,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20301738530397415,
      "num_tokens": 72173770.0,
      "step": 41605
    },
    {
      "entropy": 5.874419641494751,
      "epoch": 3.237385722622058,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003964834728179226,
      "loss": 5.0039,
      "mean_token_accuracy": 0.2070169508457184,
      "num_tokens": 72183072.0,
      "step": 41610
    },
    {
      "entropy": 5.776216316223144,
      "epoch": 3.2377747519937756,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039646014334976783,
      "loss": 4.8671,
      "mean_token_accuracy": 0.21907583475112916,
      "num_tokens": 72190494.0,
      "step": 41615
    },
    {
      "entropy": 5.813143634796143,
      "epoch": 3.2381637813654933,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039643681203864863,
      "loss": 5.0128,
      "mean_token_accuracy": 0.2085426226258278,
      "num_tokens": 72198775.0,
      "step": 41620
    },
    {
      "entropy": 5.932885456085205,
      "epoch": 3.2385528107372106,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039641347888491905,
      "loss": 5.0918,
      "mean_token_accuracy": 0.2021746188402176,
      "num_tokens": 72207137.0,
      "step": 41625
    },
    {
      "entropy": 5.8863872528076175,
      "epoch": 3.2389418401089283,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039639014388893337,
      "loss": 5.0308,
      "mean_token_accuracy": 0.2055279016494751,
      "num_tokens": 72215645.0,
      "step": 41630
    },
    {
      "entropy": 5.853871154785156,
      "epoch": 3.239330869480646,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003963668070510453,
      "loss": 4.9483,
      "mean_token_accuracy": 0.2093917265534401,
      "num_tokens": 72224723.0,
      "step": 41635
    },
    {
      "entropy": 5.8575742721557615,
      "epoch": 3.2397198988523632,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003963434683716091,
      "loss": 4.9632,
      "mean_token_accuracy": 0.21289380192756652,
      "num_tokens": 72233195.0,
      "step": 41640
    },
    {
      "entropy": 5.904401016235352,
      "epoch": 3.240108928224081,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003963201278509789,
      "loss": 4.9839,
      "mean_token_accuracy": 0.20977976620197297,
      "num_tokens": 72241850.0,
      "step": 41645
    },
    {
      "entropy": 5.8810028553009035,
      "epoch": 3.2404979575957986,
      "grad_norm": 1.25,
      "learning_rate": 0.0003962967854895089,
      "loss": 4.9841,
      "mean_token_accuracy": 0.20454856455326081,
      "num_tokens": 72250624.0,
      "step": 41650
    },
    {
      "entropy": 5.855868148803711,
      "epoch": 3.240886986967516,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003962734412875533,
      "loss": 4.991,
      "mean_token_accuracy": 0.1996140405535698,
      "num_tokens": 72258669.0,
      "step": 41655
    },
    {
      "entropy": 5.845728158950806,
      "epoch": 3.2412760163392336,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003962500952454662,
      "loss": 4.959,
      "mean_token_accuracy": 0.2090431272983551,
      "num_tokens": 72267577.0,
      "step": 41660
    },
    {
      "entropy": 5.878549480438233,
      "epoch": 3.2416650457109513,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039622674736360195,
      "loss": 5.1083,
      "mean_token_accuracy": 0.20154589861631395,
      "num_tokens": 72276008.0,
      "step": 41665
    },
    {
      "entropy": 5.894612264633179,
      "epoch": 3.2420540750826685,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039620339764231467,
      "loss": 5.0316,
      "mean_token_accuracy": 0.20562729835510254,
      "num_tokens": 72284807.0,
      "step": 41670
    },
    {
      "entropy": 5.837061595916748,
      "epoch": 3.242443104454386,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039618004608195877,
      "loss": 5.0105,
      "mean_token_accuracy": 0.20710051953792571,
      "num_tokens": 72294364.0,
      "step": 41675
    },
    {
      "entropy": 5.8676715850830075,
      "epoch": 3.242832133826104,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039615669268288863,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20890984535217286,
      "num_tokens": 72302931.0,
      "step": 41680
    },
    {
      "entropy": 5.870527601242065,
      "epoch": 3.2432211631978216,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039613333744545836,
      "loss": 4.9786,
      "mean_token_accuracy": 0.2102140411734581,
      "num_tokens": 72311911.0,
      "step": 41685
    },
    {
      "entropy": 5.799951696395874,
      "epoch": 3.243610192569539,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003961099803700226,
      "loss": 4.8935,
      "mean_token_accuracy": 0.2241044744849205,
      "num_tokens": 72320435.0,
      "step": 41690
    },
    {
      "entropy": 5.8635382652282715,
      "epoch": 3.2439992219412566,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003960866214569357,
      "loss": 4.987,
      "mean_token_accuracy": 0.20266940593719482,
      "num_tokens": 72329098.0,
      "step": 41695
    },
    {
      "entropy": 5.80028510093689,
      "epoch": 3.2443882513129743,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039606326070655204,
      "loss": 4.9314,
      "mean_token_accuracy": 0.20103475749492644,
      "num_tokens": 72337650.0,
      "step": 41700
    },
    {
      "entropy": 5.815481042861938,
      "epoch": 3.2447772806846915,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039603989811922596,
      "loss": 4.915,
      "mean_token_accuracy": 0.21421968340873718,
      "num_tokens": 72346627.0,
      "step": 41705
    },
    {
      "entropy": 5.841947555541992,
      "epoch": 3.245166310056409,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003960165336953122,
      "loss": 4.9323,
      "mean_token_accuracy": 0.21651702076196672,
      "num_tokens": 72355695.0,
      "step": 41710
    },
    {
      "entropy": 5.81978964805603,
      "epoch": 3.245555339428127,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003959931674351651,
      "loss": 4.9634,
      "mean_token_accuracy": 0.20882151126861573,
      "num_tokens": 72364078.0,
      "step": 41715
    },
    {
      "entropy": 5.828111410140991,
      "epoch": 3.2459443687998446,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003959697993391393,
      "loss": 5.0091,
      "mean_token_accuracy": 0.2093831866979599,
      "num_tokens": 72373021.0,
      "step": 41720
    },
    {
      "entropy": 5.869501399993896,
      "epoch": 3.246333398171562,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039594642940758934,
      "loss": 5.0373,
      "mean_token_accuracy": 0.20530015379190444,
      "num_tokens": 72381688.0,
      "step": 41725
    },
    {
      "entropy": 5.900115013122559,
      "epoch": 3.2467224275432796,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039592305764086984,
      "loss": 5.0487,
      "mean_token_accuracy": 0.2116955116391182,
      "num_tokens": 72389560.0,
      "step": 41730
    },
    {
      "entropy": 5.837467765808105,
      "epoch": 3.2471114569149973,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039589968403933533,
      "loss": 4.9323,
      "mean_token_accuracy": 0.21286197155714034,
      "num_tokens": 72398206.0,
      "step": 41735
    },
    {
      "entropy": 5.807404661178589,
      "epoch": 3.2475004862867145,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003958763086033406,
      "loss": 5.0395,
      "mean_token_accuracy": 0.20220747739076614,
      "num_tokens": 72407658.0,
      "step": 41740
    },
    {
      "entropy": 5.91076078414917,
      "epoch": 3.247889515658432,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003958529313332403,
      "loss": 5.0502,
      "mean_token_accuracy": 0.20345233678817748,
      "num_tokens": 72415982.0,
      "step": 41745
    },
    {
      "entropy": 5.907308769226074,
      "epoch": 3.24827854503015,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003958295522293891,
      "loss": 5.0451,
      "mean_token_accuracy": 0.2003877729177475,
      "num_tokens": 72424992.0,
      "step": 41750
    },
    {
      "entropy": 5.824803447723388,
      "epoch": 3.248667574401867,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003958061712921417,
      "loss": 4.8829,
      "mean_token_accuracy": 0.21685806512832642,
      "num_tokens": 72432925.0,
      "step": 41755
    },
    {
      "entropy": 5.887304878234863,
      "epoch": 3.249056603773585,
      "grad_norm": 1.390625,
      "learning_rate": 0.000395782788521853,
      "loss": 5.0526,
      "mean_token_accuracy": 0.20483575761318207,
      "num_tokens": 72441880.0,
      "step": 41760
    },
    {
      "entropy": 5.864757776260376,
      "epoch": 3.2494456331453025,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003957594039188778,
      "loss": 4.9652,
      "mean_token_accuracy": 0.21303342431783676,
      "num_tokens": 72449259.0,
      "step": 41765
    },
    {
      "entropy": 5.829287910461426,
      "epoch": 3.24983466251702,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003957360174835708,
      "loss": 4.9323,
      "mean_token_accuracy": 0.2128154918551445,
      "num_tokens": 72457881.0,
      "step": 41770
    },
    {
      "entropy": 5.7773590087890625,
      "epoch": 3.2502236918887375,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003957126292162868,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21630093306303025,
      "num_tokens": 72466513.0,
      "step": 41775
    },
    {
      "entropy": 5.826560735702515,
      "epoch": 3.250612721260455,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003956892391173809,
      "loss": 4.9129,
      "mean_token_accuracy": 0.220101997256279,
      "num_tokens": 72474888.0,
      "step": 41780
    },
    {
      "entropy": 5.8689844608306885,
      "epoch": 3.251001750632173,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039566584718720793,
      "loss": 5.0216,
      "mean_token_accuracy": 0.21203243732452393,
      "num_tokens": 72484071.0,
      "step": 41785
    },
    {
      "entropy": 5.819292402267456,
      "epoch": 3.25139078000389,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003956424534261227,
      "loss": 5.0007,
      "mean_token_accuracy": 0.21201223582029344,
      "num_tokens": 72492962.0,
      "step": 41790
    },
    {
      "entropy": 5.8278398513793945,
      "epoch": 3.251779809375608,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039561905783448025,
      "loss": 4.9973,
      "mean_token_accuracy": 0.20687753856182098,
      "num_tokens": 72502247.0,
      "step": 41795
    },
    {
      "entropy": 5.893687391281128,
      "epoch": 3.2521688387473255,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039559566041263565,
      "loss": 4.9904,
      "mean_token_accuracy": 0.21269286721944808,
      "num_tokens": 72510611.0,
      "step": 41800
    },
    {
      "entropy": 5.947108030319214,
      "epoch": 3.252557868119043,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003955722611609438,
      "loss": 5.0184,
      "mean_token_accuracy": 0.20822292417287827,
      "num_tokens": 72518901.0,
      "step": 41805
    },
    {
      "entropy": 5.90369930267334,
      "epoch": 3.2529468974907605,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039554886007975976,
      "loss": 5.0206,
      "mean_token_accuracy": 0.20320906341075898,
      "num_tokens": 72527422.0,
      "step": 41810
    },
    {
      "entropy": 5.827991628646851,
      "epoch": 3.253335926862478,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039552545716943873,
      "loss": 4.9727,
      "mean_token_accuracy": 0.21307283341884614,
      "num_tokens": 72536388.0,
      "step": 41815
    },
    {
      "entropy": 5.887800312042236,
      "epoch": 3.253724956234196,
      "grad_norm": 1.25,
      "learning_rate": 0.0003955020524303358,
      "loss": 5.025,
      "mean_token_accuracy": 0.20682238638401032,
      "num_tokens": 72545236.0,
      "step": 41820
    },
    {
      "entropy": 5.839358329772949,
      "epoch": 3.254113985605913,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039547864586280583,
      "loss": 4.9589,
      "mean_token_accuracy": 0.20209725052118302,
      "num_tokens": 72553966.0,
      "step": 41825
    },
    {
      "entropy": 5.845506858825684,
      "epoch": 3.254503014977631,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003954552374672042,
      "loss": 5.009,
      "mean_token_accuracy": 0.19830774515867233,
      "num_tokens": 72563050.0,
      "step": 41830
    },
    {
      "entropy": 5.842481327056885,
      "epoch": 3.2548920443493485,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003954318272438862,
      "loss": 4.9499,
      "mean_token_accuracy": 0.20917630344629287,
      "num_tokens": 72571535.0,
      "step": 41835
    },
    {
      "entropy": 5.892104434967041,
      "epoch": 3.255281073721066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003954084151932067,
      "loss": 5.0658,
      "mean_token_accuracy": 0.20261505246162415,
      "num_tokens": 72580217.0,
      "step": 41840
    },
    {
      "entropy": 5.834992218017578,
      "epoch": 3.2556701030927835,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003953850013155214,
      "loss": 4.9423,
      "mean_token_accuracy": 0.20555755048990249,
      "num_tokens": 72588173.0,
      "step": 41845
    },
    {
      "entropy": 5.811919689178467,
      "epoch": 3.256059132464501,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003953615856111852,
      "loss": 4.9188,
      "mean_token_accuracy": 0.21008205711841582,
      "num_tokens": 72596637.0,
      "step": 41850
    },
    {
      "entropy": 5.851525592803955,
      "epoch": 3.256448161836219,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039533816808055345,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21256849616765977,
      "num_tokens": 72605240.0,
      "step": 41855
    },
    {
      "entropy": 5.805837488174438,
      "epoch": 3.256837191207936,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039531474872398165,
      "loss": 4.9204,
      "mean_token_accuracy": 0.2113334909081459,
      "num_tokens": 72613566.0,
      "step": 41860
    },
    {
      "entropy": 5.837501525878906,
      "epoch": 3.257226220579654,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000395291327541825,
      "loss": 4.9438,
      "mean_token_accuracy": 0.21383706480264664,
      "num_tokens": 72622343.0,
      "step": 41865
    },
    {
      "entropy": 5.890217208862305,
      "epoch": 3.257615249951371,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039526790453443887,
      "loss": 5.0488,
      "mean_token_accuracy": 0.2085500255227089,
      "num_tokens": 72630864.0,
      "step": 41870
    },
    {
      "entropy": 5.9177717685699465,
      "epoch": 3.258004279323089,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039524447970217875,
      "loss": 5.0961,
      "mean_token_accuracy": 0.20058230459690093,
      "num_tokens": 72639990.0,
      "step": 41875
    },
    {
      "entropy": 5.86585750579834,
      "epoch": 3.2583933086948065,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039522105304540007,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21100347340106965,
      "num_tokens": 72648739.0,
      "step": 41880
    },
    {
      "entropy": 5.850292825698853,
      "epoch": 3.258782338066524,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003951976245644582,
      "loss": 4.99,
      "mean_token_accuracy": 0.21081506311893464,
      "num_tokens": 72657981.0,
      "step": 41885
    },
    {
      "entropy": 5.811745119094849,
      "epoch": 3.2591713674382414,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003951741942597087,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21941595226526261,
      "num_tokens": 72666568.0,
      "step": 41890
    },
    {
      "entropy": 5.839476442337036,
      "epoch": 3.259560396809959,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039515076213150714,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21191552132368088,
      "num_tokens": 72674752.0,
      "step": 41895
    },
    {
      "entropy": 5.840234899520874,
      "epoch": 3.259949426181677,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039512732818020903,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20724097043275833,
      "num_tokens": 72683498.0,
      "step": 41900
    },
    {
      "entropy": 5.835481309890747,
      "epoch": 3.260338455553394,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003951038924061698,
      "loss": 4.9349,
      "mean_token_accuracy": 0.21460264623165132,
      "num_tokens": 72693152.0,
      "step": 41905
    },
    {
      "entropy": 5.8911693572998045,
      "epoch": 3.2607274849251118,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039508045480974517,
      "loss": 4.9391,
      "mean_token_accuracy": 0.20620122998952867,
      "num_tokens": 72701860.0,
      "step": 41910
    },
    {
      "entropy": 5.8231535911560055,
      "epoch": 3.2611165142968295,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039505701539129087,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2125685393810272,
      "num_tokens": 72709534.0,
      "step": 41915
    },
    {
      "entropy": 5.7989319324493405,
      "epoch": 3.261505543668547,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039503357415116236,
      "loss": 4.9688,
      "mean_token_accuracy": 0.20856141448020935,
      "num_tokens": 72718275.0,
      "step": 41920
    },
    {
      "entropy": 5.89702091217041,
      "epoch": 3.2618945730402644,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039501013108971547,
      "loss": 5.0247,
      "mean_token_accuracy": 0.2130662113428116,
      "num_tokens": 72727208.0,
      "step": 41925
    },
    {
      "entropy": 5.83869686126709,
      "epoch": 3.262283602411982,
      "grad_norm": 1.375,
      "learning_rate": 0.0003949866862073059,
      "loss": 4.9114,
      "mean_token_accuracy": 0.2141250416636467,
      "num_tokens": 72735346.0,
      "step": 41930
    },
    {
      "entropy": 5.843750619888306,
      "epoch": 3.2626726317837,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039496323950428925,
      "loss": 4.9929,
      "mean_token_accuracy": 0.21010990291833878,
      "num_tokens": 72745424.0,
      "step": 41935
    },
    {
      "entropy": 5.859559059143066,
      "epoch": 3.263061661155417,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003949397909810213,
      "loss": 4.9772,
      "mean_token_accuracy": 0.2041515365242958,
      "num_tokens": 72754773.0,
      "step": 41940
    },
    {
      "entropy": 5.835443687438965,
      "epoch": 3.2634506905271348,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039491634063785813,
      "loss": 4.8451,
      "mean_token_accuracy": 0.214905746281147,
      "num_tokens": 72762761.0,
      "step": 41945
    },
    {
      "entropy": 5.8295269966125485,
      "epoch": 3.2638397198988525,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039489288847515525,
      "loss": 4.9731,
      "mean_token_accuracy": 0.20945342183113097,
      "num_tokens": 72771578.0,
      "step": 41950
    },
    {
      "entropy": 5.8990583419799805,
      "epoch": 3.26422874927057,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003948694344932687,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20886601954698564,
      "num_tokens": 72780671.0,
      "step": 41955
    },
    {
      "entropy": 5.908766222000122,
      "epoch": 3.2646177786422874,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003948459786925542,
      "loss": 4.9756,
      "mean_token_accuracy": 0.2106522426009178,
      "num_tokens": 72789452.0,
      "step": 41960
    },
    {
      "entropy": 5.8628747940063475,
      "epoch": 3.265006808014005,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003948225210733677,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2127100110054016,
      "num_tokens": 72797327.0,
      "step": 41965
    },
    {
      "entropy": 5.820798063278199,
      "epoch": 3.265395837385723,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003947990616360652,
      "loss": 4.9742,
      "mean_token_accuracy": 0.21106703132390975,
      "num_tokens": 72806344.0,
      "step": 41970
    },
    {
      "entropy": 5.8147258281707765,
      "epoch": 3.26578486675744,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003947756003810026,
      "loss": 4.9547,
      "mean_token_accuracy": 0.21391002535820008,
      "num_tokens": 72815144.0,
      "step": 41975
    },
    {
      "entropy": 5.873108053207398,
      "epoch": 3.2661738961291578,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003947521373085359,
      "loss": 5.0245,
      "mean_token_accuracy": 0.20321057140827178,
      "num_tokens": 72824160.0,
      "step": 41980
    },
    {
      "entropy": 5.915070581436157,
      "epoch": 3.2665629255008755,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003947286724190212,
      "loss": 5.0948,
      "mean_token_accuracy": 0.20254190415143966,
      "num_tokens": 72833155.0,
      "step": 41985
    },
    {
      "entropy": 5.7415882587432865,
      "epoch": 3.2669519548725927,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039470520571281443,
      "loss": 4.8283,
      "mean_token_accuracy": 0.22486157864332199,
      "num_tokens": 72842260.0,
      "step": 41990
    },
    {
      "entropy": 5.836046981811523,
      "epoch": 3.2673409842443104,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039468173719027163,
      "loss": 4.9193,
      "mean_token_accuracy": 0.21455041021108628,
      "num_tokens": 72851434.0,
      "step": 41995
    },
    {
      "entropy": 5.821157550811767,
      "epoch": 3.267730013616028,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0003946582668517491,
      "loss": 4.964,
      "mean_token_accuracy": 0.20854993760585785,
      "num_tokens": 72859521.0,
      "step": 42000
    },
    {
      "epoch": 3.267730013616028,
      "eval_entropy": 5.556047858035369,
      "eval_loss": 5.070876121520996,
      "eval_mean_token_accuracy": 0.21527995658649401,
      "eval_num_tokens": 72859521.0,
      "eval_runtime": 21.6295,
      "eval_samples_per_second": 1921.36,
      "eval_steps_per_second": 240.182,
      "step": 42000
    },
    {
      "entropy": 5.758299922943115,
      "epoch": 3.2681190429877454,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003946347946976026,
      "loss": 4.9248,
      "mean_token_accuracy": 0.20697599202394484,
      "num_tokens": 72868554.0,
      "step": 42005
    },
    {
      "entropy": 5.891294288635254,
      "epoch": 3.268508072359463,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003946113207281887,
      "loss": 5.0367,
      "mean_token_accuracy": 0.2071451172232628,
      "num_tokens": 72877360.0,
      "step": 42010
    },
    {
      "entropy": 5.8557299137115475,
      "epoch": 3.2688971017311808,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039458784494386337,
      "loss": 5.0243,
      "mean_token_accuracy": 0.200650355219841,
      "num_tokens": 72885962.0,
      "step": 42015
    },
    {
      "entropy": 5.912816572189331,
      "epoch": 3.2692861311028985,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039456436734498275,
      "loss": 5.0043,
      "mean_token_accuracy": 0.20639751702547074,
      "num_tokens": 72894976.0,
      "step": 42020
    },
    {
      "entropy": 5.794985103607178,
      "epoch": 3.2696751604746157,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039454088793190316,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20383837223052978,
      "num_tokens": 72903775.0,
      "step": 42025
    },
    {
      "entropy": 5.818513631820679,
      "epoch": 3.2700641898463334,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039451740670498094,
      "loss": 4.9464,
      "mean_token_accuracy": 0.2107660785317421,
      "num_tokens": 72912013.0,
      "step": 42030
    },
    {
      "entropy": 5.841211366653442,
      "epoch": 3.270453219218051,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003944939236645723,
      "loss": 4.9582,
      "mean_token_accuracy": 0.21072672307491302,
      "num_tokens": 72920204.0,
      "step": 42035
    },
    {
      "entropy": 5.843954801559448,
      "epoch": 3.2708422485897684,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003944704388110335,
      "loss": 4.9556,
      "mean_token_accuracy": 0.20791803002357484,
      "num_tokens": 72930125.0,
      "step": 42040
    },
    {
      "entropy": 5.79491696357727,
      "epoch": 3.271231277961486,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000394446952144721,
      "loss": 4.8988,
      "mean_token_accuracy": 0.2118603602051735,
      "num_tokens": 72938884.0,
      "step": 42045
    },
    {
      "entropy": 5.797326469421387,
      "epoch": 3.2716203073332037,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003944234636659911,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2066956341266632,
      "num_tokens": 72947303.0,
      "step": 42050
    },
    {
      "entropy": 5.853543996810913,
      "epoch": 3.2720093367049214,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003943999733752003,
      "loss": 5.009,
      "mean_token_accuracy": 0.20157743841409684,
      "num_tokens": 72955789.0,
      "step": 42055
    },
    {
      "entropy": 5.763695478439331,
      "epoch": 3.2723983660766387,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003943764812727048,
      "loss": 4.9076,
      "mean_token_accuracy": 0.221087484061718,
      "num_tokens": 72963931.0,
      "step": 42060
    },
    {
      "entropy": 5.764915323257446,
      "epoch": 3.2727873954483564,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039435298735886133,
      "loss": 4.9274,
      "mean_token_accuracy": 0.20477720946073533,
      "num_tokens": 72973094.0,
      "step": 42065
    },
    {
      "entropy": 5.8211160659790036,
      "epoch": 3.273176424820074,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003943294916340261,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2124040901660919,
      "num_tokens": 72981461.0,
      "step": 42070
    },
    {
      "entropy": 5.880550670623779,
      "epoch": 3.2735654541917913,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003943059940985559,
      "loss": 4.9494,
      "mean_token_accuracy": 0.21562696397304534,
      "num_tokens": 72989986.0,
      "step": 42075
    },
    {
      "entropy": 5.931149244308472,
      "epoch": 3.273954483563509,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039428249475280706,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21342407613992692,
      "num_tokens": 72998020.0,
      "step": 42080
    },
    {
      "entropy": 5.848946237564087,
      "epoch": 3.2743435129352267,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003942589935971363,
      "loss": 5.0347,
      "mean_token_accuracy": 0.2022048845887184,
      "num_tokens": 73006103.0,
      "step": 42085
    },
    {
      "entropy": 5.735710334777832,
      "epoch": 3.274732542306944,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003942354906319001,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21790630221366883,
      "num_tokens": 73014396.0,
      "step": 42090
    },
    {
      "entropy": 5.8317530155181885,
      "epoch": 3.2751215716786617,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003942119858574551,
      "loss": 4.9972,
      "mean_token_accuracy": 0.2070605367422104,
      "num_tokens": 73022714.0,
      "step": 42095
    },
    {
      "entropy": 5.856555604934693,
      "epoch": 3.2755106010503794,
      "grad_norm": 1.359375,
      "learning_rate": 0.000394188479274158,
      "loss": 5.0073,
      "mean_token_accuracy": 0.20847219228744507,
      "num_tokens": 73031295.0,
      "step": 42100
    },
    {
      "entropy": 5.928793907165527,
      "epoch": 3.2758996304220966,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039416497088236535,
      "loss": 5.0317,
      "mean_token_accuracy": 0.21158937066793443,
      "num_tokens": 73039728.0,
      "step": 42105
    },
    {
      "entropy": 5.860816717147827,
      "epoch": 3.2762886597938143,
      "grad_norm": 1.21875,
      "learning_rate": 0.000394141460682434,
      "loss": 4.9816,
      "mean_token_accuracy": 0.20716286897659303,
      "num_tokens": 73049125.0,
      "step": 42110
    },
    {
      "entropy": 5.820709323883056,
      "epoch": 3.276677689165532,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003941179486747206,
      "loss": 5.0058,
      "mean_token_accuracy": 0.2053380087018013,
      "num_tokens": 73056992.0,
      "step": 42115
    },
    {
      "entropy": 5.782445192337036,
      "epoch": 3.2770667185372497,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003940944348595819,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2146309033036232,
      "num_tokens": 73066866.0,
      "step": 42120
    },
    {
      "entropy": 5.85546727180481,
      "epoch": 3.277455747908967,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039407091923737475,
      "loss": 4.959,
      "mean_token_accuracy": 0.2142377495765686,
      "num_tokens": 73075961.0,
      "step": 42125
    },
    {
      "entropy": 5.894864177703857,
      "epoch": 3.2778447772806847,
      "grad_norm": 1.25,
      "learning_rate": 0.00039404740180845594,
      "loss": 5.0792,
      "mean_token_accuracy": 0.20389993935823442,
      "num_tokens": 73085711.0,
      "step": 42130
    },
    {
      "entropy": 5.827836275100708,
      "epoch": 3.2782338066524024,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039402388257318236,
      "loss": 4.8353,
      "mean_token_accuracy": 0.21816546320915223,
      "num_tokens": 73094303.0,
      "step": 42135
    },
    {
      "entropy": 5.855353927612304,
      "epoch": 3.2786228360241196,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003940003615319106,
      "loss": 4.9331,
      "mean_token_accuracy": 0.2054259806871414,
      "num_tokens": 73102703.0,
      "step": 42140
    },
    {
      "entropy": 5.827605533599853,
      "epoch": 3.2790118653958373,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039397683868499784,
      "loss": 4.9152,
      "mean_token_accuracy": 0.20903125554323196,
      "num_tokens": 73111823.0,
      "step": 42145
    },
    {
      "entropy": 5.799358367919922,
      "epoch": 3.279400894767555,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000393953314032801,
      "loss": 4.9806,
      "mean_token_accuracy": 0.20332070887088777,
      "num_tokens": 73121135.0,
      "step": 42150
    },
    {
      "entropy": 5.872877025604248,
      "epoch": 3.2797899241392727,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003939297875756769,
      "loss": 4.9686,
      "mean_token_accuracy": 0.20659374296665192,
      "num_tokens": 73129317.0,
      "step": 42155
    },
    {
      "entropy": 5.879653978347778,
      "epoch": 3.28017895351099,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003939062593139825,
      "loss": 4.9476,
      "mean_token_accuracy": 0.2098517194390297,
      "num_tokens": 73138395.0,
      "step": 42160
    },
    {
      "entropy": 5.781131315231323,
      "epoch": 3.2805679828827077,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000393882729248075,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21325925439596177,
      "num_tokens": 73146272.0,
      "step": 42165
    },
    {
      "entropy": 5.798563241958618,
      "epoch": 3.2809570122544254,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039385919737831127,
      "loss": 4.9883,
      "mean_token_accuracy": 0.2085891678929329,
      "num_tokens": 73154630.0,
      "step": 42170
    },
    {
      "entropy": 5.885260057449341,
      "epoch": 3.2813460416261426,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039383566370504837,
      "loss": 4.9125,
      "mean_token_accuracy": 0.21565323323011398,
      "num_tokens": 73162857.0,
      "step": 42175
    },
    {
      "entropy": 5.874507713317871,
      "epoch": 3.2817350709978603,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003938121282286434,
      "loss": 5.0211,
      "mean_token_accuracy": 0.20456851422786712,
      "num_tokens": 73171197.0,
      "step": 42180
    },
    {
      "entropy": 5.801846742630005,
      "epoch": 3.282124100369578,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039378859094945366,
      "loss": 4.8511,
      "mean_token_accuracy": 0.21923355311155318,
      "num_tokens": 73179863.0,
      "step": 42185
    },
    {
      "entropy": 5.806522083282471,
      "epoch": 3.2825131297412953,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003937650518678359,
      "loss": 4.9325,
      "mean_token_accuracy": 0.20902454257011413,
      "num_tokens": 73189129.0,
      "step": 42190
    },
    {
      "entropy": 5.836764907836914,
      "epoch": 3.282902159113013,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003937415109841476,
      "loss": 4.9903,
      "mean_token_accuracy": 0.1997587114572525,
      "num_tokens": 73198361.0,
      "step": 42195
    },
    {
      "entropy": 5.863855361938477,
      "epoch": 3.2832911884847307,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003937179682987459,
      "loss": 5.0249,
      "mean_token_accuracy": 0.20621461868286134,
      "num_tokens": 73206662.0,
      "step": 42200
    },
    {
      "entropy": 5.850204849243164,
      "epoch": 3.283680217856448,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039369442381198793,
      "loss": 5.0262,
      "mean_token_accuracy": 0.20960848778486252,
      "num_tokens": 73214836.0,
      "step": 42205
    },
    {
      "entropy": 5.794151735305786,
      "epoch": 3.2840692472281656,
      "grad_norm": 1.359375,
      "learning_rate": 0.000393670877524231,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22028219252824782,
      "num_tokens": 73222920.0,
      "step": 42210
    },
    {
      "entropy": 5.83977575302124,
      "epoch": 3.2844582765998833,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003936473294358325,
      "loss": 4.9501,
      "mean_token_accuracy": 0.2130502939224243,
      "num_tokens": 73232298.0,
      "step": 42215
    },
    {
      "entropy": 5.8012042999267575,
      "epoch": 3.284847305971601,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003936237795471495,
      "loss": 4.881,
      "mean_token_accuracy": 0.2207501322031021,
      "num_tokens": 73240972.0,
      "step": 42220
    },
    {
      "entropy": 5.799656915664673,
      "epoch": 3.2852363353433183,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039360022785853957,
      "loss": 4.9396,
      "mean_token_accuracy": 0.2151234045624733,
      "num_tokens": 73249140.0,
      "step": 42225
    },
    {
      "entropy": 5.875300502777099,
      "epoch": 3.285625364715036,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003935766743703599,
      "loss": 4.9697,
      "mean_token_accuracy": 0.20288372039794922,
      "num_tokens": 73258434.0,
      "step": 42230
    },
    {
      "entropy": 5.86159496307373,
      "epoch": 3.2860143940867537,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039355311908296787,
      "loss": 4.9826,
      "mean_token_accuracy": 0.2001311868429184,
      "num_tokens": 73266745.0,
      "step": 42235
    },
    {
      "entropy": 5.899941349029541,
      "epoch": 3.286403423458471,
      "grad_norm": 1.296875,
      "learning_rate": 0.000393529561996721,
      "loss": 5.0298,
      "mean_token_accuracy": 0.2089293897151947,
      "num_tokens": 73275072.0,
      "step": 42240
    },
    {
      "entropy": 5.8372656345367435,
      "epoch": 3.2867924528301886,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039350600311197667,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21633943319320678,
      "num_tokens": 73284402.0,
      "step": 42245
    },
    {
      "entropy": 5.879165506362915,
      "epoch": 3.2871814822019063,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039348244242909234,
      "loss": 5.0121,
      "mean_token_accuracy": 0.2120365619659424,
      "num_tokens": 73293870.0,
      "step": 42250
    },
    {
      "entropy": 5.887145948410034,
      "epoch": 3.287570511573624,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039345887994842566,
      "loss": 5.0188,
      "mean_token_accuracy": 0.20394021421670913,
      "num_tokens": 73303196.0,
      "step": 42255
    },
    {
      "entropy": 5.854613351821899,
      "epoch": 3.2879595409453413,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003934353156703339,
      "loss": 4.9185,
      "mean_token_accuracy": 0.20981398969888687,
      "num_tokens": 73311549.0,
      "step": 42260
    },
    {
      "entropy": 5.807502412796021,
      "epoch": 3.288348570317059,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003934117495951748,
      "loss": 4.9344,
      "mean_token_accuracy": 0.2147183522582054,
      "num_tokens": 73321126.0,
      "step": 42265
    },
    {
      "entropy": 5.787750625610352,
      "epoch": 3.2887375996887767,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003933881817233058,
      "loss": 4.8459,
      "mean_token_accuracy": 0.22305065840482713,
      "num_tokens": 73330207.0,
      "step": 42270
    },
    {
      "entropy": 5.898659706115723,
      "epoch": 3.289126629060494,
      "grad_norm": 1.375,
      "learning_rate": 0.00039336461205508456,
      "loss": 5.0216,
      "mean_token_accuracy": 0.21271206736564635,
      "num_tokens": 73339099.0,
      "step": 42275
    },
    {
      "entropy": 5.7973545551300045,
      "epoch": 3.2895156584322116,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003933410405908689,
      "loss": 4.8763,
      "mean_token_accuracy": 0.21505739390850068,
      "num_tokens": 73347566.0,
      "step": 42280
    },
    {
      "entropy": 5.874573469161987,
      "epoch": 3.2899046878039293,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039331746733101613,
      "loss": 5.0001,
      "mean_token_accuracy": 0.2091255322098732,
      "num_tokens": 73355909.0,
      "step": 42285
    },
    {
      "entropy": 5.805905771255493,
      "epoch": 3.290293717175647,
      "grad_norm": 1.421875,
      "learning_rate": 0.00039329389227588416,
      "loss": 4.9147,
      "mean_token_accuracy": 0.21106512993574142,
      "num_tokens": 73365233.0,
      "step": 42290
    },
    {
      "entropy": 5.875520658493042,
      "epoch": 3.2906827465473643,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039327031542583067,
      "loss": 5.0217,
      "mean_token_accuracy": 0.21213218122720717,
      "num_tokens": 73373623.0,
      "step": 42295
    },
    {
      "entropy": 5.827519130706787,
      "epoch": 3.291071775919082,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039324673678121344,
      "loss": 4.9402,
      "mean_token_accuracy": 0.2082584023475647,
      "num_tokens": 73381668.0,
      "step": 42300
    },
    {
      "entropy": 5.832705974578857,
      "epoch": 3.291460805290799,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003932231563423901,
      "loss": 5.0243,
      "mean_token_accuracy": 0.20912815779447555,
      "num_tokens": 73390284.0,
      "step": 42305
    },
    {
      "entropy": 5.846575975418091,
      "epoch": 3.291849834662517,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003931995741097186,
      "loss": 4.9427,
      "mean_token_accuracy": 0.2177132323384285,
      "num_tokens": 73398704.0,
      "step": 42310
    },
    {
      "entropy": 5.863977479934692,
      "epoch": 3.2922388640342346,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039317599008355665,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20509020686149598,
      "num_tokens": 73407270.0,
      "step": 42315
    },
    {
      "entropy": 5.885476016998291,
      "epoch": 3.2926278934059523,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003931524042642622,
      "loss": 5.0074,
      "mean_token_accuracy": 0.21272029280662536,
      "num_tokens": 73416158.0,
      "step": 42320
    },
    {
      "entropy": 5.883214378356934,
      "epoch": 3.2930169227776696,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003931288166521931,
      "loss": 5.0736,
      "mean_token_accuracy": 0.20418912321329116,
      "num_tokens": 73424823.0,
      "step": 42325
    },
    {
      "entropy": 5.90181074142456,
      "epoch": 3.2934059521493873,
      "grad_norm": 1.421875,
      "learning_rate": 0.00039310522724770713,
      "loss": 4.9783,
      "mean_token_accuracy": 0.21451079994440078,
      "num_tokens": 73432699.0,
      "step": 42330
    },
    {
      "entropy": 5.78568320274353,
      "epoch": 3.293794981521105,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039308163605116235,
      "loss": 4.972,
      "mean_token_accuracy": 0.20922284573316574,
      "num_tokens": 73441154.0,
      "step": 42335
    },
    {
      "entropy": 5.8774559020996096,
      "epoch": 3.294184010892822,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003930580430629167,
      "loss": 5.0968,
      "mean_token_accuracy": 0.20230432003736495,
      "num_tokens": 73449804.0,
      "step": 42340
    },
    {
      "entropy": 5.880717039108276,
      "epoch": 3.29457304026454,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003930344482833281,
      "loss": 4.9617,
      "mean_token_accuracy": 0.2074015974998474,
      "num_tokens": 73458546.0,
      "step": 42345
    },
    {
      "entropy": 5.899647378921509,
      "epoch": 3.2949620696362576,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039301085171275475,
      "loss": 4.9753,
      "mean_token_accuracy": 0.20935631692409515,
      "num_tokens": 73467404.0,
      "step": 42350
    },
    {
      "entropy": 5.814826726913452,
      "epoch": 3.2953510990079753,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003929872533515545,
      "loss": 4.9885,
      "mean_token_accuracy": 0.21057751327753066,
      "num_tokens": 73476520.0,
      "step": 42355
    },
    {
      "entropy": 5.806273746490478,
      "epoch": 3.2957401283796925,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039296365320008546,
      "loss": 5.0172,
      "mean_token_accuracy": 0.20643816143274307,
      "num_tokens": 73485927.0,
      "step": 42360
    },
    {
      "entropy": 5.890439701080322,
      "epoch": 3.2961291577514102,
      "grad_norm": 1.5,
      "learning_rate": 0.00039294005125870576,
      "loss": 4.9892,
      "mean_token_accuracy": 0.20817520320415497,
      "num_tokens": 73493971.0,
      "step": 42365
    },
    {
      "entropy": 5.789645957946777,
      "epoch": 3.296518187123128,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039291644752777346,
      "loss": 4.9021,
      "mean_token_accuracy": 0.20926578044891359,
      "num_tokens": 73502565.0,
      "step": 42370
    },
    {
      "entropy": 5.821596908569336,
      "epoch": 3.296907216494845,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039289284200764676,
      "loss": 4.9834,
      "mean_token_accuracy": 0.21939948052167893,
      "num_tokens": 73511306.0,
      "step": 42375
    },
    {
      "entropy": 5.886881351470947,
      "epoch": 3.297296245866563,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003928692346986839,
      "loss": 4.9511,
      "mean_token_accuracy": 0.20738032460212708,
      "num_tokens": 73520054.0,
      "step": 42380
    },
    {
      "entropy": 5.782700109481811,
      "epoch": 3.2976852752382806,
      "grad_norm": 1.453125,
      "learning_rate": 0.00039284562560124297,
      "loss": 4.8406,
      "mean_token_accuracy": 0.21730977892875672,
      "num_tokens": 73528108.0,
      "step": 42385
    },
    {
      "entropy": 5.829644584655762,
      "epoch": 3.2980743046099983,
      "grad_norm": 1.375,
      "learning_rate": 0.0003928220147156822,
      "loss": 5.013,
      "mean_token_accuracy": 0.2090679094195366,
      "num_tokens": 73536478.0,
      "step": 42390
    },
    {
      "entropy": 5.799276876449585,
      "epoch": 3.2984633339817155,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039279840204235994,
      "loss": 4.913,
      "mean_token_accuracy": 0.21344873309135437,
      "num_tokens": 73544550.0,
      "step": 42395
    },
    {
      "entropy": 5.8829535961151125,
      "epoch": 3.2988523633534332,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039277478758163443,
      "loss": 4.9782,
      "mean_token_accuracy": 0.2101181596517563,
      "num_tokens": 73552982.0,
      "step": 42400
    },
    {
      "entropy": 5.827280569076538,
      "epoch": 3.299241392725151,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000392751171333864,
      "loss": 4.9473,
      "mean_token_accuracy": 0.2102155014872551,
      "num_tokens": 73562177.0,
      "step": 42405
    },
    {
      "entropy": 5.900981187820435,
      "epoch": 3.299630422096868,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039272755329940683,
      "loss": 5.052,
      "mean_token_accuracy": 0.20447300970554352,
      "num_tokens": 73570552.0,
      "step": 42410
    },
    {
      "entropy": 5.8899781703948975,
      "epoch": 3.300019451468586,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039270393347862155,
      "loss": 5.07,
      "mean_token_accuracy": 0.20506585538387298,
      "num_tokens": 73580017.0,
      "step": 42415
    },
    {
      "entropy": 5.868103790283203,
      "epoch": 3.3004084808403036,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039268031187186635,
      "loss": 4.9698,
      "mean_token_accuracy": 0.21190062910318375,
      "num_tokens": 73588956.0,
      "step": 42420
    },
    {
      "entropy": 5.903385829925537,
      "epoch": 3.300797510212021,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039265668847949977,
      "loss": 5.0465,
      "mean_token_accuracy": 0.21161836981773377,
      "num_tokens": 73598440.0,
      "step": 42425
    },
    {
      "entropy": 5.825795030593872,
      "epoch": 3.3011865395837385,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003926330633018802,
      "loss": 5.0022,
      "mean_token_accuracy": 0.20690242499113082,
      "num_tokens": 73607780.0,
      "step": 42430
    },
    {
      "entropy": 5.844006967544556,
      "epoch": 3.3015755689554562,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039260943633936615,
      "loss": 5.0097,
      "mean_token_accuracy": 0.20750716626644133,
      "num_tokens": 73616816.0,
      "step": 42435
    },
    {
      "entropy": 5.814282512664795,
      "epoch": 3.3019645983271735,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039258580759231603,
      "loss": 4.9166,
      "mean_token_accuracy": 0.21747713088989257,
      "num_tokens": 73624546.0,
      "step": 42440
    },
    {
      "entropy": 5.78208818435669,
      "epoch": 3.302353627698891,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003925621770610885,
      "loss": 4.8454,
      "mean_token_accuracy": 0.22139570415019988,
      "num_tokens": 73632917.0,
      "step": 42445
    },
    {
      "entropy": 5.729669618606567,
      "epoch": 3.302742657070609,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000392538544746042,
      "loss": 4.8157,
      "mean_token_accuracy": 0.21723325550556183,
      "num_tokens": 73641262.0,
      "step": 42450
    },
    {
      "entropy": 5.7563389301300045,
      "epoch": 3.3031316864423266,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039251491064753525,
      "loss": 4.9182,
      "mean_token_accuracy": 0.20476796478033066,
      "num_tokens": 73649751.0,
      "step": 42455
    },
    {
      "entropy": 5.82024917602539,
      "epoch": 3.303520715814044,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003924912747659267,
      "loss": 5.0101,
      "mean_token_accuracy": 0.1986706718802452,
      "num_tokens": 73658697.0,
      "step": 42460
    },
    {
      "entropy": 5.820790815353393,
      "epoch": 3.3039097451857615,
      "grad_norm": 1.25,
      "learning_rate": 0.00039246763710157513,
      "loss": 4.9353,
      "mean_token_accuracy": 0.21357878893613816,
      "num_tokens": 73667232.0,
      "step": 42465
    },
    {
      "entropy": 5.828835678100586,
      "epoch": 3.3042987745574792,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003924439976548391,
      "loss": 4.9486,
      "mean_token_accuracy": 0.2115991234779358,
      "num_tokens": 73675531.0,
      "step": 42470
    },
    {
      "entropy": 5.837472343444825,
      "epoch": 3.3046878039291965,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039242035642607733,
      "loss": 4.9712,
      "mean_token_accuracy": 0.2053645834326744,
      "num_tokens": 73683312.0,
      "step": 42475
    },
    {
      "entropy": 5.880854797363281,
      "epoch": 3.305076833300914,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003923967134156486,
      "loss": 5.0949,
      "mean_token_accuracy": 0.20000273883342742,
      "num_tokens": 73691929.0,
      "step": 42480
    },
    {
      "entropy": 5.805250883102417,
      "epoch": 3.305465862672632,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003923730686239115,
      "loss": 4.9719,
      "mean_token_accuracy": 0.21509898751974105,
      "num_tokens": 73700643.0,
      "step": 42485
    },
    {
      "entropy": 5.858594608306885,
      "epoch": 3.3058548920443496,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003923494220512249,
      "loss": 4.9422,
      "mean_token_accuracy": 0.2096325933933258,
      "num_tokens": 73709027.0,
      "step": 42490
    },
    {
      "entropy": 5.889338445663452,
      "epoch": 3.306243921416067,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003923257736979476,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21169265359640121,
      "num_tokens": 73717305.0,
      "step": 42495
    },
    {
      "entropy": 5.86341118812561,
      "epoch": 3.3066329507877845,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039230212356443855,
      "loss": 5.0687,
      "mean_token_accuracy": 0.20710794925689696,
      "num_tokens": 73725900.0,
      "step": 42500
    },
    {
      "entropy": 5.826107454299927,
      "epoch": 3.307021980159502,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039227847165105635,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21795350015163423,
      "num_tokens": 73734954.0,
      "step": 42505
    },
    {
      "entropy": 5.8410398960113525,
      "epoch": 3.3074110095312195,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039225481795816016,
      "loss": 4.9491,
      "mean_token_accuracy": 0.21102227568626403,
      "num_tokens": 73743206.0,
      "step": 42510
    },
    {
      "entropy": 5.811402893066406,
      "epoch": 3.307800038902937,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003922311624861086,
      "loss": 4.9566,
      "mean_token_accuracy": 0.2177676022052765,
      "num_tokens": 73752094.0,
      "step": 42515
    },
    {
      "entropy": 5.826411390304566,
      "epoch": 3.308189068274655,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003922075052352608,
      "loss": 4.9566,
      "mean_token_accuracy": 0.2172113314270973,
      "num_tokens": 73761623.0,
      "step": 42520
    },
    {
      "entropy": 5.888113498687744,
      "epoch": 3.308578097646372,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003921838462059756,
      "loss": 5.0165,
      "mean_token_accuracy": 0.21313211768865586,
      "num_tokens": 73770125.0,
      "step": 42525
    },
    {
      "entropy": 5.8751214981079105,
      "epoch": 3.30896712701809,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039216018539861217,
      "loss": 5.0546,
      "mean_token_accuracy": 0.20197821408510208,
      "num_tokens": 73778683.0,
      "step": 42530
    },
    {
      "entropy": 5.90404748916626,
      "epoch": 3.3093561563898075,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003921365228135293,
      "loss": 4.9792,
      "mean_token_accuracy": 0.2064994290471077,
      "num_tokens": 73787014.0,
      "step": 42535
    },
    {
      "entropy": 5.813021945953369,
      "epoch": 3.3097451857615248,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039211285845108617,
      "loss": 4.9395,
      "mean_token_accuracy": 0.2167838454246521,
      "num_tokens": 73795812.0,
      "step": 42540
    },
    {
      "entropy": 5.840122270584106,
      "epoch": 3.3101342151332425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003920891923116418,
      "loss": 4.9877,
      "mean_token_accuracy": 0.20903054922819136,
      "num_tokens": 73805029.0,
      "step": 42545
    },
    {
      "entropy": 5.811117219924927,
      "epoch": 3.31052324450496,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039206552439555544,
      "loss": 4.9887,
      "mean_token_accuracy": 0.21025339215993882,
      "num_tokens": 73813456.0,
      "step": 42550
    },
    {
      "entropy": 5.930837726593017,
      "epoch": 3.310912273876678,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003920418547031859,
      "loss": 5.055,
      "mean_token_accuracy": 0.20942474901676178,
      "num_tokens": 73822048.0,
      "step": 42555
    },
    {
      "entropy": 5.830178213119507,
      "epoch": 3.311301303248395,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003920181832348926,
      "loss": 4.9764,
      "mean_token_accuracy": 0.21472700834274291,
      "num_tokens": 73831600.0,
      "step": 42560
    },
    {
      "entropy": 5.837258386611938,
      "epoch": 3.311690332620113,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003919945099910346,
      "loss": 4.9742,
      "mean_token_accuracy": 0.20774897187948227,
      "num_tokens": 73840633.0,
      "step": 42565
    },
    {
      "entropy": 5.799410963058472,
      "epoch": 3.3120793619918305,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039197083497197107,
      "loss": 4.8856,
      "mean_token_accuracy": 0.21513110846281053,
      "num_tokens": 73849329.0,
      "step": 42570
    },
    {
      "entropy": 5.817235040664673,
      "epoch": 3.3124683913635478,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003919471581780613,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21262370198965072,
      "num_tokens": 73858613.0,
      "step": 42575
    },
    {
      "entropy": 5.80995626449585,
      "epoch": 3.3128574207352655,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003919234796096646,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21544256210327148,
      "num_tokens": 73867326.0,
      "step": 42580
    },
    {
      "entropy": 5.823333644866944,
      "epoch": 3.313246450106983,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003918997992671402,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2091308131814003,
      "num_tokens": 73875902.0,
      "step": 42585
    },
    {
      "entropy": 5.943651580810547,
      "epoch": 3.313635479478701,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003918761171508475,
      "loss": 5.0334,
      "mean_token_accuracy": 0.21149638146162034,
      "num_tokens": 73885343.0,
      "step": 42590
    },
    {
      "entropy": 5.926949214935303,
      "epoch": 3.314024508850418,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003918524332611456,
      "loss": 5.0997,
      "mean_token_accuracy": 0.20311731994152069,
      "num_tokens": 73893302.0,
      "step": 42595
    },
    {
      "entropy": 5.829248237609863,
      "epoch": 3.314413538222136,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003918287475983941,
      "loss": 4.9641,
      "mean_token_accuracy": 0.21530821919441223,
      "num_tokens": 73902178.0,
      "step": 42600
    },
    {
      "entropy": 5.815505599975586,
      "epoch": 3.3148025675938535,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039180506016295226,
      "loss": 4.911,
      "mean_token_accuracy": 0.21545563340187074,
      "num_tokens": 73911345.0,
      "step": 42605
    },
    {
      "entropy": 5.84421067237854,
      "epoch": 3.3151915969655708,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039178137095517956,
      "loss": 4.9297,
      "mean_token_accuracy": 0.21338322162628173,
      "num_tokens": 73919693.0,
      "step": 42610
    },
    {
      "entropy": 5.771155118942261,
      "epoch": 3.3155806263372885,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003917576799754354,
      "loss": 4.877,
      "mean_token_accuracy": 0.21517099738121032,
      "num_tokens": 73927962.0,
      "step": 42615
    },
    {
      "entropy": 5.831697034835815,
      "epoch": 3.315969655709006,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003917339872240793,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20830419957637786,
      "num_tokens": 73936433.0,
      "step": 42620
    },
    {
      "entropy": 5.911880016326904,
      "epoch": 3.3163586850807234,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003917102927014708,
      "loss": 5.0054,
      "mean_token_accuracy": 0.20829056799411774,
      "num_tokens": 73944964.0,
      "step": 42625
    },
    {
      "entropy": 5.816055536270142,
      "epoch": 3.316747714452441,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003916865964079694,
      "loss": 4.904,
      "mean_token_accuracy": 0.2189040571451187,
      "num_tokens": 73954028.0,
      "step": 42630
    },
    {
      "entropy": 5.79279580116272,
      "epoch": 3.317136743824159,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003916628983439345,
      "loss": 4.9396,
      "mean_token_accuracy": 0.2087011843919754,
      "num_tokens": 73962923.0,
      "step": 42635
    },
    {
      "entropy": 5.7940850257873535,
      "epoch": 3.317525773195876,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039163919850972585,
      "loss": 4.991,
      "mean_token_accuracy": 0.2098859280347824,
      "num_tokens": 73971183.0,
      "step": 42640
    },
    {
      "entropy": 5.8327020645141605,
      "epoch": 3.3179148025675937,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003916154969057031,
      "loss": 4.9935,
      "mean_token_accuracy": 0.21230850964784623,
      "num_tokens": 73979248.0,
      "step": 42645
    },
    {
      "entropy": 5.856299877166748,
      "epoch": 3.3183038319393114,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039159179353222566,
      "loss": 4.9958,
      "mean_token_accuracy": 0.2050314500927925,
      "num_tokens": 73987869.0,
      "step": 42650
    },
    {
      "entropy": 5.857276868820191,
      "epoch": 3.318692861311029,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039156808838965336,
      "loss": 4.9938,
      "mean_token_accuracy": 0.20746403336524963,
      "num_tokens": 73996562.0,
      "step": 42655
    },
    {
      "entropy": 5.859234094619751,
      "epoch": 3.3190818906827464,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039154438147834595,
      "loss": 4.958,
      "mean_token_accuracy": 0.20780195146799088,
      "num_tokens": 74005202.0,
      "step": 42660
    },
    {
      "entropy": 5.865197086334229,
      "epoch": 3.319470920054464,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039152067279866294,
      "loss": 5.0235,
      "mean_token_accuracy": 0.21197966933250428,
      "num_tokens": 74014541.0,
      "step": 42665
    },
    {
      "entropy": 5.910163640975952,
      "epoch": 3.319859949426182,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003914969623509643,
      "loss": 5.0045,
      "mean_token_accuracy": 0.21013777554035187,
      "num_tokens": 74022613.0,
      "step": 42670
    },
    {
      "entropy": 5.8958807468414305,
      "epoch": 3.320248978797899,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003914732501356096,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20537243783473969,
      "num_tokens": 74031753.0,
      "step": 42675
    },
    {
      "entropy": 5.89395546913147,
      "epoch": 3.3206380081696167,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039144953615295877,
      "loss": 5.0924,
      "mean_token_accuracy": 0.20061901062726975,
      "num_tokens": 74040096.0,
      "step": 42680
    },
    {
      "entropy": 5.8199724674224855,
      "epoch": 3.3210270375413344,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039142582040337147,
      "loss": 5.0221,
      "mean_token_accuracy": 0.21078623980283737,
      "num_tokens": 74048533.0,
      "step": 42685
    },
    {
      "entropy": 5.8597313404083256,
      "epoch": 3.321416066913052,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003914021028872077,
      "loss": 4.9546,
      "mean_token_accuracy": 0.21374624818563462,
      "num_tokens": 74057871.0,
      "step": 42690
    },
    {
      "entropy": 5.894290161132813,
      "epoch": 3.3218050962847694,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003913783836048274,
      "loss": 4.9634,
      "mean_token_accuracy": 0.211209736764431,
      "num_tokens": 74066343.0,
      "step": 42695
    },
    {
      "entropy": 5.835529088973999,
      "epoch": 3.322194125656487,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003913546625565902,
      "loss": 4.9071,
      "mean_token_accuracy": 0.21516996175050734,
      "num_tokens": 74074784.0,
      "step": 42700
    },
    {
      "entropy": 5.811766481399536,
      "epoch": 3.322583155028205,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039133093974285635,
      "loss": 4.9577,
      "mean_token_accuracy": 0.2149084210395813,
      "num_tokens": 74083708.0,
      "step": 42705
    },
    {
      "entropy": 5.759116840362549,
      "epoch": 3.322972184399922,
      "grad_norm": 1.25,
      "learning_rate": 0.00039130721516398556,
      "loss": 4.8279,
      "mean_token_accuracy": 0.21692177653312683,
      "num_tokens": 74092328.0,
      "step": 42710
    },
    {
      "entropy": 5.8838074684143065,
      "epoch": 3.3233612137716397,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003912834888203379,
      "loss": 4.9995,
      "mean_token_accuracy": 0.20795153826475143,
      "num_tokens": 74100507.0,
      "step": 42715
    },
    {
      "entropy": 5.83881402015686,
      "epoch": 3.3237502431433574,
      "grad_norm": 1.375,
      "learning_rate": 0.0003912597607122734,
      "loss": 4.9744,
      "mean_token_accuracy": 0.21333924382925035,
      "num_tokens": 74109317.0,
      "step": 42720
    },
    {
      "entropy": 5.863425827026367,
      "epoch": 3.324139272515075,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039123603084015204,
      "loss": 5.022,
      "mean_token_accuracy": 0.2032216578722,
      "num_tokens": 74118726.0,
      "step": 42725
    },
    {
      "entropy": 5.853871631622314,
      "epoch": 3.3245283018867924,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039121229920433394,
      "loss": 4.9127,
      "mean_token_accuracy": 0.21436996906995773,
      "num_tokens": 74127298.0,
      "step": 42730
    },
    {
      "entropy": 5.893998765945435,
      "epoch": 3.32491733125851,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003911885658051792,
      "loss": 5.0287,
      "mean_token_accuracy": 0.2038129225373268,
      "num_tokens": 74136212.0,
      "step": 42735
    },
    {
      "entropy": 5.789471673965454,
      "epoch": 3.3253063606302278,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039116483064304794,
      "loss": 4.9307,
      "mean_token_accuracy": 0.21689655482769013,
      "num_tokens": 74144835.0,
      "step": 42740
    },
    {
      "entropy": 5.803245162963867,
      "epoch": 3.325695390001945,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003911410937183002,
      "loss": 4.9584,
      "mean_token_accuracy": 0.20901769548654556,
      "num_tokens": 74153296.0,
      "step": 42745
    },
    {
      "entropy": 5.752967166900635,
      "epoch": 3.3260844193736627,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039111735503129626,
      "loss": 4.8736,
      "mean_token_accuracy": 0.218157622218132,
      "num_tokens": 74161157.0,
      "step": 42750
    },
    {
      "entropy": 5.841610050201416,
      "epoch": 3.3264734487453804,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003910936145823962,
      "loss": 5.0231,
      "mean_token_accuracy": 0.2035321295261383,
      "num_tokens": 74170035.0,
      "step": 42755
    },
    {
      "entropy": 5.863004350662232,
      "epoch": 3.3268624781170977,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003910698723719604,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20073511302471161,
      "num_tokens": 74178785.0,
      "step": 42760
    },
    {
      "entropy": 5.902700519561767,
      "epoch": 3.3272515074888154,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039104612840034904,
      "loss": 4.9881,
      "mean_token_accuracy": 0.2015640452504158,
      "num_tokens": 74187421.0,
      "step": 42765
    },
    {
      "entropy": 5.906836986541748,
      "epoch": 3.327640536860533,
      "grad_norm": 1.375,
      "learning_rate": 0.00039102238266792234,
      "loss": 4.929,
      "mean_token_accuracy": 0.21622657030820847,
      "num_tokens": 74195774.0,
      "step": 42770
    },
    {
      "entropy": 5.90985631942749,
      "epoch": 3.3280295662322503,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039099863517504064,
      "loss": 5.0196,
      "mean_token_accuracy": 0.20650187730789185,
      "num_tokens": 74204441.0,
      "step": 42775
    },
    {
      "entropy": 5.790881061553955,
      "epoch": 3.328418595603968,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003909748859220644,
      "loss": 4.9451,
      "mean_token_accuracy": 0.20908161550760268,
      "num_tokens": 74213042.0,
      "step": 42780
    },
    {
      "entropy": 5.799608945846558,
      "epoch": 3.3288076249756857,
      "grad_norm": 1.34375,
      "learning_rate": 0.00039095113490935377,
      "loss": 4.9164,
      "mean_token_accuracy": 0.21475863754749297,
      "num_tokens": 74221463.0,
      "step": 42785
    },
    {
      "entropy": 5.83936710357666,
      "epoch": 3.3291966543474034,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003909273821372692,
      "loss": 4.9468,
      "mean_token_accuracy": 0.2198818266391754,
      "num_tokens": 74230019.0,
      "step": 42790
    },
    {
      "entropy": 5.824632453918457,
      "epoch": 3.3295856837191207,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039090362760617114,
      "loss": 4.9757,
      "mean_token_accuracy": 0.20985606461763381,
      "num_tokens": 74238352.0,
      "step": 42795
    },
    {
      "entropy": 5.833367824554443,
      "epoch": 3.3299747130908384,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039087987131642,
      "loss": 5.0008,
      "mean_token_accuracy": 0.21044310480356215,
      "num_tokens": 74246699.0,
      "step": 42800
    },
    {
      "entropy": 5.833177089691162,
      "epoch": 3.330363742462556,
      "grad_norm": 1.25,
      "learning_rate": 0.0003908561132683762,
      "loss": 4.9739,
      "mean_token_accuracy": 0.21834137588739394,
      "num_tokens": 74255546.0,
      "step": 42805
    },
    {
      "entropy": 5.904789543151855,
      "epoch": 3.3307527718342733,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039083235346240044,
      "loss": 4.986,
      "mean_token_accuracy": 0.20526485592126847,
      "num_tokens": 74264777.0,
      "step": 42810
    },
    {
      "entropy": 5.8385556697845455,
      "epoch": 3.331141801205991,
      "grad_norm": 1.21875,
      "learning_rate": 0.000390808591898853,
      "loss": 4.9807,
      "mean_token_accuracy": 0.2095123291015625,
      "num_tokens": 74274521.0,
      "step": 42815
    },
    {
      "entropy": 5.778361415863037,
      "epoch": 3.3315308305777087,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003907848285780945,
      "loss": 4.9139,
      "mean_token_accuracy": 0.21654916554689407,
      "num_tokens": 74283521.0,
      "step": 42820
    },
    {
      "entropy": 5.8118720054626465,
      "epoch": 3.3319198599494264,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039076106350048557,
      "loss": 4.958,
      "mean_token_accuracy": 0.20990127176046372,
      "num_tokens": 74292474.0,
      "step": 42825
    },
    {
      "entropy": 5.82964072227478,
      "epoch": 3.3323088893211437,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003907372966663867,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21784617900848388,
      "num_tokens": 74300990.0,
      "step": 42830
    },
    {
      "entropy": 5.923128080368042,
      "epoch": 3.3326979186928614,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003907135280761585,
      "loss": 5.1353,
      "mean_token_accuracy": 0.19662207067012788,
      "num_tokens": 74310350.0,
      "step": 42835
    },
    {
      "entropy": 5.8685098648071286,
      "epoch": 3.333086948064579,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003906897577301618,
      "loss": 4.9921,
      "mean_token_accuracy": 0.2115735113620758,
      "num_tokens": 74319468.0,
      "step": 42840
    },
    {
      "entropy": 5.822714471817017,
      "epoch": 3.3334759774362963,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003906659856287571,
      "loss": 4.9594,
      "mean_token_accuracy": 0.21357482820749282,
      "num_tokens": 74327802.0,
      "step": 42845
    },
    {
      "entropy": 5.895671558380127,
      "epoch": 3.333865006808014,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039064221177230517,
      "loss": 5.0615,
      "mean_token_accuracy": 0.2010016843676567,
      "num_tokens": 74337632.0,
      "step": 42850
    },
    {
      "entropy": 5.845259714126587,
      "epoch": 3.3342540361797317,
      "grad_norm": 1.25,
      "learning_rate": 0.00039061843616116683,
      "loss": 5.028,
      "mean_token_accuracy": 0.20468274056911467,
      "num_tokens": 74346082.0,
      "step": 42855
    },
    {
      "entropy": 5.813741731643677,
      "epoch": 3.334643065551449,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039059465879570264,
      "loss": 4.9501,
      "mean_token_accuracy": 0.20808258801698684,
      "num_tokens": 74354238.0,
      "step": 42860
    },
    {
      "entropy": 5.8350170135498045,
      "epoch": 3.3350320949231667,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003905708796762736,
      "loss": 4.988,
      "mean_token_accuracy": 0.2023255154490471,
      "num_tokens": 74362576.0,
      "step": 42865
    },
    {
      "entropy": 5.8157310962677,
      "epoch": 3.3354211242948844,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003905470988032403,
      "loss": 4.9254,
      "mean_token_accuracy": 0.21614105254411697,
      "num_tokens": 74371188.0,
      "step": 42870
    },
    {
      "entropy": 5.876766395568848,
      "epoch": 3.3358101536666016,
      "grad_norm": 1.375,
      "learning_rate": 0.0003905233161769637,
      "loss": 4.9798,
      "mean_token_accuracy": 0.21647816598415376,
      "num_tokens": 74379832.0,
      "step": 42875
    },
    {
      "entropy": 5.812815952301025,
      "epoch": 3.3361991830383193,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003904995317978048,
      "loss": 4.9451,
      "mean_token_accuracy": 0.21059848070144654,
      "num_tokens": 74388691.0,
      "step": 42880
    },
    {
      "entropy": 5.783127737045288,
      "epoch": 3.336588212410037,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039047574566612416,
      "loss": 4.928,
      "mean_token_accuracy": 0.20944904386997223,
      "num_tokens": 74397776.0,
      "step": 42885
    },
    {
      "entropy": 5.839664602279663,
      "epoch": 3.3369772417817547,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039045195778228295,
      "loss": 5.0184,
      "mean_token_accuracy": 0.20882997065782546,
      "num_tokens": 74406616.0,
      "step": 42890
    },
    {
      "entropy": 5.843802070617675,
      "epoch": 3.337366271153472,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003904281681466421,
      "loss": 4.9999,
      "mean_token_accuracy": 0.209655998647213,
      "num_tokens": 74415029.0,
      "step": 42895
    },
    {
      "entropy": 5.884590768814087,
      "epoch": 3.3377553005251897,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003904043767595624,
      "loss": 4.9598,
      "mean_token_accuracy": 0.21364667862653733,
      "num_tokens": 74423582.0,
      "step": 42900
    },
    {
      "entropy": 5.919558763504028,
      "epoch": 3.3381443298969073,
      "grad_norm": 1.453125,
      "learning_rate": 0.00039038058362140507,
      "loss": 5.0479,
      "mean_token_accuracy": 0.20443604290485382,
      "num_tokens": 74432458.0,
      "step": 42905
    },
    {
      "entropy": 5.840308618545532,
      "epoch": 3.3385333592686246,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039035678873253096,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20856208503246307,
      "num_tokens": 74441414.0,
      "step": 42910
    },
    {
      "entropy": 5.9096990585327145,
      "epoch": 3.3389223886403423,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003903329920933012,
      "loss": 5.0461,
      "mean_token_accuracy": 0.2043795868754387,
      "num_tokens": 74450382.0,
      "step": 42915
    },
    {
      "entropy": 5.927686023712158,
      "epoch": 3.33931141801206,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003903091937040769,
      "loss": 5.059,
      "mean_token_accuracy": 0.21061530262231826,
      "num_tokens": 74459404.0,
      "step": 42920
    },
    {
      "entropy": 5.8168638229370115,
      "epoch": 3.3397004473837777,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003902853935652191,
      "loss": 4.8902,
      "mean_token_accuracy": 0.21427226513624192,
      "num_tokens": 74468389.0,
      "step": 42925
    },
    {
      "entropy": 5.7763043403625485,
      "epoch": 3.340089476755495,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003902615916770889,
      "loss": 4.9324,
      "mean_token_accuracy": 0.21388742029666902,
      "num_tokens": 74476349.0,
      "step": 42930
    },
    {
      "entropy": 5.838443040847778,
      "epoch": 3.3404785061272126,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039023778804004753,
      "loss": 4.9995,
      "mean_token_accuracy": 0.20598880499601363,
      "num_tokens": 74485294.0,
      "step": 42935
    },
    {
      "entropy": 5.86090579032898,
      "epoch": 3.3408675354989303,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039021398265445616,
      "loss": 5.0203,
      "mean_token_accuracy": 0.2073439359664917,
      "num_tokens": 74493978.0,
      "step": 42940
    },
    {
      "entropy": 5.919500827789307,
      "epoch": 3.3412565648706476,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000390190175520676,
      "loss": 5.0132,
      "mean_token_accuracy": 0.213208669424057,
      "num_tokens": 74503415.0,
      "step": 42945
    },
    {
      "entropy": 5.863336229324341,
      "epoch": 3.3416455942423653,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003901663666390683,
      "loss": 4.9932,
      "mean_token_accuracy": 0.21455423384904862,
      "num_tokens": 74512096.0,
      "step": 42950
    },
    {
      "entropy": 5.808670568466186,
      "epoch": 3.342034623614083,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003901425560099943,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21028883904218673,
      "num_tokens": 74520197.0,
      "step": 42955
    },
    {
      "entropy": 5.780158424377442,
      "epoch": 3.3424236529858002,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039011874363381523,
      "loss": 4.9633,
      "mean_token_accuracy": 0.20705867409706116,
      "num_tokens": 74529583.0,
      "step": 42960
    },
    {
      "entropy": 5.851327610015869,
      "epoch": 3.342812682357518,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039009492951089236,
      "loss": 4.9114,
      "mean_token_accuracy": 0.218940132856369,
      "num_tokens": 74538269.0,
      "step": 42965
    },
    {
      "entropy": 5.832162666320801,
      "epoch": 3.3432017117292356,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003900711136415873,
      "loss": 4.9216,
      "mean_token_accuracy": 0.21349160373210907,
      "num_tokens": 74547157.0,
      "step": 42970
    },
    {
      "entropy": 5.806814050674438,
      "epoch": 3.343590741100953,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003900472960262611,
      "loss": 5.0349,
      "mean_token_accuracy": 0.20200174450874328,
      "num_tokens": 74555374.0,
      "step": 42975
    },
    {
      "entropy": 5.88667311668396,
      "epoch": 3.3439797704726706,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039002347666527537,
      "loss": 5.0326,
      "mean_token_accuracy": 0.20361700057983398,
      "num_tokens": 74564401.0,
      "step": 42980
    },
    {
      "entropy": 5.832461166381836,
      "epoch": 3.3443687998443883,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038999965555899145,
      "loss": 4.8121,
      "mean_token_accuracy": 0.22651437371969224,
      "num_tokens": 74573080.0,
      "step": 42985
    },
    {
      "entropy": 5.879475688934326,
      "epoch": 3.344757829216106,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003899758327077709,
      "loss": 5.0213,
      "mean_token_accuracy": 0.20589971095323562,
      "num_tokens": 74582267.0,
      "step": 42990
    },
    {
      "entropy": 5.910618638992309,
      "epoch": 3.3451468585878232,
      "grad_norm": 1.375,
      "learning_rate": 0.00038995200811197496,
      "loss": 4.986,
      "mean_token_accuracy": 0.21082333773374556,
      "num_tokens": 74590637.0,
      "step": 42995
    },
    {
      "entropy": 5.795453882217407,
      "epoch": 3.345535887959541,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038992818177196526,
      "loss": 5.0,
      "mean_token_accuracy": 0.21156125217676164,
      "num_tokens": 74599132.0,
      "step": 43000
    },
    {
      "entropy": 5.811682319641113,
      "epoch": 3.3459249173312586,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003899043536881034,
      "loss": 4.918,
      "mean_token_accuracy": 0.21848187744617462,
      "num_tokens": 74607161.0,
      "step": 43005
    },
    {
      "entropy": 5.839553308486939,
      "epoch": 3.346313946702976,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038988052386075084,
      "loss": 5.0316,
      "mean_token_accuracy": 0.20822179019451142,
      "num_tokens": 74616020.0,
      "step": 43010
    },
    {
      "entropy": 5.893888664245606,
      "epoch": 3.3467029760746936,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038985669229026917,
      "loss": 5.0102,
      "mean_token_accuracy": 0.2064686894416809,
      "num_tokens": 74624410.0,
      "step": 43015
    },
    {
      "entropy": 5.910970211029053,
      "epoch": 3.3470920054464113,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038983285897702005,
      "loss": 4.97,
      "mean_token_accuracy": 0.20782451033592225,
      "num_tokens": 74632702.0,
      "step": 43020
    },
    {
      "entropy": 5.827113294601441,
      "epoch": 3.347481034818129,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038980902392136496,
      "loss": 4.9015,
      "mean_token_accuracy": 0.21118486374616624,
      "num_tokens": 74641310.0,
      "step": 43025
    },
    {
      "entropy": 5.777649450302124,
      "epoch": 3.3478700641898462,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003897851871236657,
      "loss": 4.9196,
      "mean_token_accuracy": 0.21113961488008498,
      "num_tokens": 74649931.0,
      "step": 43030
    },
    {
      "entropy": 5.825043773651123,
      "epoch": 3.348259093561564,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038976134858428397,
      "loss": 4.9619,
      "mean_token_accuracy": 0.20550073087215423,
      "num_tokens": 74658110.0,
      "step": 43035
    },
    {
      "entropy": 5.868655014038086,
      "epoch": 3.3486481229332816,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038973750830358144,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21235011667013168,
      "num_tokens": 74667177.0,
      "step": 43040
    },
    {
      "entropy": 5.806269693374634,
      "epoch": 3.349037152304999,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038971366628191986,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21716970354318618,
      "num_tokens": 74676066.0,
      "step": 43045
    },
    {
      "entropy": 5.818996858596802,
      "epoch": 3.3494261816767166,
      "grad_norm": 1.375,
      "learning_rate": 0.000389689822519661,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21810292154550553,
      "num_tokens": 74683879.0,
      "step": 43050
    },
    {
      "entropy": 5.846711206436157,
      "epoch": 3.3498152110484343,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038966597701716653,
      "loss": 4.9581,
      "mean_token_accuracy": 0.21114077270030976,
      "num_tokens": 74692594.0,
      "step": 43055
    },
    {
      "entropy": 5.890863561630249,
      "epoch": 3.350204240420152,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003896421297747985,
      "loss": 4.9865,
      "mean_token_accuracy": 0.2034825414419174,
      "num_tokens": 74701674.0,
      "step": 43060
    },
    {
      "entropy": 5.814179229736328,
      "epoch": 3.350593269791869,
      "grad_norm": 1.375,
      "learning_rate": 0.0003896182807929185,
      "loss": 4.9013,
      "mean_token_accuracy": 0.21337458193302156,
      "num_tokens": 74709864.0,
      "step": 43065
    },
    {
      "entropy": 5.862629318237305,
      "epoch": 3.350982299163587,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003895944300718885,
      "loss": 5.0481,
      "mean_token_accuracy": 0.20782252550125122,
      "num_tokens": 74718364.0,
      "step": 43070
    },
    {
      "entropy": 5.844099760055542,
      "epoch": 3.351371328535304,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038957057761207053,
      "loss": 4.9831,
      "mean_token_accuracy": 0.2061554566025734,
      "num_tokens": 74726823.0,
      "step": 43075
    },
    {
      "entropy": 5.815512466430664,
      "epoch": 3.351760357907022,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038954672341382635,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20451629608869554,
      "num_tokens": 74735758.0,
      "step": 43080
    },
    {
      "entropy": 5.857950878143311,
      "epoch": 3.3521493872787396,
      "grad_norm": 1.34375,
      "learning_rate": 0.000389522867477518,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21689166724681855,
      "num_tokens": 74744444.0,
      "step": 43085
    },
    {
      "entropy": 5.8396467685699465,
      "epoch": 3.3525384166504573,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003894990098035074,
      "loss": 5.0143,
      "mean_token_accuracy": 0.2069427028298378,
      "num_tokens": 74753420.0,
      "step": 43090
    },
    {
      "entropy": 5.816665363311768,
      "epoch": 3.3529274460221745,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003894751503921566,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21620472967624665,
      "num_tokens": 74762641.0,
      "step": 43095
    },
    {
      "entropy": 5.797010326385498,
      "epoch": 3.353316475393892,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003894512892438275,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21144979447126389,
      "num_tokens": 74771481.0,
      "step": 43100
    },
    {
      "entropy": 5.871252775192261,
      "epoch": 3.35370550476561,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038942742635888236,
      "loss": 4.9338,
      "mean_token_accuracy": 0.21203863471746445,
      "num_tokens": 74779589.0,
      "step": 43105
    },
    {
      "entropy": 5.756110000610351,
      "epoch": 3.354094534137327,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038940356173768303,
      "loss": 4.8421,
      "mean_token_accuracy": 0.2209274113178253,
      "num_tokens": 74788110.0,
      "step": 43110
    },
    {
      "entropy": 5.80942735671997,
      "epoch": 3.354483563509045,
      "grad_norm": 1.578125,
      "learning_rate": 0.0003893796953805919,
      "loss": 4.9669,
      "mean_token_accuracy": 0.21442490965127944,
      "num_tokens": 74797015.0,
      "step": 43115
    },
    {
      "entropy": 5.830813407897949,
      "epoch": 3.3548725928807626,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038935582728797085,
      "loss": 4.9791,
      "mean_token_accuracy": 0.20420684963464736,
      "num_tokens": 74807027.0,
      "step": 43120
    },
    {
      "entropy": 5.858003950119018,
      "epoch": 3.3552616222524803,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038933195746018213,
      "loss": 5.0044,
      "mean_token_accuracy": 0.2104016825556755,
      "num_tokens": 74815306.0,
      "step": 43125
    },
    {
      "entropy": 5.825858545303345,
      "epoch": 3.3556506516241975,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038930808589758797,
      "loss": 4.9989,
      "mean_token_accuracy": 0.20586484670639038,
      "num_tokens": 74823443.0,
      "step": 43130
    },
    {
      "entropy": 5.764940595626831,
      "epoch": 3.356039680995915,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003892842126005506,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22201027870178222,
      "num_tokens": 74831787.0,
      "step": 43135
    },
    {
      "entropy": 5.836208152770996,
      "epoch": 3.356428710367633,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003892603375694321,
      "loss": 4.9409,
      "mean_token_accuracy": 0.215311960875988,
      "num_tokens": 74840804.0,
      "step": 43140
    },
    {
      "entropy": 5.8840662956237795,
      "epoch": 3.35681773973935,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038923646080459484,
      "loss": 4.9941,
      "mean_token_accuracy": 0.20953678488731384,
      "num_tokens": 74849558.0,
      "step": 43145
    },
    {
      "entropy": 5.817625427246094,
      "epoch": 3.357206769111068,
      "grad_norm": 1.515625,
      "learning_rate": 0.00038921258230640123,
      "loss": 4.9066,
      "mean_token_accuracy": 0.21595184803009032,
      "num_tokens": 74857923.0,
      "step": 43150
    },
    {
      "entropy": 5.763276386260986,
      "epoch": 3.3575957984827856,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038918870207521334,
      "loss": 4.9122,
      "mean_token_accuracy": 0.20636270344257354,
      "num_tokens": 74866227.0,
      "step": 43155
    },
    {
      "entropy": 5.903433465957642,
      "epoch": 3.3579848278545033,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038916482011139373,
      "loss": 5.0184,
      "mean_token_accuracy": 0.2082407534122467,
      "num_tokens": 74874420.0,
      "step": 43160
    },
    {
      "entropy": 5.8736159324646,
      "epoch": 3.3583738572262205,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003891409364153047,
      "loss": 5.0176,
      "mean_token_accuracy": 0.20799576938152314,
      "num_tokens": 74883888.0,
      "step": 43165
    },
    {
      "entropy": 5.841304349899292,
      "epoch": 3.358762886597938,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003891170509873086,
      "loss": 4.961,
      "mean_token_accuracy": 0.21401079446077348,
      "num_tokens": 74893873.0,
      "step": 43170
    },
    {
      "entropy": 5.880174446105957,
      "epoch": 3.359151915969656,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038909316382776784,
      "loss": 5.0335,
      "mean_token_accuracy": 0.2036713793873787,
      "num_tokens": 74903460.0,
      "step": 43175
    },
    {
      "entropy": 5.893458414077759,
      "epoch": 3.359540945341373,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003890692749370449,
      "loss": 5.073,
      "mean_token_accuracy": 0.20022270381450652,
      "num_tokens": 74912960.0,
      "step": 43180
    },
    {
      "entropy": 5.855274248123169,
      "epoch": 3.359929974713091,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038904538431550235,
      "loss": 4.9705,
      "mean_token_accuracy": 0.21769245862960815,
      "num_tokens": 74921983.0,
      "step": 43185
    },
    {
      "entropy": 5.912438249588012,
      "epoch": 3.3603190040848085,
      "grad_norm": 1.25,
      "learning_rate": 0.0003890214919635026,
      "loss": 5.0675,
      "mean_token_accuracy": 0.20508663207292557,
      "num_tokens": 74931883.0,
      "step": 43190
    },
    {
      "entropy": 5.804451990127563,
      "epoch": 3.360708033456526,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003889975978814082,
      "loss": 4.9189,
      "mean_token_accuracy": 0.21209487617015838,
      "num_tokens": 74940852.0,
      "step": 43195
    },
    {
      "entropy": 5.819003009796143,
      "epoch": 3.3610970628282435,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003889737020695817,
      "loss": 4.9151,
      "mean_token_accuracy": 0.20883516669273378,
      "num_tokens": 74949562.0,
      "step": 43200
    },
    {
      "entropy": 5.871111488342285,
      "epoch": 3.361486092199961,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038894980452838573,
      "loss": 5.0377,
      "mean_token_accuracy": 0.21060646176338196,
      "num_tokens": 74958199.0,
      "step": 43205
    },
    {
      "entropy": 5.874205541610718,
      "epoch": 3.3618751215716784,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003889259052581828,
      "loss": 4.9609,
      "mean_token_accuracy": 0.19816185235977174,
      "num_tokens": 74966362.0,
      "step": 43210
    },
    {
      "entropy": 5.796246480941773,
      "epoch": 3.362264150943396,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038890200425933545,
      "loss": 4.9491,
      "mean_token_accuracy": 0.21735860854387284,
      "num_tokens": 74974767.0,
      "step": 43215
    },
    {
      "entropy": 5.871094751358032,
      "epoch": 3.362653180315114,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003888781015322067,
      "loss": 5.0361,
      "mean_token_accuracy": 0.20685530602931976,
      "num_tokens": 74983604.0,
      "step": 43220
    },
    {
      "entropy": 5.9332959175109865,
      "epoch": 3.3630422096868315,
      "grad_norm": 1.375,
      "learning_rate": 0.00038885419707715885,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20884458273649215,
      "num_tokens": 74991866.0,
      "step": 43225
    },
    {
      "entropy": 5.86727819442749,
      "epoch": 3.363431239058549,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038883029089455487,
      "loss": 4.9716,
      "mean_token_accuracy": 0.215176260471344,
      "num_tokens": 75001004.0,
      "step": 43230
    },
    {
      "entropy": 5.764481401443481,
      "epoch": 3.3638202684302665,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003888063829847574,
      "loss": 4.8599,
      "mean_token_accuracy": 0.2211586967110634,
      "num_tokens": 75009766.0,
      "step": 43235
    },
    {
      "entropy": 5.926794242858887,
      "epoch": 3.364209297801984,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003887824733481292,
      "loss": 5.0279,
      "mean_token_accuracy": 0.20281373858451843,
      "num_tokens": 75018167.0,
      "step": 43240
    },
    {
      "entropy": 5.89974570274353,
      "epoch": 3.3645983271737014,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038875856198503306,
      "loss": 4.9606,
      "mean_token_accuracy": 0.21420126855373384,
      "num_tokens": 75027084.0,
      "step": 43245
    },
    {
      "entropy": 5.799437952041626,
      "epoch": 3.364987356545419,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038873464889583177,
      "loss": 4.9342,
      "mean_token_accuracy": 0.21189447790384291,
      "num_tokens": 75035069.0,
      "step": 43250
    },
    {
      "entropy": 5.766240167617798,
      "epoch": 3.365376385917137,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038871073408088826,
      "loss": 4.9029,
      "mean_token_accuracy": 0.21025425493717192,
      "num_tokens": 75043961.0,
      "step": 43255
    },
    {
      "entropy": 5.805471992492675,
      "epoch": 3.3657654152888545,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003886868175405653,
      "loss": 4.9308,
      "mean_token_accuracy": 0.210835362970829,
      "num_tokens": 75052942.0,
      "step": 43260
    },
    {
      "entropy": 5.874312448501587,
      "epoch": 3.366154444660572,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003886628992752259,
      "loss": 4.9833,
      "mean_token_accuracy": 0.20985704213380812,
      "num_tokens": 75061058.0,
      "step": 43265
    },
    {
      "entropy": 5.875028610229492,
      "epoch": 3.3665434740322895,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003886389792852329,
      "loss": 4.9562,
      "mean_token_accuracy": 0.20855135023593901,
      "num_tokens": 75069757.0,
      "step": 43270
    },
    {
      "entropy": 5.817039442062378,
      "epoch": 3.366932503404007,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038861505757094923,
      "loss": 4.9055,
      "mean_token_accuracy": 0.2089798480272293,
      "num_tokens": 75078488.0,
      "step": 43275
    },
    {
      "entropy": 5.920904159545898,
      "epoch": 3.3673215327757244,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003885911341327379,
      "loss": 5.0914,
      "mean_token_accuracy": 0.19929058253765106,
      "num_tokens": 75087220.0,
      "step": 43280
    },
    {
      "entropy": 5.959580278396606,
      "epoch": 3.367710562147442,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000388567208970962,
      "loss": 5.0755,
      "mean_token_accuracy": 0.20068210661411284,
      "num_tokens": 75095412.0,
      "step": 43285
    },
    {
      "entropy": 5.903345203399658,
      "epoch": 3.36809959151916,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003885432820859844,
      "loss": 5.0743,
      "mean_token_accuracy": 0.20170137286186218,
      "num_tokens": 75104783.0,
      "step": 43290
    },
    {
      "entropy": 5.858115386962891,
      "epoch": 3.368488620890877,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038851935347816827,
      "loss": 4.9679,
      "mean_token_accuracy": 0.2076390027999878,
      "num_tokens": 75113801.0,
      "step": 43295
    },
    {
      "entropy": 5.935214185714722,
      "epoch": 3.368877650262595,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003884954231478764,
      "loss": 5.0379,
      "mean_token_accuracy": 0.20119248777627946,
      "num_tokens": 75123116.0,
      "step": 43300
    },
    {
      "entropy": 5.8980601787567135,
      "epoch": 3.3692666796343125,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003884714910954724,
      "loss": 4.9691,
      "mean_token_accuracy": 0.20547809451818466,
      "num_tokens": 75131662.0,
      "step": 43305
    },
    {
      "entropy": 5.763066244125366,
      "epoch": 3.3696557090060297,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000388447557321319,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2145422324538231,
      "num_tokens": 75140897.0,
      "step": 43310
    },
    {
      "entropy": 5.79844069480896,
      "epoch": 3.3700447383777474,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003884236218257795,
      "loss": 4.975,
      "mean_token_accuracy": 0.20715581327676774,
      "num_tokens": 75150189.0,
      "step": 43315
    },
    {
      "entropy": 5.803768062591553,
      "epoch": 3.370433767749465,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003883996846092171,
      "loss": 4.8551,
      "mean_token_accuracy": 0.2269786313176155,
      "num_tokens": 75158677.0,
      "step": 43320
    },
    {
      "entropy": 5.8337572574615475,
      "epoch": 3.370822797121183,
      "grad_norm": 1.25,
      "learning_rate": 0.000388375745671995,
      "loss": 4.9536,
      "mean_token_accuracy": 0.21378158777952194,
      "num_tokens": 75167669.0,
      "step": 43325
    },
    {
      "entropy": 5.8386555194854735,
      "epoch": 3.3712118264929,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038835180501447646,
      "loss": 4.9744,
      "mean_token_accuracy": 0.20647733956575393,
      "num_tokens": 75176379.0,
      "step": 43330
    },
    {
      "entropy": 5.848819971084595,
      "epoch": 3.3716008558646178,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003883278626370245,
      "loss": 5.0038,
      "mean_token_accuracy": 0.21000365614891053,
      "num_tokens": 75185634.0,
      "step": 43335
    },
    {
      "entropy": 5.855559587478638,
      "epoch": 3.3719898852363355,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038830391854000285,
      "loss": 4.9353,
      "mean_token_accuracy": 0.2142452836036682,
      "num_tokens": 75194101.0,
      "step": 43340
    },
    {
      "entropy": 5.896668577194214,
      "epoch": 3.3723789146080527,
      "grad_norm": 1.375,
      "learning_rate": 0.00038827997272377447,
      "loss": 4.9958,
      "mean_token_accuracy": 0.20538676977157594,
      "num_tokens": 75202738.0,
      "step": 43345
    },
    {
      "entropy": 5.835319900512696,
      "epoch": 3.3727679439797704,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038825602518870276,
      "loss": 4.844,
      "mean_token_accuracy": 0.21763889938592912,
      "num_tokens": 75211771.0,
      "step": 43350
    },
    {
      "entropy": 5.882034206390381,
      "epoch": 3.373156973351488,
      "grad_norm": 1.375,
      "learning_rate": 0.0003882320759351512,
      "loss": 5.0023,
      "mean_token_accuracy": 0.20337293595075606,
      "num_tokens": 75221270.0,
      "step": 43355
    },
    {
      "entropy": 5.901565217971802,
      "epoch": 3.373546002723206,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038820812496348305,
      "loss": 4.9566,
      "mean_token_accuracy": 0.211650288105011,
      "num_tokens": 75229864.0,
      "step": 43360
    },
    {
      "entropy": 5.85054202079773,
      "epoch": 3.373935032094923,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038818417227406185,
      "loss": 5.0378,
      "mean_token_accuracy": 0.20025272518396378,
      "num_tokens": 75239446.0,
      "step": 43365
    },
    {
      "entropy": 5.816908597946167,
      "epoch": 3.3743240614666408,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038816021786725096,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20730853378772734,
      "num_tokens": 75247997.0,
      "step": 43370
    },
    {
      "entropy": 5.810194301605224,
      "epoch": 3.3747130908383585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038813626174341386,
      "loss": 4.9124,
      "mean_token_accuracy": 0.21124526411294936,
      "num_tokens": 75256492.0,
      "step": 43375
    },
    {
      "entropy": 5.895193004608155,
      "epoch": 3.3751021202100757,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003881123039029141,
      "loss": 5.0018,
      "mean_token_accuracy": 0.20909362584352492,
      "num_tokens": 75265057.0,
      "step": 43380
    },
    {
      "entropy": 5.810988187789917,
      "epoch": 3.3754911495817934,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003880883443461152,
      "loss": 4.9708,
      "mean_token_accuracy": 0.21300434470176696,
      "num_tokens": 75274685.0,
      "step": 43385
    },
    {
      "entropy": 5.8117406368255615,
      "epoch": 3.375880178953511,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003880643830733806,
      "loss": 4.9595,
      "mean_token_accuracy": 0.210467866063118,
      "num_tokens": 75283053.0,
      "step": 43390
    },
    {
      "entropy": 5.828567171096802,
      "epoch": 3.3762692083252284,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038804042008507396,
      "loss": 5.0016,
      "mean_token_accuracy": 0.21121222525835037,
      "num_tokens": 75291823.0,
      "step": 43395
    },
    {
      "entropy": 5.876355361938477,
      "epoch": 3.376658237696946,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038801645538155883,
      "loss": 5.0575,
      "mean_token_accuracy": 0.1978149428963661,
      "num_tokens": 75299858.0,
      "step": 43400
    },
    {
      "entropy": 5.927226829528808,
      "epoch": 3.3770472670686638,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038799248896319896,
      "loss": 5.063,
      "mean_token_accuracy": 0.20472633987665176,
      "num_tokens": 75308157.0,
      "step": 43405
    },
    {
      "entropy": 5.909140729904175,
      "epoch": 3.377436296440381,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038796852083035794,
      "loss": 5.0365,
      "mean_token_accuracy": 0.20736172795295715,
      "num_tokens": 75316589.0,
      "step": 43410
    },
    {
      "entropy": 5.8654601097106935,
      "epoch": 3.3778253258120987,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003879445509833993,
      "loss": 4.9277,
      "mean_token_accuracy": 0.20812382847070693,
      "num_tokens": 75326219.0,
      "step": 43415
    },
    {
      "entropy": 5.900867748260498,
      "epoch": 3.3782143551838164,
      "grad_norm": 1.296875,
      "learning_rate": 0.000387920579422687,
      "loss": 5.0468,
      "mean_token_accuracy": 0.20216209143400193,
      "num_tokens": 75334348.0,
      "step": 43420
    },
    {
      "entropy": 5.921757555007934,
      "epoch": 3.378603384555534,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003878966061485845,
      "loss": 5.0935,
      "mean_token_accuracy": 0.19990944117307663,
      "num_tokens": 75343242.0,
      "step": 43425
    },
    {
      "entropy": 5.921330881118775,
      "epoch": 3.3789924139272514,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038787263116145577,
      "loss": 5.062,
      "mean_token_accuracy": 0.20078145563602448,
      "num_tokens": 75351760.0,
      "step": 43430
    },
    {
      "entropy": 5.83110933303833,
      "epoch": 3.379381443298969,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038784865446166454,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2138667047023773,
      "num_tokens": 75360605.0,
      "step": 43435
    },
    {
      "entropy": 5.8273553371429445,
      "epoch": 3.3797704726706868,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038782467604957455,
      "loss": 4.8848,
      "mean_token_accuracy": 0.21912334561347963,
      "num_tokens": 75368801.0,
      "step": 43440
    },
    {
      "entropy": 5.7931946277618405,
      "epoch": 3.380159502042404,
      "grad_norm": 1.46875,
      "learning_rate": 0.00038780069592554977,
      "loss": 4.9603,
      "mean_token_accuracy": 0.20600615292787552,
      "num_tokens": 75377466.0,
      "step": 43445
    },
    {
      "entropy": 5.869297027587891,
      "epoch": 3.3805485314141217,
      "grad_norm": 1.25,
      "learning_rate": 0.0003877767140899539,
      "loss": 5.0287,
      "mean_token_accuracy": 0.20937345921993256,
      "num_tokens": 75386831.0,
      "step": 43450
    },
    {
      "entropy": 5.948837471008301,
      "epoch": 3.3809375607858394,
      "grad_norm": 1.375,
      "learning_rate": 0.00038775273054315087,
      "loss": 4.9764,
      "mean_token_accuracy": 0.20542542338371278,
      "num_tokens": 75394949.0,
      "step": 43455
    },
    {
      "entropy": 5.837774038314819,
      "epoch": 3.381326590157557,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003877287452855046,
      "loss": 4.8724,
      "mean_token_accuracy": 0.22140044271945952,
      "num_tokens": 75403678.0,
      "step": 43460
    },
    {
      "entropy": 5.809627914428711,
      "epoch": 3.3817156195292744,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038770475831737905,
      "loss": 4.9158,
      "mean_token_accuracy": 0.2186850443482399,
      "num_tokens": 75412195.0,
      "step": 43465
    },
    {
      "entropy": 5.804500532150269,
      "epoch": 3.382104648900992,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038768076963913823,
      "loss": 4.8925,
      "mean_token_accuracy": 0.21176016926765442,
      "num_tokens": 75421383.0,
      "step": 43470
    },
    {
      "entropy": 5.80602536201477,
      "epoch": 3.3824936782727097,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038765677925114614,
      "loss": 4.8965,
      "mean_token_accuracy": 0.21284267157316208,
      "num_tokens": 75430248.0,
      "step": 43475
    },
    {
      "entropy": 5.76225733757019,
      "epoch": 3.382882707644427,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003876327871537667,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21789791733026503,
      "num_tokens": 75438849.0,
      "step": 43480
    },
    {
      "entropy": 5.868874168395996,
      "epoch": 3.3832717370161447,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038760879334736394,
      "loss": 5.0149,
      "mean_token_accuracy": 0.21043260246515275,
      "num_tokens": 75447562.0,
      "step": 43485
    },
    {
      "entropy": 5.911243915557861,
      "epoch": 3.3836607663878624,
      "grad_norm": 1.453125,
      "learning_rate": 0.00038758479783230197,
      "loss": 5.0046,
      "mean_token_accuracy": 0.20378794521093369,
      "num_tokens": 75456364.0,
      "step": 43490
    },
    {
      "entropy": 5.8536820888519285,
      "epoch": 3.38404979575958,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003875608006089449,
      "loss": 4.9462,
      "mean_token_accuracy": 0.2092744752764702,
      "num_tokens": 75465612.0,
      "step": 43495
    },
    {
      "entropy": 5.845042514801025,
      "epoch": 3.3844388251312973,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003875368016776569,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21377458423376083,
      "num_tokens": 75474604.0,
      "step": 43500
    },
    {
      "entropy": 5.839162445068359,
      "epoch": 3.384827854503015,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000387512801038802,
      "loss": 4.9051,
      "mean_token_accuracy": 0.21862321794033052,
      "num_tokens": 75483193.0,
      "step": 43505
    },
    {
      "entropy": 5.810390472412109,
      "epoch": 3.3852168838747323,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003874887986927444,
      "loss": 4.9152,
      "mean_token_accuracy": 0.21272409707307816,
      "num_tokens": 75491813.0,
      "step": 43510
    },
    {
      "entropy": 5.923397445678711,
      "epoch": 3.38560591324645,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003874647946398484,
      "loss": 5.0523,
      "mean_token_accuracy": 0.20736235082149507,
      "num_tokens": 75500520.0,
      "step": 43515
    },
    {
      "entropy": 5.863705635070801,
      "epoch": 3.3859949426181677,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003874407888804781,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2147574320435524,
      "num_tokens": 75509249.0,
      "step": 43520
    },
    {
      "entropy": 5.8056402683258055,
      "epoch": 3.3863839719898854,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038741678141499783,
      "loss": 4.9036,
      "mean_token_accuracy": 0.2207706719636917,
      "num_tokens": 75517654.0,
      "step": 43525
    },
    {
      "entropy": 5.9058952808380125,
      "epoch": 3.3867730013616026,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003873927722437718,
      "loss": 5.0419,
      "mean_token_accuracy": 0.20983896702528,
      "num_tokens": 75526995.0,
      "step": 43530
    },
    {
      "entropy": 5.991169404983521,
      "epoch": 3.3871620307333203,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003873687613671643,
      "loss": 5.0747,
      "mean_token_accuracy": 0.20524675995111466,
      "num_tokens": 75536399.0,
      "step": 43535
    },
    {
      "entropy": 5.760081148147583,
      "epoch": 3.387551060105038,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003873447487855398,
      "loss": 4.8234,
      "mean_token_accuracy": 0.22108521908521653,
      "num_tokens": 75544436.0,
      "step": 43540
    },
    {
      "entropy": 5.863285827636719,
      "epoch": 3.3879400894767553,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038732073449926244,
      "loss": 5.0564,
      "mean_token_accuracy": 0.21082198321819307,
      "num_tokens": 75552757.0,
      "step": 43545
    },
    {
      "entropy": 5.914246225357056,
      "epoch": 3.388329118848473,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003872967185086968,
      "loss": 5.062,
      "mean_token_accuracy": 0.207522015273571,
      "num_tokens": 75560850.0,
      "step": 43550
    },
    {
      "entropy": 5.915704298019409,
      "epoch": 3.3887181482201907,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003872727008142072,
      "loss": 4.955,
      "mean_token_accuracy": 0.2098520129919052,
      "num_tokens": 75569054.0,
      "step": 43555
    },
    {
      "entropy": 5.866705513000488,
      "epoch": 3.3891071775919084,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038724868141615807,
      "loss": 5.0568,
      "mean_token_accuracy": 0.20347207635641099,
      "num_tokens": 75578008.0,
      "step": 43560
    },
    {
      "entropy": 5.869704389572144,
      "epoch": 3.3894962069636256,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003872246603149138,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21469050496816636,
      "num_tokens": 75586332.0,
      "step": 43565
    },
    {
      "entropy": 5.829465579986572,
      "epoch": 3.3898852363353433,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003872006375108389,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20756751000881196,
      "num_tokens": 75594217.0,
      "step": 43570
    },
    {
      "entropy": 5.783953905105591,
      "epoch": 3.390274265707061,
      "grad_norm": 1.40625,
      "learning_rate": 0.000387176613004298,
      "loss": 4.8175,
      "mean_token_accuracy": 0.2321384385228157,
      "num_tokens": 75602287.0,
      "step": 43575
    },
    {
      "entropy": 5.775793027877808,
      "epoch": 3.3906632950787783,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003871525867956555,
      "loss": 4.8751,
      "mean_token_accuracy": 0.21968195140361785,
      "num_tokens": 75610765.0,
      "step": 43580
    },
    {
      "entropy": 5.864161729812622,
      "epoch": 3.391052324450496,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038712855888527604,
      "loss": 5.0517,
      "mean_token_accuracy": 0.2058502823114395,
      "num_tokens": 75620103.0,
      "step": 43585
    },
    {
      "entropy": 5.892338562011719,
      "epoch": 3.3914413538222137,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038710452927352414,
      "loss": 4.9866,
      "mean_token_accuracy": 0.2094047486782074,
      "num_tokens": 75629191.0,
      "step": 43590
    },
    {
      "entropy": 5.920384263992309,
      "epoch": 3.3918303831939314,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003870804979607645,
      "loss": 4.9333,
      "mean_token_accuracy": 0.2104494258761406,
      "num_tokens": 75637575.0,
      "step": 43595
    },
    {
      "entropy": 5.866631937026978,
      "epoch": 3.3922194125656486,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038705646494736163,
      "loss": 5.0833,
      "mean_token_accuracy": 0.20507924854755402,
      "num_tokens": 75645925.0,
      "step": 43600
    },
    {
      "entropy": 5.782024717330932,
      "epoch": 3.3926084419373663,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003870324302336802,
      "loss": 4.8682,
      "mean_token_accuracy": 0.21200764626264573,
      "num_tokens": 75654269.0,
      "step": 43605
    },
    {
      "entropy": 5.891292524337769,
      "epoch": 3.392997471309084,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003870083938200851,
      "loss": 5.0051,
      "mean_token_accuracy": 0.2076784312725067,
      "num_tokens": 75663019.0,
      "step": 43610
    },
    {
      "entropy": 5.897125005722046,
      "epoch": 3.3933865006808013,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038698435570694084,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20762305557727814,
      "num_tokens": 75671246.0,
      "step": 43615
    },
    {
      "entropy": 5.835525703430176,
      "epoch": 3.393775530052519,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003869603158946122,
      "loss": 4.8711,
      "mean_token_accuracy": 0.2226286932826042,
      "num_tokens": 75679793.0,
      "step": 43620
    },
    {
      "entropy": 5.863463973999023,
      "epoch": 3.3941645594242367,
      "grad_norm": 1.359375,
      "learning_rate": 0.000386936274383464,
      "loss": 4.9725,
      "mean_token_accuracy": 0.205870421230793,
      "num_tokens": 75688356.0,
      "step": 43625
    },
    {
      "entropy": 5.8460582256317135,
      "epoch": 3.394553588795954,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000386912231173861,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21406291723251342,
      "num_tokens": 75697252.0,
      "step": 43630
    },
    {
      "entropy": 5.8522539138793945,
      "epoch": 3.3949426181676716,
      "grad_norm": 1.3125,
      "learning_rate": 0.000386888186266168,
      "loss": 4.9782,
      "mean_token_accuracy": 0.21982807219028472,
      "num_tokens": 75705947.0,
      "step": 43635
    },
    {
      "entropy": 5.879702949523926,
      "epoch": 3.3953316475393893,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003868641396607499,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20264589190483093,
      "num_tokens": 75714381.0,
      "step": 43640
    },
    {
      "entropy": 5.809407901763916,
      "epoch": 3.3957206769111066,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003868400913579715,
      "loss": 4.9523,
      "mean_token_accuracy": 0.20542743802070618,
      "num_tokens": 75723170.0,
      "step": 43645
    },
    {
      "entropy": 5.809116172790527,
      "epoch": 3.3961097062828243,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00038681604135819774,
      "loss": 4.9627,
      "mean_token_accuracy": 0.20993778258562087,
      "num_tokens": 75732304.0,
      "step": 43650
    },
    {
      "entropy": 5.904730796813965,
      "epoch": 3.396498735654542,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038679198966179353,
      "loss": 5.0289,
      "mean_token_accuracy": 0.20943732410669327,
      "num_tokens": 75740967.0,
      "step": 43655
    },
    {
      "entropy": 5.820071744918823,
      "epoch": 3.3968877650262597,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038676793626912375,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21390833109617233,
      "num_tokens": 75749352.0,
      "step": 43660
    },
    {
      "entropy": 5.855237674713135,
      "epoch": 3.397276794397977,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038674388118055354,
      "loss": 4.9574,
      "mean_token_accuracy": 0.21729274988174438,
      "num_tokens": 75758435.0,
      "step": 43665
    },
    {
      "entropy": 5.836182498931885,
      "epoch": 3.3976658237696946,
      "grad_norm": 1.25,
      "learning_rate": 0.0003867198243964477,
      "loss": 4.9589,
      "mean_token_accuracy": 0.2146657720208168,
      "num_tokens": 75767559.0,
      "step": 43670
    },
    {
      "entropy": 5.876359748840332,
      "epoch": 3.3980548531414123,
      "grad_norm": 1.375,
      "learning_rate": 0.0003866957659171714,
      "loss": 4.9858,
      "mean_token_accuracy": 0.20794467478990555,
      "num_tokens": 75776344.0,
      "step": 43675
    },
    {
      "entropy": 5.944918823242188,
      "epoch": 3.3984438825131296,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003866717057430896,
      "loss": 5.0568,
      "mean_token_accuracy": 0.2109245017170906,
      "num_tokens": 75785488.0,
      "step": 43680
    },
    {
      "entropy": 5.861222410202027,
      "epoch": 3.3988329118848473,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003866476438745675,
      "loss": 4.9458,
      "mean_token_accuracy": 0.2139919251203537,
      "num_tokens": 75793501.0,
      "step": 43685
    },
    {
      "entropy": 5.856098937988281,
      "epoch": 3.399221941256565,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038662358031197,
      "loss": 4.9324,
      "mean_token_accuracy": 0.21445069313049317,
      "num_tokens": 75801365.0,
      "step": 43690
    },
    {
      "entropy": 5.804084348678589,
      "epoch": 3.3996109706282827,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003865995150556624,
      "loss": 4.8584,
      "mean_token_accuracy": 0.2285398542881012,
      "num_tokens": 75810061.0,
      "step": 43695
    },
    {
      "entropy": 5.9310651302337645,
      "epoch": 3.4,
      "grad_norm": 1.453125,
      "learning_rate": 0.00038657544810600966,
      "loss": 4.9682,
      "mean_token_accuracy": 0.21041859537363053,
      "num_tokens": 75817574.0,
      "step": 43700
    },
    {
      "entropy": 5.899451208114624,
      "epoch": 3.4003890293717176,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003865513794633773,
      "loss": 5.0983,
      "mean_token_accuracy": 0.20494943410158156,
      "num_tokens": 75826198.0,
      "step": 43705
    },
    {
      "entropy": 5.8465088367462155,
      "epoch": 3.4007780587434353,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003865273091281301,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21291194707155228,
      "num_tokens": 75835043.0,
      "step": 43710
    },
    {
      "entropy": 5.902456331253052,
      "epoch": 3.4011670881151526,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003865032371006336,
      "loss": 5.0552,
      "mean_token_accuracy": 0.2049863561987877,
      "num_tokens": 75843191.0,
      "step": 43715
    },
    {
      "entropy": 5.861647129058838,
      "epoch": 3.4015561174868703,
      "grad_norm": 1.34375,
      "learning_rate": 0.000386479163381253,
      "loss": 4.9346,
      "mean_token_accuracy": 0.2209335222840309,
      "num_tokens": 75852024.0,
      "step": 43720
    },
    {
      "entropy": 5.7848883152008055,
      "epoch": 3.401945146858588,
      "grad_norm": 1.390625,
      "learning_rate": 0.00038645508797035345,
      "loss": 4.8375,
      "mean_token_accuracy": 0.22590504288673402,
      "num_tokens": 75860605.0,
      "step": 43725
    },
    {
      "entropy": 5.814254236221314,
      "epoch": 3.402334176230305,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038643101086830036,
      "loss": 4.9871,
      "mean_token_accuracy": 0.21688619405031204,
      "num_tokens": 75868861.0,
      "step": 43730
    },
    {
      "entropy": 5.800155735015869,
      "epoch": 3.402723205602023,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000386406932075459,
      "loss": 4.946,
      "mean_token_accuracy": 0.21470569372177123,
      "num_tokens": 75876840.0,
      "step": 43735
    },
    {
      "entropy": 5.779456233978271,
      "epoch": 3.4031122349737406,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003863828515921948,
      "loss": 4.97,
      "mean_token_accuracy": 0.21442488580942154,
      "num_tokens": 75885348.0,
      "step": 43740
    },
    {
      "entropy": 5.816876316070557,
      "epoch": 3.403501264345458,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003863587694188732,
      "loss": 5.0002,
      "mean_token_accuracy": 0.21057252436876298,
      "num_tokens": 75894062.0,
      "step": 43745
    },
    {
      "entropy": 5.899986314773559,
      "epoch": 3.4038902937171756,
      "grad_norm": 1.390625,
      "learning_rate": 0.00038633468555585943,
      "loss": 5.0266,
      "mean_token_accuracy": 0.21033218652009963,
      "num_tokens": 75902103.0,
      "step": 43750
    },
    {
      "entropy": 5.907148170471191,
      "epoch": 3.4042793230888932,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003863106000035191,
      "loss": 4.9876,
      "mean_token_accuracy": 0.20815979987382888,
      "num_tokens": 75910702.0,
      "step": 43755
    },
    {
      "entropy": 5.867507839202881,
      "epoch": 3.404668352460611,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003862865127622175,
      "loss": 4.9561,
      "mean_token_accuracy": 0.22112989127635957,
      "num_tokens": 75919085.0,
      "step": 43760
    },
    {
      "entropy": 5.88025918006897,
      "epoch": 3.405057381832328,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003862624238323201,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21158987581729888,
      "num_tokens": 75927421.0,
      "step": 43765
    },
    {
      "entropy": 5.7898355484008786,
      "epoch": 3.405446411204046,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003862383332141927,
      "loss": 4.9978,
      "mean_token_accuracy": 0.21203978806734086,
      "num_tokens": 75935719.0,
      "step": 43770
    },
    {
      "entropy": 5.9160990715026855,
      "epoch": 3.4058354405757636,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003862142409082006,
      "loss": 4.9555,
      "mean_token_accuracy": 0.20951660722494125,
      "num_tokens": 75944159.0,
      "step": 43775
    },
    {
      "entropy": 5.8175774097442625,
      "epoch": 3.406224469947481,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003861901469147093,
      "loss": 4.8185,
      "mean_token_accuracy": 0.22251327335834503,
      "num_tokens": 75951551.0,
      "step": 43780
    },
    {
      "entropy": 5.763814496994018,
      "epoch": 3.4066134993191985,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038616605123408464,
      "loss": 4.9362,
      "mean_token_accuracy": 0.208657206594944,
      "num_tokens": 75960422.0,
      "step": 43785
    },
    {
      "entropy": 5.808055257797241,
      "epoch": 3.4070025286909162,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038614195386669207,
      "loss": 4.9795,
      "mean_token_accuracy": 0.20771874934434892,
      "num_tokens": 75969318.0,
      "step": 43790
    },
    {
      "entropy": 5.824270296096802,
      "epoch": 3.407391558062634,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003861178548128972,
      "loss": 4.9797,
      "mean_token_accuracy": 0.2069397136569023,
      "num_tokens": 75978367.0,
      "step": 43795
    },
    {
      "entropy": 5.971016263961792,
      "epoch": 3.407780587434351,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038609375407306566,
      "loss": 5.0583,
      "mean_token_accuracy": 0.2057945042848587,
      "num_tokens": 75987211.0,
      "step": 43800
    },
    {
      "entropy": 5.894583606719971,
      "epoch": 3.408169616806069,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038606965164756334,
      "loss": 5.0389,
      "mean_token_accuracy": 0.20768342167139053,
      "num_tokens": 75996892.0,
      "step": 43805
    },
    {
      "entropy": 5.832450723648071,
      "epoch": 3.4085586461777866,
      "grad_norm": 1.375,
      "learning_rate": 0.00038604554753675585,
      "loss": 4.9801,
      "mean_token_accuracy": 0.21070825904607773,
      "num_tokens": 76005160.0,
      "step": 43810
    },
    {
      "entropy": 5.869582462310791,
      "epoch": 3.408947675549504,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003860214417410089,
      "loss": 4.9942,
      "mean_token_accuracy": 0.21537842303514482,
      "num_tokens": 76013623.0,
      "step": 43815
    },
    {
      "entropy": 5.902950668334961,
      "epoch": 3.4093367049212215,
      "grad_norm": 1.25,
      "learning_rate": 0.00038599733426068826,
      "loss": 5.0183,
      "mean_token_accuracy": 0.20660870671272277,
      "num_tokens": 76022735.0,
      "step": 43820
    },
    {
      "entropy": 5.846972274780273,
      "epoch": 3.4097257342929392,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00038597322509615977,
      "loss": 4.8706,
      "mean_token_accuracy": 0.22903195321559905,
      "num_tokens": 76031430.0,
      "step": 43825
    },
    {
      "entropy": 5.823302173614502,
      "epoch": 3.4101147636646565,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038594911424778925,
      "loss": 4.9398,
      "mean_token_accuracy": 0.21545860022306443,
      "num_tokens": 76039387.0,
      "step": 43830
    },
    {
      "entropy": 5.845300722122192,
      "epoch": 3.410503793036374,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038592500171594245,
      "loss": 4.9027,
      "mean_token_accuracy": 0.20591666400432587,
      "num_tokens": 76047461.0,
      "step": 43835
    },
    {
      "entropy": 5.792436504364014,
      "epoch": 3.410892822408092,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038590088750098536,
      "loss": 4.9468,
      "mean_token_accuracy": 0.211945278942585,
      "num_tokens": 76056194.0,
      "step": 43840
    },
    {
      "entropy": 5.907734680175781,
      "epoch": 3.411281851779809,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038587677160328377,
      "loss": 5.1354,
      "mean_token_accuracy": 0.1988607481122017,
      "num_tokens": 76064848.0,
      "step": 43845
    },
    {
      "entropy": 5.824720764160157,
      "epoch": 3.411670881151527,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003858526540232037,
      "loss": 4.9603,
      "mean_token_accuracy": 0.21031819730997087,
      "num_tokens": 76073972.0,
      "step": 43850
    },
    {
      "entropy": 5.840104770660401,
      "epoch": 3.4120599105232445,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038582853476111094,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20448665022850038,
      "num_tokens": 76082763.0,
      "step": 43855
    },
    {
      "entropy": 5.819099235534668,
      "epoch": 3.4124489398949622,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003858044138173717,
      "loss": 4.9475,
      "mean_token_accuracy": 0.21147241443395615,
      "num_tokens": 76091093.0,
      "step": 43860
    },
    {
      "entropy": 5.822482490539551,
      "epoch": 3.4128379692666795,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038578029119235176,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21598471254110335,
      "num_tokens": 76099633.0,
      "step": 43865
    },
    {
      "entropy": 5.87732162475586,
      "epoch": 3.413226998638397,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038575616688641724,
      "loss": 4.9629,
      "mean_token_accuracy": 0.20790741443634034,
      "num_tokens": 76108519.0,
      "step": 43870
    },
    {
      "entropy": 5.807954216003418,
      "epoch": 3.413616028010115,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003857320408999342,
      "loss": 4.8855,
      "mean_token_accuracy": 0.2212754800915718,
      "num_tokens": 76116700.0,
      "step": 43875
    },
    {
      "entropy": 5.782957935333252,
      "epoch": 3.414005057381832,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003857079132332687,
      "loss": 4.9387,
      "mean_token_accuracy": 0.21093517392873765,
      "num_tokens": 76125200.0,
      "step": 43880
    },
    {
      "entropy": 5.824362802505493,
      "epoch": 3.41439408675355,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038568378388678675,
      "loss": 5.0074,
      "mean_token_accuracy": 0.20993922650814056,
      "num_tokens": 76134586.0,
      "step": 43885
    },
    {
      "entropy": 5.886890172958374,
      "epoch": 3.4147831161252675,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003856596528608546,
      "loss": 4.9753,
      "mean_token_accuracy": 0.2100835531949997,
      "num_tokens": 76143334.0,
      "step": 43890
    },
    {
      "entropy": 5.945027256011963,
      "epoch": 3.4151721454969852,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003856355201558384,
      "loss": 5.0795,
      "mean_token_accuracy": 0.20208909660577773,
      "num_tokens": 76152896.0,
      "step": 43895
    },
    {
      "entropy": 5.7728355884552,
      "epoch": 3.4155611748687025,
      "grad_norm": 1.390625,
      "learning_rate": 0.00038561138577210426,
      "loss": 4.8603,
      "mean_token_accuracy": 0.21950095742940903,
      "num_tokens": 76161288.0,
      "step": 43900
    },
    {
      "entropy": 5.8665062427520756,
      "epoch": 3.41595020424042,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038558724971001835,
      "loss": 4.9127,
      "mean_token_accuracy": 0.21739864349365234,
      "num_tokens": 76169517.0,
      "step": 43905
    },
    {
      "entropy": 5.836590194702149,
      "epoch": 3.416339233612138,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000385563111969947,
      "loss": 5.0133,
      "mean_token_accuracy": 0.21415163874626159,
      "num_tokens": 76178536.0,
      "step": 43910
    },
    {
      "entropy": 5.789945602416992,
      "epoch": 3.416728262983855,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038553897255225633,
      "loss": 4.982,
      "mean_token_accuracy": 0.2053741455078125,
      "num_tokens": 76186660.0,
      "step": 43915
    },
    {
      "entropy": 5.7879496097564695,
      "epoch": 3.417117292355573,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038551483145731277,
      "loss": 4.8446,
      "mean_token_accuracy": 0.2163775384426117,
      "num_tokens": 76195185.0,
      "step": 43920
    },
    {
      "entropy": 5.8237841606140135,
      "epoch": 3.4175063217272905,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003854906886854825,
      "loss": 4.8956,
      "mean_token_accuracy": 0.21647748351097107,
      "num_tokens": 76204060.0,
      "step": 43925
    },
    {
      "entropy": 5.831596708297729,
      "epoch": 3.417895351099008,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038546654423713187,
      "loss": 4.9763,
      "mean_token_accuracy": 0.2152456447482109,
      "num_tokens": 76212007.0,
      "step": 43930
    },
    {
      "entropy": 5.827530479431152,
      "epoch": 3.4182843804707255,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003854423981126273,
      "loss": 4.9797,
      "mean_token_accuracy": 0.20387081205844879,
      "num_tokens": 76220258.0,
      "step": 43935
    },
    {
      "entropy": 5.852058744430542,
      "epoch": 3.418673409842443,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003854182503123351,
      "loss": 4.9467,
      "mean_token_accuracy": 0.21196922957897185,
      "num_tokens": 76228097.0,
      "step": 43940
    },
    {
      "entropy": 5.882218360900879,
      "epoch": 3.4190624392141604,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038539410083662166,
      "loss": 4.9854,
      "mean_token_accuracy": 0.20443921983242036,
      "num_tokens": 76236227.0,
      "step": 43945
    },
    {
      "entropy": 5.854920196533203,
      "epoch": 3.419451468585878,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038536994968585355,
      "loss": 4.9848,
      "mean_token_accuracy": 0.20739775896072388,
      "num_tokens": 76245009.0,
      "step": 43950
    },
    {
      "entropy": 5.919456148147583,
      "epoch": 3.419840497957596,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038534579686039705,
      "loss": 5.0868,
      "mean_token_accuracy": 0.1970907285809517,
      "num_tokens": 76254553.0,
      "step": 43955
    },
    {
      "entropy": 5.854041337966919,
      "epoch": 3.4202295273293135,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038532164236061874,
      "loss": 5.0032,
      "mean_token_accuracy": 0.21112479418516159,
      "num_tokens": 76263822.0,
      "step": 43960
    },
    {
      "entropy": 5.764766693115234,
      "epoch": 3.4206185567010308,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000385297486186885,
      "loss": 4.7816,
      "mean_token_accuracy": 0.21902240961790084,
      "num_tokens": 76273103.0,
      "step": 43965
    },
    {
      "entropy": 5.8248310565948485,
      "epoch": 3.4210075860727485,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003852733283395626,
      "loss": 4.9962,
      "mean_token_accuracy": 0.21147226691246032,
      "num_tokens": 76282248.0,
      "step": 43970
    },
    {
      "entropy": 5.8378016471862795,
      "epoch": 3.421396615444466,
      "grad_norm": 1.359375,
      "learning_rate": 0.000385249168819018,
      "loss": 4.9349,
      "mean_token_accuracy": 0.2157124400138855,
      "num_tokens": 76291147.0,
      "step": 43975
    },
    {
      "entropy": 5.88880090713501,
      "epoch": 3.4217856448161834,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003852250076256177,
      "loss": 4.9535,
      "mean_token_accuracy": 0.21774826645851136,
      "num_tokens": 76299713.0,
      "step": 43980
    },
    {
      "entropy": 5.772650575637817,
      "epoch": 3.422174674187901,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003852008447597283,
      "loss": 4.9252,
      "mean_token_accuracy": 0.21542322039604186,
      "num_tokens": 76309661.0,
      "step": 43985
    },
    {
      "entropy": 5.832460546493531,
      "epoch": 3.422563703559619,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038517668022171656,
      "loss": 5.0137,
      "mean_token_accuracy": 0.20649162232875823,
      "num_tokens": 76318473.0,
      "step": 43990
    },
    {
      "entropy": 5.9087177276611325,
      "epoch": 3.4229527329313365,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003851525140119489,
      "loss": 4.9825,
      "mean_token_accuracy": 0.21006109863519667,
      "num_tokens": 76327503.0,
      "step": 43995
    },
    {
      "entropy": 5.845364809036255,
      "epoch": 3.4233417623030538,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038512834613079226,
      "loss": 5.0194,
      "mean_token_accuracy": 0.2066958725452423,
      "num_tokens": 76335582.0,
      "step": 44000
    },
    {
      "entropy": 5.83451771736145,
      "epoch": 3.4237307916747715,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003851041765786133,
      "loss": 4.9893,
      "mean_token_accuracy": 0.21417795568704606,
      "num_tokens": 76344978.0,
      "step": 44005
    },
    {
      "entropy": 5.888021802902221,
      "epoch": 3.424119821046489,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038508000535577864,
      "loss": 5.002,
      "mean_token_accuracy": 0.2062072053551674,
      "num_tokens": 76353616.0,
      "step": 44010
    },
    {
      "entropy": 5.935391187667847,
      "epoch": 3.4245088504182064,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003850558324626552,
      "loss": 4.9977,
      "mean_token_accuracy": 0.20849503129720687,
      "num_tokens": 76361468.0,
      "step": 44015
    },
    {
      "entropy": 5.852973890304566,
      "epoch": 3.424897879789924,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003850316578996096,
      "loss": 4.9743,
      "mean_token_accuracy": 0.21723642796278,
      "num_tokens": 76370100.0,
      "step": 44020
    },
    {
      "entropy": 5.849349880218506,
      "epoch": 3.425286909161642,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003850074816670087,
      "loss": 4.9949,
      "mean_token_accuracy": 0.21705928146839143,
      "num_tokens": 76379167.0,
      "step": 44025
    },
    {
      "entropy": 5.760069799423218,
      "epoch": 3.4256759385333595,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038498330376521945,
      "loss": 4.7983,
      "mean_token_accuracy": 0.22141379117965698,
      "num_tokens": 76387456.0,
      "step": 44030
    },
    {
      "entropy": 5.929547882080078,
      "epoch": 3.4260649679050768,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003849591241946085,
      "loss": 5.0886,
      "mean_token_accuracy": 0.1950438439846039,
      "num_tokens": 76397139.0,
      "step": 44035
    },
    {
      "entropy": 5.864507007598877,
      "epoch": 3.4264539972767944,
      "grad_norm": 1.328125,
      "learning_rate": 0.000384934942955543,
      "loss": 4.944,
      "mean_token_accuracy": 0.21518368273973465,
      "num_tokens": 76405198.0,
      "step": 44040
    },
    {
      "entropy": 5.887068271636963,
      "epoch": 3.426843026648512,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038491076004838965,
      "loss": 4.9874,
      "mean_token_accuracy": 0.20953595340251924,
      "num_tokens": 76413237.0,
      "step": 44045
    },
    {
      "entropy": 5.875796461105347,
      "epoch": 3.4272320560202294,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038488657547351544,
      "loss": 5.0019,
      "mean_token_accuracy": 0.21083976328372955,
      "num_tokens": 76421675.0,
      "step": 44050
    },
    {
      "entropy": 5.869209432601929,
      "epoch": 3.427621085391947,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003848623892312874,
      "loss": 4.9695,
      "mean_token_accuracy": 0.21260202527046204,
      "num_tokens": 76430338.0,
      "step": 44055
    },
    {
      "entropy": 5.924650764465332,
      "epoch": 3.428010114763665,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003848382013220725,
      "loss": 5.131,
      "mean_token_accuracy": 0.19443540424108505,
      "num_tokens": 76439001.0,
      "step": 44060
    },
    {
      "entropy": 5.879854774475097,
      "epoch": 3.428399144135382,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003848140117462376,
      "loss": 5.0134,
      "mean_token_accuracy": 0.2093374699354172,
      "num_tokens": 76447695.0,
      "step": 44065
    },
    {
      "entropy": 5.895300579071045,
      "epoch": 3.4287881735070997,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003847898205041499,
      "loss": 4.9874,
      "mean_token_accuracy": 0.20581070482730865,
      "num_tokens": 76456297.0,
      "step": 44070
    },
    {
      "entropy": 5.920821714401245,
      "epoch": 3.4291772028788174,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038476562759617644,
      "loss": 5.1016,
      "mean_token_accuracy": 0.20470652729272842,
      "num_tokens": 76464971.0,
      "step": 44075
    },
    {
      "entropy": 5.776926708221436,
      "epoch": 3.4295662322505347,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038474143302268425,
      "loss": 4.9144,
      "mean_token_accuracy": 0.21680910438299178,
      "num_tokens": 76474134.0,
      "step": 44080
    },
    {
      "entropy": 5.946998453140258,
      "epoch": 3.4299552616222524,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003847172367840406,
      "loss": 5.1546,
      "mean_token_accuracy": 0.1979419022798538,
      "num_tokens": 76482902.0,
      "step": 44085
    },
    {
      "entropy": 5.924585819244385,
      "epoch": 3.43034429099397,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003846930388806124,
      "loss": 5.0078,
      "mean_token_accuracy": 0.20791225135326385,
      "num_tokens": 76490935.0,
      "step": 44090
    },
    {
      "entropy": 5.905887079238892,
      "epoch": 3.430733320365688,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038466883931276703,
      "loss": 5.0027,
      "mean_token_accuracy": 0.21106399446725846,
      "num_tokens": 76498663.0,
      "step": 44095
    },
    {
      "entropy": 5.861296272277832,
      "epoch": 3.431122349737405,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038464463808087146,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20566127598285674,
      "num_tokens": 76507353.0,
      "step": 44100
    },
    {
      "entropy": 5.855917930603027,
      "epoch": 3.4315113791091227,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003846204351852931,
      "loss": 4.989,
      "mean_token_accuracy": 0.21015361398458482,
      "num_tokens": 76515396.0,
      "step": 44105
    },
    {
      "entropy": 5.898738670349121,
      "epoch": 3.4319004084808404,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038459623062639907,
      "loss": 4.9455,
      "mean_token_accuracy": 0.20896498411893843,
      "num_tokens": 76523715.0,
      "step": 44110
    },
    {
      "entropy": 5.811381864547729,
      "epoch": 3.4322894378525577,
      "grad_norm": 1.25,
      "learning_rate": 0.00038457202440455673,
      "loss": 4.9097,
      "mean_token_accuracy": 0.21946179866790771,
      "num_tokens": 76532576.0,
      "step": 44115
    },
    {
      "entropy": 5.8292468070983885,
      "epoch": 3.4326784672242754,
      "grad_norm": 2.171875,
      "learning_rate": 0.0003845478165201333,
      "loss": 4.8454,
      "mean_token_accuracy": 0.23248335719108582,
      "num_tokens": 76541747.0,
      "step": 44120
    },
    {
      "entropy": 5.804116487503052,
      "epoch": 3.433067496595993,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038452360697349615,
      "loss": 5.0084,
      "mean_token_accuracy": 0.21379686146974564,
      "num_tokens": 76550399.0,
      "step": 44125
    },
    {
      "entropy": 5.889016342163086,
      "epoch": 3.433456525967711,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003844993957650125,
      "loss": 5.0462,
      "mean_token_accuracy": 0.19934344291687012,
      "num_tokens": 76559222.0,
      "step": 44130
    },
    {
      "entropy": 5.883822774887085,
      "epoch": 3.433845555339428,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003844751828950499,
      "loss": 5.0451,
      "mean_token_accuracy": 0.20084514915943147,
      "num_tokens": 76568097.0,
      "step": 44135
    },
    {
      "entropy": 5.8990936279296875,
      "epoch": 3.4342345847111457,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038445096836397564,
      "loss": 5.0429,
      "mean_token_accuracy": 0.20541415959596634,
      "num_tokens": 76577014.0,
      "step": 44140
    },
    {
      "entropy": 5.864108610153198,
      "epoch": 3.4346236140828634,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038442675217215713,
      "loss": 5.0092,
      "mean_token_accuracy": 0.2038536861538887,
      "num_tokens": 76585787.0,
      "step": 44145
    },
    {
      "entropy": 5.853636264801025,
      "epoch": 3.4350126434545807,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038440253431996175,
      "loss": 4.9823,
      "mean_token_accuracy": 0.21245600432157516,
      "num_tokens": 76594211.0,
      "step": 44150
    },
    {
      "entropy": 5.81405348777771,
      "epoch": 3.4354016728262984,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038437831480775717,
      "loss": 4.9106,
      "mean_token_accuracy": 0.21415332555770875,
      "num_tokens": 76602536.0,
      "step": 44155
    },
    {
      "entropy": 5.893564939498901,
      "epoch": 3.435790702198016,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00038435409363591066,
      "loss": 5.0546,
      "mean_token_accuracy": 0.20723741501569748,
      "num_tokens": 76611138.0,
      "step": 44160
    },
    {
      "entropy": 5.875454521179199,
      "epoch": 3.4361797315697333,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038432987080478986,
      "loss": 5.0067,
      "mean_token_accuracy": 0.21487057507038115,
      "num_tokens": 76619377.0,
      "step": 44165
    },
    {
      "entropy": 5.916072511672974,
      "epoch": 3.436568760941451,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003843056463147623,
      "loss": 5.1238,
      "mean_token_accuracy": 0.20404950082302092,
      "num_tokens": 76627860.0,
      "step": 44170
    },
    {
      "entropy": 5.7767781734466555,
      "epoch": 3.4369577903131687,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038428142016619555,
      "loss": 4.9788,
      "mean_token_accuracy": 0.2046142816543579,
      "num_tokens": 76637195.0,
      "step": 44175
    },
    {
      "entropy": 5.896539878845215,
      "epoch": 3.437346819684886,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003842571923594572,
      "loss": 4.9718,
      "mean_token_accuracy": 0.2172362193465233,
      "num_tokens": 76645988.0,
      "step": 44180
    },
    {
      "entropy": 5.899210834503174,
      "epoch": 3.4377358490566037,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038423296289491477,
      "loss": 5.0173,
      "mean_token_accuracy": 0.21032586097717285,
      "num_tokens": 76654375.0,
      "step": 44185
    },
    {
      "entropy": 5.798703622817993,
      "epoch": 3.4381248784283214,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038420873177293606,
      "loss": 4.9709,
      "mean_token_accuracy": 0.21007563471794127,
      "num_tokens": 76663217.0,
      "step": 44190
    },
    {
      "entropy": 5.9870216846466064,
      "epoch": 3.438513907800039,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003841844989938886,
      "loss": 5.1064,
      "mean_token_accuracy": 0.20313764363527298,
      "num_tokens": 76672328.0,
      "step": 44195
    },
    {
      "entropy": 5.89600977897644,
      "epoch": 3.4389029371717563,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003841602645581402,
      "loss": 4.9856,
      "mean_token_accuracy": 0.21131310611963272,
      "num_tokens": 76681005.0,
      "step": 44200
    },
    {
      "entropy": 5.809788608551026,
      "epoch": 3.439291966543474,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00038413602846605854,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21348177939653395,
      "num_tokens": 76689883.0,
      "step": 44205
    },
    {
      "entropy": 5.860534811019898,
      "epoch": 3.4396809959151917,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003841117907180114,
      "loss": 5.0145,
      "mean_token_accuracy": 0.20696192234754562,
      "num_tokens": 76698953.0,
      "step": 44210
    },
    {
      "entropy": 5.922059202194214,
      "epoch": 3.440070025286909,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038408755131436647,
      "loss": 5.0559,
      "mean_token_accuracy": 0.20818686485290527,
      "num_tokens": 76708531.0,
      "step": 44215
    },
    {
      "entropy": 5.926369619369507,
      "epoch": 3.4404590546586267,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00038406331025549157,
      "loss": 5.0223,
      "mean_token_accuracy": 0.20300254672765733,
      "num_tokens": 76716715.0,
      "step": 44220
    },
    {
      "entropy": 5.89971604347229,
      "epoch": 3.4408480840303444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003840390675417545,
      "loss": 4.9489,
      "mean_token_accuracy": 0.2173616334795952,
      "num_tokens": 76725081.0,
      "step": 44225
    },
    {
      "entropy": 5.841331148147583,
      "epoch": 3.441237113402062,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003840148231735231,
      "loss": 4.9434,
      "mean_token_accuracy": 0.21376111209392548,
      "num_tokens": 76732995.0,
      "step": 44230
    },
    {
      "entropy": 5.848056125640869,
      "epoch": 3.4416261427737793,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038399057715116536,
      "loss": 4.9683,
      "mean_token_accuracy": 0.21805143505334854,
      "num_tokens": 76741863.0,
      "step": 44235
    },
    {
      "entropy": 5.911487197875976,
      "epoch": 3.442015172145497,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038396632947504904,
      "loss": 4.9797,
      "mean_token_accuracy": 0.21770156025886536,
      "num_tokens": 76751281.0,
      "step": 44240
    },
    {
      "entropy": 5.901050424575805,
      "epoch": 3.4424042015172147,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003839420801455422,
      "loss": 4.9413,
      "mean_token_accuracy": 0.21233413368463516,
      "num_tokens": 76758803.0,
      "step": 44245
    },
    {
      "entropy": 5.874415683746338,
      "epoch": 3.442793230888932,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003839178291630126,
      "loss": 5.0513,
      "mean_token_accuracy": 0.20249511152505875,
      "num_tokens": 76767833.0,
      "step": 44250
    },
    {
      "entropy": 5.943673086166382,
      "epoch": 3.4431822602606497,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038389357652782836,
      "loss": 5.0273,
      "mean_token_accuracy": 0.21299265176057816,
      "num_tokens": 76776046.0,
      "step": 44255
    },
    {
      "entropy": 5.907406759262085,
      "epoch": 3.4435712896323674,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038386932224035734,
      "loss": 4.928,
      "mean_token_accuracy": 0.2177903324365616,
      "num_tokens": 76784753.0,
      "step": 44260
    },
    {
      "entropy": 5.845219230651855,
      "epoch": 3.443960319004085,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00038384506630096765,
      "loss": 5.018,
      "mean_token_accuracy": 0.20693393647670746,
      "num_tokens": 76793920.0,
      "step": 44265
    },
    {
      "entropy": 5.890315151214599,
      "epoch": 3.4443493483758023,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003838208087100274,
      "loss": 5.0756,
      "mean_token_accuracy": 0.20843016654253005,
      "num_tokens": 76802575.0,
      "step": 44270
    },
    {
      "entropy": 5.786786127090454,
      "epoch": 3.44473837774752,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003837965494679045,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2169828549027443,
      "num_tokens": 76810962.0,
      "step": 44275
    },
    {
      "entropy": 5.792293930053711,
      "epoch": 3.4451274071192373,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003837722885749672,
      "loss": 4.9449,
      "mean_token_accuracy": 0.2142533466219902,
      "num_tokens": 76819408.0,
      "step": 44280
    },
    {
      "entropy": 5.872972440719605,
      "epoch": 3.445516436490955,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003837480260315835,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2159590482711792,
      "num_tokens": 76826888.0,
      "step": 44285
    },
    {
      "entropy": 5.913843059539795,
      "epoch": 3.4459054658626727,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038372376183812165,
      "loss": 4.9754,
      "mean_token_accuracy": 0.213849338889122,
      "num_tokens": 76836649.0,
      "step": 44290
    },
    {
      "entropy": 5.83642692565918,
      "epoch": 3.4462944952343904,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003836994959949497,
      "loss": 4.9742,
      "mean_token_accuracy": 0.2113280713558197,
      "num_tokens": 76845123.0,
      "step": 44295
    },
    {
      "entropy": 5.826214122772217,
      "epoch": 3.4466835246061076,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038367522850243593,
      "loss": 4.8854,
      "mean_token_accuracy": 0.21889735907316207,
      "num_tokens": 76853595.0,
      "step": 44300
    },
    {
      "entropy": 5.82852578163147,
      "epoch": 3.4470725539778253,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038365095936094857,
      "loss": 4.9745,
      "mean_token_accuracy": 0.21296181827783583,
      "num_tokens": 76862765.0,
      "step": 44305
    },
    {
      "entropy": 5.814942216873169,
      "epoch": 3.447461583349543,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003836266885708558,
      "loss": 4.9498,
      "mean_token_accuracy": 0.20942875146865844,
      "num_tokens": 76870825.0,
      "step": 44310
    },
    {
      "entropy": 5.942566967010498,
      "epoch": 3.4478506127212603,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038360241613252593,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20974048525094985,
      "num_tokens": 76880085.0,
      "step": 44315
    },
    {
      "entropy": 5.872527027130127,
      "epoch": 3.448239642092978,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003835781420463273,
      "loss": 4.9944,
      "mean_token_accuracy": 0.20799621492624282,
      "num_tokens": 76889231.0,
      "step": 44320
    },
    {
      "entropy": 5.826364469528198,
      "epoch": 3.4486286714646956,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003835538663126282,
      "loss": 4.9981,
      "mean_token_accuracy": 0.2091787949204445,
      "num_tokens": 76898151.0,
      "step": 44325
    },
    {
      "entropy": 5.9477729320526125,
      "epoch": 3.4490177008364133,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003835295889317969,
      "loss": 5.0476,
      "mean_token_accuracy": 0.20028073191642762,
      "num_tokens": 76906638.0,
      "step": 44330
    },
    {
      "entropy": 5.872470569610596,
      "epoch": 3.4494067302081306,
      "grad_norm": 1.421875,
      "learning_rate": 0.00038350530990420177,
      "loss": 4.9842,
      "mean_token_accuracy": 0.21813303232192993,
      "num_tokens": 76914986.0,
      "step": 44335
    },
    {
      "entropy": 5.848671007156372,
      "epoch": 3.4497957595798483,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003834810292302114,
      "loss": 4.9083,
      "mean_token_accuracy": 0.2166862815618515,
      "num_tokens": 76924290.0,
      "step": 44340
    },
    {
      "entropy": 5.8444701671600345,
      "epoch": 3.450184788951566,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000383456746910194,
      "loss": 4.9959,
      "mean_token_accuracy": 0.21167974621057511,
      "num_tokens": 76933159.0,
      "step": 44345
    },
    {
      "entropy": 5.853628253936767,
      "epoch": 3.4505738183232832,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003834324629445182,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2064526528120041,
      "num_tokens": 76941436.0,
      "step": 44350
    },
    {
      "entropy": 5.873377418518066,
      "epoch": 3.450962847695001,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003834081773335522,
      "loss": 4.9882,
      "mean_token_accuracy": 0.20205493569374083,
      "num_tokens": 76949760.0,
      "step": 44355
    },
    {
      "entropy": 5.836436176300049,
      "epoch": 3.4513518770667186,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038338389007766473,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21715426146984101,
      "num_tokens": 76958941.0,
      "step": 44360
    },
    {
      "entropy": 5.879078435897827,
      "epoch": 3.4517409064384363,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003833596011772242,
      "loss": 5.0474,
      "mean_token_accuracy": 0.20372789800167085,
      "num_tokens": 76967881.0,
      "step": 44365
    },
    {
      "entropy": 5.856692743301392,
      "epoch": 3.4521299358101536,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003833353106325993,
      "loss": 4.9891,
      "mean_token_accuracy": 0.20820915549993516,
      "num_tokens": 76976059.0,
      "step": 44370
    },
    {
      "entropy": 5.8218165874481205,
      "epoch": 3.4525189651818713,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003833110184441584,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21285669654607772,
      "num_tokens": 76985224.0,
      "step": 44375
    },
    {
      "entropy": 5.927899169921875,
      "epoch": 3.452907994553589,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038328672461227024,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20375314950942994,
      "num_tokens": 76994189.0,
      "step": 44380
    },
    {
      "entropy": 5.892507648468017,
      "epoch": 3.4532970239253062,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003832624291373034,
      "loss": 4.9696,
      "mean_token_accuracy": 0.2096343070268631,
      "num_tokens": 77003701.0,
      "step": 44385
    },
    {
      "entropy": 5.822946643829345,
      "epoch": 3.453686053297024,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038323813201962645,
      "loss": 4.9515,
      "mean_token_accuracy": 0.21411342322826385,
      "num_tokens": 77012929.0,
      "step": 44390
    },
    {
      "entropy": 5.805064916610718,
      "epoch": 3.4540750826687416,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038321383325960816,
      "loss": 4.9317,
      "mean_token_accuracy": 0.2116946056485176,
      "num_tokens": 77022045.0,
      "step": 44395
    },
    {
      "entropy": 5.8638969421386715,
      "epoch": 3.454464112040459,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003831895328576172,
      "loss": 4.9474,
      "mean_token_accuracy": 0.22019802033901215,
      "num_tokens": 77030197.0,
      "step": 44400
    },
    {
      "entropy": 5.773309946060181,
      "epoch": 3.4548531414121766,
      "grad_norm": 1.25,
      "learning_rate": 0.0003831652308140223,
      "loss": 4.9242,
      "mean_token_accuracy": 0.2069219633936882,
      "num_tokens": 77038939.0,
      "step": 44405
    },
    {
      "entropy": 5.828426218032837,
      "epoch": 3.4552421707838943,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003831409271291921,
      "loss": 5.0086,
      "mean_token_accuracy": 0.2117981106042862,
      "num_tokens": 77047314.0,
      "step": 44410
    },
    {
      "entropy": 5.8918601989746096,
      "epoch": 3.4556312001556115,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038311662180349554,
      "loss": 4.999,
      "mean_token_accuracy": 0.2089047908782959,
      "num_tokens": 77055100.0,
      "step": 44415
    },
    {
      "entropy": 5.835245084762573,
      "epoch": 3.4560202295273292,
      "grad_norm": 1.375,
      "learning_rate": 0.00038309231483730133,
      "loss": 5.0103,
      "mean_token_accuracy": 0.20763815939426422,
      "num_tokens": 77064240.0,
      "step": 44420
    },
    {
      "entropy": 5.823662900924683,
      "epoch": 3.456409258899047,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003830680062309782,
      "loss": 4.8898,
      "mean_token_accuracy": 0.21590039432048796,
      "num_tokens": 77073274.0,
      "step": 44425
    },
    {
      "entropy": 5.84425539970398,
      "epoch": 3.4567982882707646,
      "grad_norm": 1.421875,
      "learning_rate": 0.00038304369598489516,
      "loss": 4.9975,
      "mean_token_accuracy": 0.21598760187625884,
      "num_tokens": 77081157.0,
      "step": 44430
    },
    {
      "entropy": 5.850561714172363,
      "epoch": 3.457187317642482,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000383019384099421,
      "loss": 4.9716,
      "mean_token_accuracy": 0.20854909271001815,
      "num_tokens": 77090045.0,
      "step": 44435
    },
    {
      "entropy": 5.890961742401123,
      "epoch": 3.4575763470141996,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003829950705749246,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21714490205049514,
      "num_tokens": 77098559.0,
      "step": 44440
    },
    {
      "entropy": 5.919901704788208,
      "epoch": 3.4579653763859173,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038297075541177496,
      "loss": 5.0545,
      "mean_token_accuracy": 0.212209253013134,
      "num_tokens": 77107033.0,
      "step": 44445
    },
    {
      "entropy": 5.8210916996002195,
      "epoch": 3.4583544057576345,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038294643861034087,
      "loss": 4.9056,
      "mean_token_accuracy": 0.2134287789463997,
      "num_tokens": 77115867.0,
      "step": 44450
    },
    {
      "entropy": 5.866188669204712,
      "epoch": 3.4587434351293522,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003829221201709914,
      "loss": 5.0265,
      "mean_token_accuracy": 0.2061479091644287,
      "num_tokens": 77124938.0,
      "step": 44455
    },
    {
      "entropy": 5.863016796112061,
      "epoch": 3.45913246450107,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003828978000940955,
      "loss": 4.9785,
      "mean_token_accuracy": 0.20372442454099654,
      "num_tokens": 77134362.0,
      "step": 44460
    },
    {
      "entropy": 5.932760715484619,
      "epoch": 3.4595214938727876,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003828734783800223,
      "loss": 4.9893,
      "mean_token_accuracy": 0.208194862306118,
      "num_tokens": 77142633.0,
      "step": 44465
    },
    {
      "entropy": 5.86742844581604,
      "epoch": 3.459910523244505,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003828491550291408,
      "loss": 4.942,
      "mean_token_accuracy": 0.21366048008203506,
      "num_tokens": 77151313.0,
      "step": 44470
    },
    {
      "entropy": 5.86872444152832,
      "epoch": 3.4602995526162226,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003828248300418199,
      "loss": 5.0029,
      "mean_token_accuracy": 0.20709519237279891,
      "num_tokens": 77159139.0,
      "step": 44475
    },
    {
      "entropy": 5.870968437194824,
      "epoch": 3.4606885819879403,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038280050341842894,
      "loss": 5.0399,
      "mean_token_accuracy": 0.1981135219335556,
      "num_tokens": 77167747.0,
      "step": 44480
    },
    {
      "entropy": 5.8439706325531,
      "epoch": 3.4610776113596575,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003827761751593369,
      "loss": 4.9297,
      "mean_token_accuracy": 0.21200936883687974,
      "num_tokens": 77176505.0,
      "step": 44485
    },
    {
      "entropy": 5.869344091415405,
      "epoch": 3.461466640731375,
      "grad_norm": 1.25,
      "learning_rate": 0.0003827518452649129,
      "loss": 4.9957,
      "mean_token_accuracy": 0.20692320317029952,
      "num_tokens": 77185396.0,
      "step": 44490
    },
    {
      "entropy": 5.939345455169677,
      "epoch": 3.461855670103093,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003827275137355261,
      "loss": 5.0454,
      "mean_token_accuracy": 0.2015735313296318,
      "num_tokens": 77193558.0,
      "step": 44495
    },
    {
      "entropy": 5.917511701583862,
      "epoch": 3.46224469947481,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003827031805715458,
      "loss": 5.0037,
      "mean_token_accuracy": 0.20787128210067748,
      "num_tokens": 77202783.0,
      "step": 44500
    },
    {
      "entropy": 5.846156978607178,
      "epoch": 3.462633728846528,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038267884577334124,
      "loss": 4.9522,
      "mean_token_accuracy": 0.21177664250135422,
      "num_tokens": 77211356.0,
      "step": 44505
    },
    {
      "entropy": 5.9159839153289795,
      "epoch": 3.4630227582182456,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003826545093412815,
      "loss": 4.9756,
      "mean_token_accuracy": 0.2117762193083763,
      "num_tokens": 77219401.0,
      "step": 44510
    },
    {
      "entropy": 5.872131109237671,
      "epoch": 3.463411787589963,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038263017127573596,
      "loss": 5.035,
      "mean_token_accuracy": 0.20968535989522935,
      "num_tokens": 77228436.0,
      "step": 44515
    },
    {
      "entropy": 5.885397911071777,
      "epoch": 3.4638008169616805,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003826058315770739,
      "loss": 5.0436,
      "mean_token_accuracy": 0.20858116149902345,
      "num_tokens": 77237547.0,
      "step": 44520
    },
    {
      "entropy": 5.8960357189178465,
      "epoch": 3.464189846333398,
      "grad_norm": 1.390625,
      "learning_rate": 0.00038258149024566465,
      "loss": 4.9909,
      "mean_token_accuracy": 0.21531523615121842,
      "num_tokens": 77247512.0,
      "step": 44525
    },
    {
      "entropy": 5.895198917388916,
      "epoch": 3.464578875705116,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003825571472818774,
      "loss": 5.0325,
      "mean_token_accuracy": 0.20045370906591414,
      "num_tokens": 77256511.0,
      "step": 44530
    },
    {
      "entropy": 5.95631685256958,
      "epoch": 3.464967905076833,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003825328026860817,
      "loss": 5.0717,
      "mean_token_accuracy": 0.20416852086782455,
      "num_tokens": 77264986.0,
      "step": 44535
    },
    {
      "entropy": 5.919108772277832,
      "epoch": 3.465356934448551,
      "grad_norm": 1.375,
      "learning_rate": 0.00038250845645864686,
      "loss": 5.0445,
      "mean_token_accuracy": 0.21106908321380616,
      "num_tokens": 77273531.0,
      "step": 44540
    },
    {
      "entropy": 5.832286596298218,
      "epoch": 3.4657459638202686,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003824841085999423,
      "loss": 4.9387,
      "mean_token_accuracy": 0.2083718404173851,
      "num_tokens": 77282279.0,
      "step": 44545
    },
    {
      "entropy": 5.890779495239258,
      "epoch": 3.466134993191986,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003824597591103375,
      "loss": 4.9544,
      "mean_token_accuracy": 0.2134363666176796,
      "num_tokens": 77290584.0,
      "step": 44550
    },
    {
      "entropy": 5.874242687225342,
      "epoch": 3.4665240225637035,
      "grad_norm": 1.390625,
      "learning_rate": 0.00038243540799020193,
      "loss": 5.0525,
      "mean_token_accuracy": 0.20557764172554016,
      "num_tokens": 77299706.0,
      "step": 44555
    },
    {
      "entropy": 5.918213748931885,
      "epoch": 3.466913051935421,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038241105523990497,
      "loss": 4.9997,
      "mean_token_accuracy": 0.20792745500802995,
      "num_tokens": 77308616.0,
      "step": 44560
    },
    {
      "entropy": 5.88009877204895,
      "epoch": 3.467302081307139,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003823867008598163,
      "loss": 5.0212,
      "mean_token_accuracy": 0.20499771684408188,
      "num_tokens": 77317740.0,
      "step": 44565
    },
    {
      "entropy": 5.8326293468475345,
      "epoch": 3.467691110678856,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038236234485030524,
      "loss": 4.9518,
      "mean_token_accuracy": 0.2144574925303459,
      "num_tokens": 77326868.0,
      "step": 44570
    },
    {
      "entropy": 5.876618671417236,
      "epoch": 3.468080140050574,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038233798721174156,
      "loss": 5.0065,
      "mean_token_accuracy": 0.21197029799222947,
      "num_tokens": 77335276.0,
      "step": 44575
    },
    {
      "entropy": 5.834342813491821,
      "epoch": 3.4684691694222916,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003823136279444948,
      "loss": 4.9459,
      "mean_token_accuracy": 0.22099836468696593,
      "num_tokens": 77343765.0,
      "step": 44580
    },
    {
      "entropy": 5.844537162780762,
      "epoch": 3.468858198794009,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003822892670489345,
      "loss": 4.948,
      "mean_token_accuracy": 0.21600262969732284,
      "num_tokens": 77352326.0,
      "step": 44585
    },
    {
      "entropy": 5.882081317901611,
      "epoch": 3.4692472281657265,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038226490452543036,
      "loss": 5.0025,
      "mean_token_accuracy": 0.2068750113248825,
      "num_tokens": 77360893.0,
      "step": 44590
    },
    {
      "entropy": 5.845548629760742,
      "epoch": 3.469636257537444,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000382240540374352,
      "loss": 4.9289,
      "mean_token_accuracy": 0.21735809296369552,
      "num_tokens": 77370376.0,
      "step": 44595
    },
    {
      "entropy": 5.8438496589660645,
      "epoch": 3.4700252869091615,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003822161745960691,
      "loss": 4.9853,
      "mean_token_accuracy": 0.21068590730428696,
      "num_tokens": 77379800.0,
      "step": 44600
    },
    {
      "entropy": 5.827046060562134,
      "epoch": 3.470414316280879,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003821918071909515,
      "loss": 4.9233,
      "mean_token_accuracy": 0.2136423483490944,
      "num_tokens": 77388128.0,
      "step": 44605
    },
    {
      "entropy": 5.915372848510742,
      "epoch": 3.470803345652597,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003821674381593687,
      "loss": 5.0954,
      "mean_token_accuracy": 0.20389765352010727,
      "num_tokens": 77396320.0,
      "step": 44610
    },
    {
      "entropy": 5.861430883407593,
      "epoch": 3.471192375024314,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003821430675016907,
      "loss": 5.0244,
      "mean_token_accuracy": 0.20270322114229203,
      "num_tokens": 77405980.0,
      "step": 44615
    },
    {
      "entropy": 5.8808784008026125,
      "epoch": 3.471581404396032,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038211869521828716,
      "loss": 5.0032,
      "mean_token_accuracy": 0.21069419980049134,
      "num_tokens": 77414634.0,
      "step": 44620
    },
    {
      "entropy": 5.84755539894104,
      "epoch": 3.4719704337677495,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000382094321309528,
      "loss": 4.8783,
      "mean_token_accuracy": 0.21882979869842528,
      "num_tokens": 77422945.0,
      "step": 44625
    },
    {
      "entropy": 5.819676160812378,
      "epoch": 3.472359463139467,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003820699457757829,
      "loss": 4.9839,
      "mean_token_accuracy": 0.21350980997085572,
      "num_tokens": 77431848.0,
      "step": 44630
    },
    {
      "entropy": 5.827579593658447,
      "epoch": 3.4727484925111844,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003820455686174218,
      "loss": 5.0319,
      "mean_token_accuracy": 0.19985178709030152,
      "num_tokens": 77440822.0,
      "step": 44635
    },
    {
      "entropy": 5.917607307434082,
      "epoch": 3.473137521882902,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038202118983481456,
      "loss": 5.0998,
      "mean_token_accuracy": 0.20173296332359314,
      "num_tokens": 77449320.0,
      "step": 44640
    },
    {
      "entropy": 5.771625709533692,
      "epoch": 3.47352655125462,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003819968094283311,
      "loss": 4.789,
      "mean_token_accuracy": 0.22145420908927918,
      "num_tokens": 77457841.0,
      "step": 44645
    },
    {
      "entropy": 5.839608478546142,
      "epoch": 3.473915580626337,
      "grad_norm": 1.25,
      "learning_rate": 0.0003819724273983414,
      "loss": 4.9398,
      "mean_token_accuracy": 0.218255715072155,
      "num_tokens": 77466751.0,
      "step": 44650
    },
    {
      "entropy": 5.9424254417419435,
      "epoch": 3.474304609998055,
      "grad_norm": 1.421875,
      "learning_rate": 0.00038194804374521536,
      "loss": 5.1393,
      "mean_token_accuracy": 0.19994935691356658,
      "num_tokens": 77475355.0,
      "step": 44655
    },
    {
      "entropy": 5.897541332244873,
      "epoch": 3.4746936393697725,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003819236584693229,
      "loss": 4.9414,
      "mean_token_accuracy": 0.2099450409412384,
      "num_tokens": 77483929.0,
      "step": 44660
    },
    {
      "entropy": 5.84615569114685,
      "epoch": 3.47508266874149,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038189927157103427,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2127579554915428,
      "num_tokens": 77492599.0,
      "step": 44665
    },
    {
      "entropy": 5.758064460754395,
      "epoch": 3.4754716981132074,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003818748830507192,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21731504052877426,
      "num_tokens": 77501856.0,
      "step": 44670
    },
    {
      "entropy": 5.807080173492432,
      "epoch": 3.475860727484925,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000381850492908748,
      "loss": 4.9219,
      "mean_token_accuracy": 0.22511528432369232,
      "num_tokens": 77510557.0,
      "step": 44675
    },
    {
      "entropy": 5.840623903274536,
      "epoch": 3.476249756856643,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038182610114549057,
      "loss": 4.9322,
      "mean_token_accuracy": 0.20995031893253327,
      "num_tokens": 77518737.0,
      "step": 44680
    },
    {
      "entropy": 5.786155462265015,
      "epoch": 3.47663878622836,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038180170776131714,
      "loss": 4.8287,
      "mean_token_accuracy": 0.21844499111175536,
      "num_tokens": 77527433.0,
      "step": 44685
    },
    {
      "entropy": 5.874866056442261,
      "epoch": 3.477027815600078,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003817773127565977,
      "loss": 4.9639,
      "mean_token_accuracy": 0.21056477576494217,
      "num_tokens": 77536441.0,
      "step": 44690
    },
    {
      "entropy": 5.855126190185547,
      "epoch": 3.4774168449717955,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038175291613170247,
      "loss": 5.017,
      "mean_token_accuracy": 0.2122418537735939,
      "num_tokens": 77545772.0,
      "step": 44695
    },
    {
      "entropy": 5.8513476848602295,
      "epoch": 3.477805874343513,
      "grad_norm": 1.25,
      "learning_rate": 0.00038172851788700174,
      "loss": 4.8914,
      "mean_token_accuracy": 0.22107330113649368,
      "num_tokens": 77554631.0,
      "step": 44700
    },
    {
      "entropy": 5.9209259986877445,
      "epoch": 3.4781949037152304,
      "grad_norm": 1.375,
      "learning_rate": 0.0003817041180228657,
      "loss": 5.0275,
      "mean_token_accuracy": 0.1992516651749611,
      "num_tokens": 77563026.0,
      "step": 44705
    },
    {
      "entropy": 5.863089084625244,
      "epoch": 3.478583933086948,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003816797165396643,
      "loss": 5.0584,
      "mean_token_accuracy": 0.20576203614473343,
      "num_tokens": 77571559.0,
      "step": 44710
    },
    {
      "entropy": 5.824941730499267,
      "epoch": 3.4789729624586654,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038165531343776815,
      "loss": 4.9011,
      "mean_token_accuracy": 0.20984706729650499,
      "num_tokens": 77579615.0,
      "step": 44715
    },
    {
      "entropy": 5.8749120235443115,
      "epoch": 3.479361991830383,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003816309087175474,
      "loss": 5.0771,
      "mean_token_accuracy": 0.20524393022060394,
      "num_tokens": 77587967.0,
      "step": 44720
    },
    {
      "entropy": 5.930889320373535,
      "epoch": 3.479751021202101,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038160650237937225,
      "loss": 5.0614,
      "mean_token_accuracy": 0.19971709847450256,
      "num_tokens": 77597358.0,
      "step": 44725
    },
    {
      "entropy": 5.905979633331299,
      "epoch": 3.4801400505738185,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003815820944236131,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21185854077339172,
      "num_tokens": 77605354.0,
      "step": 44730
    },
    {
      "entropy": 5.873653888702393,
      "epoch": 3.4805290799455357,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038155768485064037,
      "loss": 5.0383,
      "mean_token_accuracy": 0.20807576179504395,
      "num_tokens": 77614879.0,
      "step": 44735
    },
    {
      "entropy": 5.900464630126953,
      "epoch": 3.4809181093172534,
      "grad_norm": 1.46875,
      "learning_rate": 0.00038153327366082434,
      "loss": 5.0624,
      "mean_token_accuracy": 0.20154136568307876,
      "num_tokens": 77623437.0,
      "step": 44740
    },
    {
      "entropy": 5.845536184310913,
      "epoch": 3.481307138688971,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038150886085453546,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20731067657470703,
      "num_tokens": 77632054.0,
      "step": 44745
    },
    {
      "entropy": 5.900174951553344,
      "epoch": 3.4816961680606884,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038148444643214415,
      "loss": 5.0281,
      "mean_token_accuracy": 0.20529271215200423,
      "num_tokens": 77640616.0,
      "step": 44750
    },
    {
      "entropy": 5.976004028320313,
      "epoch": 3.482085197432406,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003814600303940208,
      "loss": 5.0875,
      "mean_token_accuracy": 0.1962366670370102,
      "num_tokens": 77649533.0,
      "step": 44755
    },
    {
      "entropy": 5.828184747695923,
      "epoch": 3.4824742268041238,
      "grad_norm": 1.265625,
      "learning_rate": 0.000381435612740536,
      "loss": 4.916,
      "mean_token_accuracy": 0.2239959254860878,
      "num_tokens": 77657929.0,
      "step": 44760
    },
    {
      "entropy": 5.843001508712769,
      "epoch": 3.4828632561758415,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038141119347206016,
      "loss": 5.0152,
      "mean_token_accuracy": 0.2024470493197441,
      "num_tokens": 77666861.0,
      "step": 44765
    },
    {
      "entropy": 5.862302446365357,
      "epoch": 3.4832522855475587,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038138677258896383,
      "loss": 4.9019,
      "mean_token_accuracy": 0.21564800292253494,
      "num_tokens": 77675033.0,
      "step": 44770
    },
    {
      "entropy": 5.850249814987182,
      "epoch": 3.4836413149192764,
      "grad_norm": 1.375,
      "learning_rate": 0.0003813623500916176,
      "loss": 4.9659,
      "mean_token_accuracy": 0.2174173876643181,
      "num_tokens": 77682971.0,
      "step": 44775
    },
    {
      "entropy": 5.8964393615722654,
      "epoch": 3.484030344290994,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038133792598039196,
      "loss": 4.9971,
      "mean_token_accuracy": 0.20960344821214677,
      "num_tokens": 77691735.0,
      "step": 44780
    },
    {
      "entropy": 5.908811759948731,
      "epoch": 3.4844193736627114,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003813135002556575,
      "loss": 5.0289,
      "mean_token_accuracy": 0.2087090104818344,
      "num_tokens": 77700416.0,
      "step": 44785
    },
    {
      "entropy": 5.822631597518921,
      "epoch": 3.484808403034429,
      "grad_norm": 1.375,
      "learning_rate": 0.00038128907291778497,
      "loss": 4.8984,
      "mean_token_accuracy": 0.22063064873218535,
      "num_tokens": 77708814.0,
      "step": 44790
    },
    {
      "entropy": 5.839618444442749,
      "epoch": 3.4851974324061468,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038126464396714494,
      "loss": 5.0031,
      "mean_token_accuracy": 0.21548448652029037,
      "num_tokens": 77717065.0,
      "step": 44795
    },
    {
      "entropy": 5.830079507827759,
      "epoch": 3.4855864617778645,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038124021340410795,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2075408712029457,
      "num_tokens": 77725866.0,
      "step": 44800
    },
    {
      "entropy": 5.932476949691773,
      "epoch": 3.4859754911495817,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00038121578122904496,
      "loss": 4.983,
      "mean_token_accuracy": 0.21454715877771377,
      "num_tokens": 77734281.0,
      "step": 44805
    },
    {
      "entropy": 5.820304489135742,
      "epoch": 3.4863645205212994,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003811913474423265,
      "loss": 4.9114,
      "mean_token_accuracy": 0.22208681404590608,
      "num_tokens": 77742836.0,
      "step": 44810
    },
    {
      "entropy": 5.883713817596435,
      "epoch": 3.486753549893017,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003811669120443234,
      "loss": 5.0072,
      "mean_token_accuracy": 0.21091004610061645,
      "num_tokens": 77751495.0,
      "step": 44815
    },
    {
      "entropy": 5.940904998779297,
      "epoch": 3.4871425792647344,
      "grad_norm": 1.25,
      "learning_rate": 0.0003811424750354063,
      "loss": 5.0609,
      "mean_token_accuracy": 0.20464243739843369,
      "num_tokens": 77760705.0,
      "step": 44820
    },
    {
      "entropy": 5.885308408737183,
      "epoch": 3.487531608636452,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003811180364159462,
      "loss": 4.8928,
      "mean_token_accuracy": 0.2120475649833679,
      "num_tokens": 77769383.0,
      "step": 44825
    },
    {
      "entropy": 5.859677791595459,
      "epoch": 3.4879206380081698,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038109359618631374,
      "loss": 4.9854,
      "mean_token_accuracy": 0.20647010952234268,
      "num_tokens": 77778338.0,
      "step": 44830
    },
    {
      "entropy": 5.794975614547729,
      "epoch": 3.488309667379887,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038106915434687983,
      "loss": 4.9201,
      "mean_token_accuracy": 0.21771511137485505,
      "num_tokens": 77786585.0,
      "step": 44835
    },
    {
      "entropy": 5.90019965171814,
      "epoch": 3.4886986967516047,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003810447108980153,
      "loss": 4.9992,
      "mean_token_accuracy": 0.2086635485291481,
      "num_tokens": 77794469.0,
      "step": 44840
    },
    {
      "entropy": 5.775630187988281,
      "epoch": 3.4890877261233224,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003810202658400911,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20990603119134904,
      "num_tokens": 77803026.0,
      "step": 44845
    },
    {
      "entropy": 5.92583122253418,
      "epoch": 3.4894767554950397,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003809958191734781,
      "loss": 5.0523,
      "mean_token_accuracy": 0.2005174219608307,
      "num_tokens": 77811581.0,
      "step": 44850
    },
    {
      "entropy": 5.912703657150269,
      "epoch": 3.4898657848667574,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038097137089854725,
      "loss": 5.0035,
      "mean_token_accuracy": 0.20876436531543732,
      "num_tokens": 77819405.0,
      "step": 44855
    },
    {
      "entropy": 5.967259788513184,
      "epoch": 3.490254814238475,
      "grad_norm": 1.375,
      "learning_rate": 0.00038094692101566955,
      "loss": 5.1288,
      "mean_token_accuracy": 0.20421157479286195,
      "num_tokens": 77828189.0,
      "step": 44860
    },
    {
      "entropy": 5.846439361572266,
      "epoch": 3.4906438436101928,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003809224695252159,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2208015576004982,
      "num_tokens": 77837377.0,
      "step": 44865
    },
    {
      "entropy": 5.833496713638306,
      "epoch": 3.49103287298191,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038089801642755746,
      "loss": 4.9918,
      "mean_token_accuracy": 0.20895759016275406,
      "num_tokens": 77846155.0,
      "step": 44870
    },
    {
      "entropy": 5.852547931671142,
      "epoch": 3.4914219023536277,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038087356172306514,
      "loss": 4.9697,
      "mean_token_accuracy": 0.2119869962334633,
      "num_tokens": 77854498.0,
      "step": 44875
    },
    {
      "entropy": 5.8093791007995605,
      "epoch": 3.4918109317253454,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038084910541211,
      "loss": 4.9714,
      "mean_token_accuracy": 0.21122411489486695,
      "num_tokens": 77862859.0,
      "step": 44880
    },
    {
      "entropy": 5.915842294692993,
      "epoch": 3.4921999610970627,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00038082464749506314,
      "loss": 5.0712,
      "mean_token_accuracy": 0.21275665760040283,
      "num_tokens": 77871363.0,
      "step": 44885
    },
    {
      "entropy": 5.803472375869751,
      "epoch": 3.4925889904687804,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038080018797229576,
      "loss": 4.9128,
      "mean_token_accuracy": 0.21097746044397353,
      "num_tokens": 77880392.0,
      "step": 44890
    },
    {
      "entropy": 5.871707153320313,
      "epoch": 3.492978019840498,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003807757268441789,
      "loss": 4.9971,
      "mean_token_accuracy": 0.21479251980781555,
      "num_tokens": 77888481.0,
      "step": 44895
    },
    {
      "entropy": 5.778295421600342,
      "epoch": 3.4933670492122157,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038075126411108367,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21414764523506163,
      "num_tokens": 77896607.0,
      "step": 44900
    },
    {
      "entropy": 5.849267196655274,
      "epoch": 3.493756078583933,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038072679977338143,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20337370336055755,
      "num_tokens": 77905449.0,
      "step": 44905
    },
    {
      "entropy": 5.913154363632202,
      "epoch": 3.4941451079556507,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038070233383144324,
      "loss": 4.9107,
      "mean_token_accuracy": 0.21128833293914795,
      "num_tokens": 77914307.0,
      "step": 44910
    },
    {
      "entropy": 5.84696855545044,
      "epoch": 3.4945341373273684,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003806778662856404,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21114497780799865,
      "num_tokens": 77923023.0,
      "step": 44915
    },
    {
      "entropy": 5.834674930572509,
      "epoch": 3.4949231666990856,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003806533971363441,
      "loss": 4.9813,
      "mean_token_accuracy": 0.2066175788640976,
      "num_tokens": 77932213.0,
      "step": 44920
    },
    {
      "entropy": 5.8748985767364506,
      "epoch": 3.4953121960708033,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003806289263839257,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20919941365718842,
      "num_tokens": 77941232.0,
      "step": 44925
    },
    {
      "entropy": 5.918903636932373,
      "epoch": 3.495701225442521,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003806044540287564,
      "loss": 4.9976,
      "mean_token_accuracy": 0.20974234938621522,
      "num_tokens": 77949783.0,
      "step": 44930
    },
    {
      "entropy": 5.880999326705933,
      "epoch": 3.4960902548142383,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038057998007120776,
      "loss": 4.99,
      "mean_token_accuracy": 0.2064649760723114,
      "num_tokens": 77958330.0,
      "step": 44935
    },
    {
      "entropy": 5.828049182891846,
      "epoch": 3.496479284185956,
      "grad_norm": 1.375,
      "learning_rate": 0.00038055550451165086,
      "loss": 4.9575,
      "mean_token_accuracy": 0.2095625177025795,
      "num_tokens": 77966628.0,
      "step": 44940
    },
    {
      "entropy": 5.8363207340240475,
      "epoch": 3.4968683135576737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003805310273504572,
      "loss": 5.044,
      "mean_token_accuracy": 0.20599292516708373,
      "num_tokens": 77975823.0,
      "step": 44945
    },
    {
      "entropy": 5.9194543838500975,
      "epoch": 3.497257342929391,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003805065485879982,
      "loss": 5.0697,
      "mean_token_accuracy": 0.20149530619382858,
      "num_tokens": 77984437.0,
      "step": 44950
    },
    {
      "entropy": 5.919117164611817,
      "epoch": 3.4976463723011086,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00038048206822464514,
      "loss": 4.9149,
      "mean_token_accuracy": 0.2154546320438385,
      "num_tokens": 77992260.0,
      "step": 44955
    },
    {
      "entropy": 5.907607555389404,
      "epoch": 3.4980354016728263,
      "grad_norm": 1.53125,
      "learning_rate": 0.00038045758626076965,
      "loss": 5.0668,
      "mean_token_accuracy": 0.21097473353147506,
      "num_tokens": 78000454.0,
      "step": 44960
    },
    {
      "entropy": 5.802175760269165,
      "epoch": 3.498424431044544,
      "grad_norm": 1.375,
      "learning_rate": 0.0003804331026967432,
      "loss": 4.9139,
      "mean_token_accuracy": 0.21668574810028077,
      "num_tokens": 78009664.0,
      "step": 44965
    },
    {
      "entropy": 5.813556051254272,
      "epoch": 3.4988134604162613,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003804086175329372,
      "loss": 4.8542,
      "mean_token_accuracy": 0.21460025161504745,
      "num_tokens": 78017422.0,
      "step": 44970
    },
    {
      "entropy": 5.8515842914581295,
      "epoch": 3.499202489787979,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003803841307697232,
      "loss": 4.9672,
      "mean_token_accuracy": 0.2082340255379677,
      "num_tokens": 78026128.0,
      "step": 44975
    },
    {
      "entropy": 5.796940326690674,
      "epoch": 3.4995915191596967,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038035964240747274,
      "loss": 4.9629,
      "mean_token_accuracy": 0.21070099025964736,
      "num_tokens": 78034030.0,
      "step": 44980
    },
    {
      "entropy": 5.821059226989746,
      "epoch": 3.499980548531414,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003803351524465574,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21264251470565795,
      "num_tokens": 78042090.0,
      "step": 44985
    },
    {
      "entropy": 5.8644993782043455,
      "epoch": 3.5003695779031316,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038031066088734883,
      "loss": 5.0061,
      "mean_token_accuracy": 0.21654389798641205,
      "num_tokens": 78050484.0,
      "step": 44990
    },
    {
      "entropy": 5.927216005325318,
      "epoch": 3.5007586072748493,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003802861677302185,
      "loss": 5.0192,
      "mean_token_accuracy": 0.20676873177289962,
      "num_tokens": 78058850.0,
      "step": 44995
    },
    {
      "entropy": 5.829816102981567,
      "epoch": 3.501147636646567,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038026167297553826,
      "loss": 4.9489,
      "mean_token_accuracy": 0.21237030029296874,
      "num_tokens": 78068088.0,
      "step": 45000
    },
    {
      "epoch": 3.501147636646567,
      "eval_entropy": 5.626769499094929,
      "eval_loss": 5.066978454589844,
      "eval_mean_token_accuracy": 0.21624493246702647,
      "eval_num_tokens": 78068088.0,
      "eval_runtime": 21.8291,
      "eval_samples_per_second": 1903.79,
      "eval_steps_per_second": 237.985,
      "step": 45000
    },
    {
      "entropy": 5.932721376419067,
      "epoch": 3.5015366660182843,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038023717662367956,
      "loss": 5.0747,
      "mean_token_accuracy": 0.20003687590360641,
      "num_tokens": 78076894.0,
      "step": 45005
    },
    {
      "entropy": 5.869003057479858,
      "epoch": 3.501925695390002,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003802126786750143,
      "loss": 4.9043,
      "mean_token_accuracy": 0.22397035658359526,
      "num_tokens": 78085267.0,
      "step": 45010
    },
    {
      "entropy": 5.817862606048584,
      "epoch": 3.5023147247617192,
      "grad_norm": 1.5625,
      "learning_rate": 0.00038018817912991413,
      "loss": 4.9629,
      "mean_token_accuracy": 0.21774177849292756,
      "num_tokens": 78093702.0,
      "step": 45015
    },
    {
      "entropy": 5.88811182975769,
      "epoch": 3.502703754133437,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038016367798875074,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21332552134990693,
      "num_tokens": 78102816.0,
      "step": 45020
    },
    {
      "entropy": 5.929887914657593,
      "epoch": 3.5030927835051546,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003801391752518959,
      "loss": 5.0668,
      "mean_token_accuracy": 0.1998712107539177,
      "num_tokens": 78110918.0,
      "step": 45025
    },
    {
      "entropy": 5.822823858261108,
      "epoch": 3.5034818128768723,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00038011467091972134,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20817546248435975,
      "num_tokens": 78120344.0,
      "step": 45030
    },
    {
      "entropy": 5.971571969985962,
      "epoch": 3.50387084224859,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003800901649925991,
      "loss": 5.1052,
      "mean_token_accuracy": 0.19882993400096893,
      "num_tokens": 78128836.0,
      "step": 45035
    },
    {
      "entropy": 5.927473306655884,
      "epoch": 3.5042598716203073,
      "grad_norm": 1.421875,
      "learning_rate": 0.00038006565747090076,
      "loss": 5.0189,
      "mean_token_accuracy": 0.20979005098342896,
      "num_tokens": 78138487.0,
      "step": 45040
    },
    {
      "entropy": 5.856084394454956,
      "epoch": 3.504648900992025,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038004114835499833,
      "loss": 4.901,
      "mean_token_accuracy": 0.21462844908237458,
      "num_tokens": 78147851.0,
      "step": 45045
    },
    {
      "entropy": 5.88827223777771,
      "epoch": 3.5050379303637422,
      "grad_norm": 1.421875,
      "learning_rate": 0.00038001663764526374,
      "loss": 4.9873,
      "mean_token_accuracy": 0.20696263164281845,
      "num_tokens": 78155748.0,
      "step": 45050
    },
    {
      "entropy": 5.93864254951477,
      "epoch": 3.50542695973546,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037999212534206876,
      "loss": 5.0868,
      "mean_token_accuracy": 0.20654332339763642,
      "num_tokens": 78165050.0,
      "step": 45055
    },
    {
      "entropy": 5.926043510437012,
      "epoch": 3.5058159891071776,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003799676114457853,
      "loss": 5.0449,
      "mean_token_accuracy": 0.20292008966207503,
      "num_tokens": 78173550.0,
      "step": 45060
    },
    {
      "entropy": 5.846756076812744,
      "epoch": 3.5062050184788953,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037994309595678553,
      "loss": 5.0245,
      "mean_token_accuracy": 0.2092999204993248,
      "num_tokens": 78182909.0,
      "step": 45065
    },
    {
      "entropy": 5.884186315536499,
      "epoch": 3.5065940478506126,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003799185788754413,
      "loss": 4.9256,
      "mean_token_accuracy": 0.21486144363880158,
      "num_tokens": 78190814.0,
      "step": 45070
    },
    {
      "entropy": 5.829663944244385,
      "epoch": 3.5069830772223303,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003798940602021247,
      "loss": 4.9574,
      "mean_token_accuracy": 0.2149548202753067,
      "num_tokens": 78199634.0,
      "step": 45075
    },
    {
      "entropy": 5.823628044128418,
      "epoch": 3.507372106594048,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003798695399372076,
      "loss": 4.9367,
      "mean_token_accuracy": 0.21718997210264207,
      "num_tokens": 78208650.0,
      "step": 45080
    },
    {
      "entropy": 5.925009059906006,
      "epoch": 3.507761135965765,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003798450180810621,
      "loss": 5.0294,
      "mean_token_accuracy": 0.2071436658501625,
      "num_tokens": 78217451.0,
      "step": 45085
    },
    {
      "entropy": 5.9026854038238525,
      "epoch": 3.508150165337483,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037982049463406044,
      "loss": 5.0397,
      "mean_token_accuracy": 0.20616032779216767,
      "num_tokens": 78226985.0,
      "step": 45090
    },
    {
      "entropy": 5.849848413467408,
      "epoch": 3.5085391947092006,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003797959695965745,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21660025417804718,
      "num_tokens": 78235318.0,
      "step": 45095
    },
    {
      "entropy": 5.815673971176148,
      "epoch": 3.5089282240809183,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037977144296897665,
      "loss": 4.979,
      "mean_token_accuracy": 0.21527139246463775,
      "num_tokens": 78243582.0,
      "step": 45100
    },
    {
      "entropy": 5.7990875244140625,
      "epoch": 3.5093172534526356,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003797469147516388,
      "loss": 4.9171,
      "mean_token_accuracy": 0.21587480902671813,
      "num_tokens": 78252638.0,
      "step": 45105
    },
    {
      "entropy": 5.911598873138428,
      "epoch": 3.5097062828243533,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037972238494493333,
      "loss": 5.0188,
      "mean_token_accuracy": 0.2049779549241066,
      "num_tokens": 78261442.0,
      "step": 45110
    },
    {
      "entropy": 5.832846593856812,
      "epoch": 3.510095312196071,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003796978535492323,
      "loss": 4.9046,
      "mean_token_accuracy": 0.2232723578810692,
      "num_tokens": 78270282.0,
      "step": 45115
    },
    {
      "entropy": 5.8629225730896,
      "epoch": 3.510484341567788,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000379673320564908,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20384276956319808,
      "num_tokens": 78278906.0,
      "step": 45120
    },
    {
      "entropy": 5.885653972625732,
      "epoch": 3.510873370939506,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037964878599233264,
      "loss": 4.9873,
      "mean_token_accuracy": 0.2142725721001625,
      "num_tokens": 78286834.0,
      "step": 45125
    },
    {
      "entropy": 5.835375261306763,
      "epoch": 3.5112624003112236,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037962424983187856,
      "loss": 5.0482,
      "mean_token_accuracy": 0.2007294178009033,
      "num_tokens": 78295841.0,
      "step": 45130
    },
    {
      "entropy": 5.894398164749146,
      "epoch": 3.5116514296829413,
      "grad_norm": 1.421875,
      "learning_rate": 0.000379599712083918,
      "loss": 4.9814,
      "mean_token_accuracy": 0.21155588775873185,
      "num_tokens": 78304020.0,
      "step": 45135
    },
    {
      "entropy": 5.902297782897949,
      "epoch": 3.5120404590546586,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003795751727488233,
      "loss": 4.9315,
      "mean_token_accuracy": 0.2119233250617981,
      "num_tokens": 78312447.0,
      "step": 45140
    },
    {
      "entropy": 5.800845336914063,
      "epoch": 3.5124294884263763,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037955063182696676,
      "loss": 4.9219,
      "mean_token_accuracy": 0.2202686548233032,
      "num_tokens": 78321321.0,
      "step": 45145
    },
    {
      "entropy": 5.878175354003906,
      "epoch": 3.5128185177980935,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037952608931872086,
      "loss": 4.9838,
      "mean_token_accuracy": 0.20980445742607118,
      "num_tokens": 78330413.0,
      "step": 45150
    },
    {
      "entropy": 5.877806091308594,
      "epoch": 3.513207547169811,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037950154522445784,
      "loss": 4.8946,
      "mean_token_accuracy": 0.2179975315928459,
      "num_tokens": 78338228.0,
      "step": 45155
    },
    {
      "entropy": 5.847499847412109,
      "epoch": 3.513596576541529,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003794769995445502,
      "loss": 5.0257,
      "mean_token_accuracy": 0.20664810240268708,
      "num_tokens": 78346702.0,
      "step": 45160
    },
    {
      "entropy": 5.804989624023437,
      "epoch": 3.5139856059132466,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003794524522793705,
      "loss": 4.9988,
      "mean_token_accuracy": 0.2157856285572052,
      "num_tokens": 78355289.0,
      "step": 45165
    },
    {
      "entropy": 5.8072878360748295,
      "epoch": 3.514374635284964,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037942790342929106,
      "loss": 4.9127,
      "mean_token_accuracy": 0.21594730466604234,
      "num_tokens": 78364884.0,
      "step": 45170
    },
    {
      "entropy": 5.8253847599029545,
      "epoch": 3.5147636646566816,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037940335299468437,
      "loss": 4.9234,
      "mean_token_accuracy": 0.217653089761734,
      "num_tokens": 78373486.0,
      "step": 45175
    },
    {
      "entropy": 5.853621482849121,
      "epoch": 3.5151526940283992,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037937880097592294,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21658690571784972,
      "num_tokens": 78382169.0,
      "step": 45180
    },
    {
      "entropy": 5.8187802791595455,
      "epoch": 3.5155417234001165,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037935424737337936,
      "loss": 4.9875,
      "mean_token_accuracy": 0.21169629245996474,
      "num_tokens": 78391127.0,
      "step": 45185
    },
    {
      "entropy": 5.821754407882691,
      "epoch": 3.515930752771834,
      "grad_norm": 1.25,
      "learning_rate": 0.00037932969218742616,
      "loss": 4.8746,
      "mean_token_accuracy": 0.21045132577419282,
      "num_tokens": 78399773.0,
      "step": 45190
    },
    {
      "entropy": 5.825250148773193,
      "epoch": 3.516319782143552,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037930513541843593,
      "loss": 4.9783,
      "mean_token_accuracy": 0.20947145968675612,
      "num_tokens": 78409403.0,
      "step": 45195
    },
    {
      "entropy": 5.840211248397827,
      "epoch": 3.5167088115152696,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037928057706678136,
      "loss": 4.9572,
      "mean_token_accuracy": 0.21833537817001342,
      "num_tokens": 78418284.0,
      "step": 45200
    },
    {
      "entropy": 5.879475498199463,
      "epoch": 3.517097840886987,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037925601713283504,
      "loss": 4.9617,
      "mean_token_accuracy": 0.21302390992641448,
      "num_tokens": 78426570.0,
      "step": 45205
    },
    {
      "entropy": 5.87636775970459,
      "epoch": 3.5174868702587045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003792314556169695,
      "loss": 4.992,
      "mean_token_accuracy": 0.21208970695734025,
      "num_tokens": 78435816.0,
      "step": 45210
    },
    {
      "entropy": 5.871246194839477,
      "epoch": 3.5178758996304222,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003792068925195576,
      "loss": 4.9547,
      "mean_token_accuracy": 0.22153990268707274,
      "num_tokens": 78444002.0,
      "step": 45215
    },
    {
      "entropy": 5.826976299285889,
      "epoch": 3.5182649290021395,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003791823278409719,
      "loss": 5.005,
      "mean_token_accuracy": 0.21417439728975296,
      "num_tokens": 78452175.0,
      "step": 45220
    },
    {
      "entropy": 5.91998701095581,
      "epoch": 3.518653958373857,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037915776158158515,
      "loss": 5.0136,
      "mean_token_accuracy": 0.20154198706150056,
      "num_tokens": 78460506.0,
      "step": 45225
    },
    {
      "entropy": 5.804496097564697,
      "epoch": 3.519042987745575,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003791331937417703,
      "loss": 4.994,
      "mean_token_accuracy": 0.20605075657367705,
      "num_tokens": 78469771.0,
      "step": 45230
    },
    {
      "entropy": 5.872729778289795,
      "epoch": 3.5194320171172926,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037910862432189986,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20851699262857437,
      "num_tokens": 78478666.0,
      "step": 45235
    },
    {
      "entropy": 5.822877597808838,
      "epoch": 3.51982104648901,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003790840533223467,
      "loss": 4.9639,
      "mean_token_accuracy": 0.21568614095449448,
      "num_tokens": 78487534.0,
      "step": 45240
    },
    {
      "entropy": 5.7737963676452635,
      "epoch": 3.5202100758607275,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037905948074348387,
      "loss": 4.9243,
      "mean_token_accuracy": 0.2086750715970993,
      "num_tokens": 78496255.0,
      "step": 45245
    },
    {
      "entropy": 5.911154079437256,
      "epoch": 3.520599105232445,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000379034906585684,
      "loss": 4.988,
      "mean_token_accuracy": 0.21195125132799147,
      "num_tokens": 78504045.0,
      "step": 45250
    },
    {
      "entropy": 5.849690198898315,
      "epoch": 3.5209881346041625,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037901033084932,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21154258251190186,
      "num_tokens": 78512696.0,
      "step": 45255
    },
    {
      "entropy": 5.838649940490723,
      "epoch": 3.52137716397588,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003789857535347648,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20899676978588105,
      "num_tokens": 78521378.0,
      "step": 45260
    },
    {
      "entropy": 5.839627265930176,
      "epoch": 3.521766193347598,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003789611746423912,
      "loss": 5.0142,
      "mean_token_accuracy": 0.20644188970327376,
      "num_tokens": 78529975.0,
      "step": 45265
    },
    {
      "entropy": 5.89998607635498,
      "epoch": 3.5221552227193156,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037893659417257233,
      "loss": 4.9489,
      "mean_token_accuracy": 0.20823965966701508,
      "num_tokens": 78538486.0,
      "step": 45270
    },
    {
      "entropy": 5.783394622802734,
      "epoch": 3.522544252091033,
      "grad_norm": 1.46875,
      "learning_rate": 0.00037891201212568107,
      "loss": 4.9557,
      "mean_token_accuracy": 0.21207451522350312,
      "num_tokens": 78546761.0,
      "step": 45275
    },
    {
      "entropy": 5.801366090774536,
      "epoch": 3.5229332814627505,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037888742850209037,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21571864187717438,
      "num_tokens": 78555361.0,
      "step": 45280
    },
    {
      "entropy": 5.825440120697022,
      "epoch": 3.523322310834468,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003788628433021733,
      "loss": 4.9634,
      "mean_token_accuracy": 0.21500884741544724,
      "num_tokens": 78564037.0,
      "step": 45285
    },
    {
      "entropy": 5.86397910118103,
      "epoch": 3.5237113402061855,
      "grad_norm": 1.34375,
      "learning_rate": 0.000378838256526303,
      "loss": 5.0592,
      "mean_token_accuracy": 0.1980877235531807,
      "num_tokens": 78572576.0,
      "step": 45290
    },
    {
      "entropy": 5.963622808456421,
      "epoch": 3.524100369577903,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003788136681748523,
      "loss": 5.1019,
      "mean_token_accuracy": 0.2039520338177681,
      "num_tokens": 78581400.0,
      "step": 45295
    },
    {
      "entropy": 5.77984299659729,
      "epoch": 3.524489398949621,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003787890782481945,
      "loss": 4.8295,
      "mean_token_accuracy": 0.22604072391986846,
      "num_tokens": 78589633.0,
      "step": 45300
    },
    {
      "entropy": 5.782489204406739,
      "epoch": 3.524878428321338,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003787644867467026,
      "loss": 4.9667,
      "mean_token_accuracy": 0.2139367714524269,
      "num_tokens": 78598807.0,
      "step": 45305
    },
    {
      "entropy": 5.8394664287567135,
      "epoch": 3.525267457693056,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037873989367074977,
      "loss": 5.0033,
      "mean_token_accuracy": 0.20914938449859619,
      "num_tokens": 78607470.0,
      "step": 45310
    },
    {
      "entropy": 5.936884593963623,
      "epoch": 3.5256564870647735,
      "grad_norm": 1.484375,
      "learning_rate": 0.00037871529902070924,
      "loss": 5.0529,
      "mean_token_accuracy": 0.20796950459480285,
      "num_tokens": 78615907.0,
      "step": 45315
    },
    {
      "entropy": 5.90245475769043,
      "epoch": 3.526045516436491,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037869070279695404,
      "loss": 5.1249,
      "mean_token_accuracy": 0.2019887700676918,
      "num_tokens": 78625009.0,
      "step": 45320
    },
    {
      "entropy": 5.8416054248809814,
      "epoch": 3.5264345458082085,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037866610499985755,
      "loss": 4.9607,
      "mean_token_accuracy": 0.21765768826007842,
      "num_tokens": 78633854.0,
      "step": 45325
    },
    {
      "entropy": 5.834599304199219,
      "epoch": 3.526823575179926,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003786415056297929,
      "loss": 4.9581,
      "mean_token_accuracy": 0.2087261900305748,
      "num_tokens": 78641957.0,
      "step": 45330
    },
    {
      "entropy": 5.891305923461914,
      "epoch": 3.527212604551644,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003786169046871333,
      "loss": 5.0325,
      "mean_token_accuracy": 0.2005322515964508,
      "num_tokens": 78650984.0,
      "step": 45335
    },
    {
      "entropy": 5.850249338150024,
      "epoch": 3.527601633923361,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003785923021722521,
      "loss": 4.9397,
      "mean_token_accuracy": 0.20504023730754853,
      "num_tokens": 78659216.0,
      "step": 45340
    },
    {
      "entropy": 5.862275791168213,
      "epoch": 3.527990663295079,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037856769808552267,
      "loss": 4.9691,
      "mean_token_accuracy": 0.2144316166639328,
      "num_tokens": 78668099.0,
      "step": 45345
    },
    {
      "entropy": 5.804915237426758,
      "epoch": 3.528379692666796,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003785430924273182,
      "loss": 4.8929,
      "mean_token_accuracy": 0.21531765162944794,
      "num_tokens": 78677022.0,
      "step": 45350
    },
    {
      "entropy": 5.906276798248291,
      "epoch": 3.5287687220385138,
      "grad_norm": 1.40625,
      "learning_rate": 0.00037851848519801214,
      "loss": 5.0762,
      "mean_token_accuracy": 0.20642807483673095,
      "num_tokens": 78685942.0,
      "step": 45355
    },
    {
      "entropy": 5.87044939994812,
      "epoch": 3.5291577514102315,
      "grad_norm": 1.25,
      "learning_rate": 0.0003784938763979778,
      "loss": 4.9622,
      "mean_token_accuracy": 0.2147611767053604,
      "num_tokens": 78694522.0,
      "step": 45360
    },
    {
      "entropy": 5.835910081863403,
      "epoch": 3.529546780781949,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003784692660275886,
      "loss": 4.9012,
      "mean_token_accuracy": 0.21867570877075196,
      "num_tokens": 78704053.0,
      "step": 45365
    },
    {
      "entropy": 5.88354606628418,
      "epoch": 3.529935810153667,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003784446540872181,
      "loss": 4.9192,
      "mean_token_accuracy": 0.2147444888949394,
      "num_tokens": 78712721.0,
      "step": 45370
    },
    {
      "entropy": 5.826020336151123,
      "epoch": 3.530324839525384,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037842004057723957,
      "loss": 4.9652,
      "mean_token_accuracy": 0.20936896055936813,
      "num_tokens": 78722087.0,
      "step": 45375
    },
    {
      "entropy": 5.853988265991211,
      "epoch": 3.530713868897102,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003783954254980265,
      "loss": 5.0198,
      "mean_token_accuracy": 0.20846703946590422,
      "num_tokens": 78730140.0,
      "step": 45380
    },
    {
      "entropy": 5.930027103424072,
      "epoch": 3.531102898268819,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003783708088499524,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20584528744220734,
      "num_tokens": 78738353.0,
      "step": 45385
    },
    {
      "entropy": 5.886236190795898,
      "epoch": 3.5314919276405368,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037834619063339085,
      "loss": 5.0156,
      "mean_token_accuracy": 0.21124156266450883,
      "num_tokens": 78747163.0,
      "step": 45390
    },
    {
      "entropy": 5.802527332305909,
      "epoch": 3.5318809570122545,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003783215708487153,
      "loss": 5.0105,
      "mean_token_accuracy": 0.20071533769369126,
      "num_tokens": 78756274.0,
      "step": 45395
    },
    {
      "entropy": 5.8513579845428465,
      "epoch": 3.532269986383972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003782969494962994,
      "loss": 4.9735,
      "mean_token_accuracy": 0.21177105456590653,
      "num_tokens": 78765553.0,
      "step": 45400
    },
    {
      "entropy": 5.87096848487854,
      "epoch": 3.5326590157556894,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003782723265765167,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2158966615796089,
      "num_tokens": 78774275.0,
      "step": 45405
    },
    {
      "entropy": 5.889529371261597,
      "epoch": 3.533048045127407,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037824770208974085,
      "loss": 5.0182,
      "mean_token_accuracy": 0.20445874482393264,
      "num_tokens": 78783339.0,
      "step": 45410
    },
    {
      "entropy": 5.7906488418579105,
      "epoch": 3.533437074499125,
      "grad_norm": 1.25,
      "learning_rate": 0.0003782230760363454,
      "loss": 4.9012,
      "mean_token_accuracy": 0.217826946079731,
      "num_tokens": 78792247.0,
      "step": 45415
    },
    {
      "entropy": 5.851281929016113,
      "epoch": 3.533826103870842,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037819844841670416,
      "loss": 5.0328,
      "mean_token_accuracy": 0.20981960594654084,
      "num_tokens": 78800331.0,
      "step": 45420
    },
    {
      "entropy": 5.807776546478271,
      "epoch": 3.5342151332425598,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003781738192311906,
      "loss": 4.8983,
      "mean_token_accuracy": 0.21264511495828628,
      "num_tokens": 78808984.0,
      "step": 45425
    },
    {
      "entropy": 5.855939197540283,
      "epoch": 3.5346041626142775,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037814918848017855,
      "loss": 4.9918,
      "mean_token_accuracy": 0.20888191908597947,
      "num_tokens": 78817485.0,
      "step": 45430
    },
    {
      "entropy": 5.877398872375489,
      "epoch": 3.534993191985995,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003781245561640418,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20741984248161316,
      "num_tokens": 78826134.0,
      "step": 45435
    },
    {
      "entropy": 5.841350841522217,
      "epoch": 3.5353822213577124,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003780999222831541,
      "loss": 4.9922,
      "mean_token_accuracy": 0.21840549111366273,
      "num_tokens": 78834569.0,
      "step": 45440
    },
    {
      "entropy": 5.91485366821289,
      "epoch": 3.53577125072943,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003780752868378892,
      "loss": 5.0287,
      "mean_token_accuracy": 0.21029309779405594,
      "num_tokens": 78843395.0,
      "step": 45445
    },
    {
      "entropy": 5.903423643112182,
      "epoch": 3.536160280101148,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00037805064982862076,
      "loss": 4.9826,
      "mean_token_accuracy": 0.20569756478071213,
      "num_tokens": 78851800.0,
      "step": 45450
    },
    {
      "entropy": 5.866128063201904,
      "epoch": 3.536549309472865,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003780260112557228,
      "loss": 5.0591,
      "mean_token_accuracy": 0.20324162840843202,
      "num_tokens": 78859926.0,
      "step": 45455
    },
    {
      "entropy": 5.871235179901123,
      "epoch": 3.5369383388445828,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037800137111956904,
      "loss": 5.0113,
      "mean_token_accuracy": 0.20618481189012527,
      "num_tokens": 78868654.0,
      "step": 45460
    },
    {
      "entropy": 5.910176372528076,
      "epoch": 3.5373273682163004,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037797672942053344,
      "loss": 4.9873,
      "mean_token_accuracy": 0.2132035657763481,
      "num_tokens": 78877545.0,
      "step": 45465
    },
    {
      "entropy": 5.877975988388061,
      "epoch": 3.537716397588018,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037795208615898997,
      "loss": 5.0361,
      "mean_token_accuracy": 0.20163301527500152,
      "num_tokens": 78886794.0,
      "step": 45470
    },
    {
      "entropy": 5.880583810806274,
      "epoch": 3.5381054269597354,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003779274413353124,
      "loss": 5.02,
      "mean_token_accuracy": 0.20416702032089235,
      "num_tokens": 78895617.0,
      "step": 45475
    },
    {
      "entropy": 5.934981203079223,
      "epoch": 3.538494456331453,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037790279494987475,
      "loss": 5.0685,
      "mean_token_accuracy": 0.21209078878164292,
      "num_tokens": 78904300.0,
      "step": 45480
    },
    {
      "entropy": 5.888391971588135,
      "epoch": 3.5388834857031704,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037787814700305096,
      "loss": 4.9937,
      "mean_token_accuracy": 0.21303920000791549,
      "num_tokens": 78913470.0,
      "step": 45485
    },
    {
      "entropy": 5.871713066101075,
      "epoch": 3.539272515074888,
      "grad_norm": 1.203125,
      "learning_rate": 0.000377853497495215,
      "loss": 4.9907,
      "mean_token_accuracy": 0.20933856815099716,
      "num_tokens": 78921923.0,
      "step": 45490
    },
    {
      "entropy": 5.800688886642456,
      "epoch": 3.5396615444466057,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037782884642674085,
      "loss": 4.9148,
      "mean_token_accuracy": 0.21321197599172592,
      "num_tokens": 78930218.0,
      "step": 45495
    },
    {
      "entropy": 5.813690423965454,
      "epoch": 3.5400505738183234,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003778041937980028,
      "loss": 4.9546,
      "mean_token_accuracy": 0.2156895264983177,
      "num_tokens": 78938272.0,
      "step": 45500
    },
    {
      "entropy": 5.758527088165283,
      "epoch": 3.5404396031900407,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037777953960937467,
      "loss": 4.8974,
      "mean_token_accuracy": 0.22340509444475173,
      "num_tokens": 78946859.0,
      "step": 45505
    },
    {
      "entropy": 5.832810926437378,
      "epoch": 3.5408286325617584,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00037775488386123057,
      "loss": 5.0599,
      "mean_token_accuracy": 0.20361458212137223,
      "num_tokens": 78957029.0,
      "step": 45510
    },
    {
      "entropy": 5.964425086975098,
      "epoch": 3.541217661933476,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037773022655394466,
      "loss": 5.0582,
      "mean_token_accuracy": 0.20582681596279145,
      "num_tokens": 78965943.0,
      "step": 45515
    },
    {
      "entropy": 5.832622718811035,
      "epoch": 3.5416066913051933,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037770556768789115,
      "loss": 4.8393,
      "mean_token_accuracy": 0.21669335663318634,
      "num_tokens": 78974713.0,
      "step": 45520
    },
    {
      "entropy": 5.877196550369263,
      "epoch": 3.541995720676911,
      "grad_norm": 1.484375,
      "learning_rate": 0.00037768090726344414,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20935473144054412,
      "num_tokens": 78983489.0,
      "step": 45525
    },
    {
      "entropy": 5.904327821731568,
      "epoch": 3.5423847500486287,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003776562452809777,
      "loss": 5.0342,
      "mean_token_accuracy": 0.20669298619031906,
      "num_tokens": 78992355.0,
      "step": 45530
    },
    {
      "entropy": 5.863565254211426,
      "epoch": 3.5427737794203464,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00037763158174086616,
      "loss": 4.9824,
      "mean_token_accuracy": 0.21712840348482132,
      "num_tokens": 79000505.0,
      "step": 45535
    },
    {
      "entropy": 5.849595737457276,
      "epoch": 3.5431628087920637,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037760691664348367,
      "loss": 5.0163,
      "mean_token_accuracy": 0.20989047586917878,
      "num_tokens": 79008697.0,
      "step": 45540
    },
    {
      "entropy": 5.905746793746948,
      "epoch": 3.5435518381637814,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003775822499892046,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21464051604270934,
      "num_tokens": 79017636.0,
      "step": 45545
    },
    {
      "entropy": 5.795956754684449,
      "epoch": 3.543940867535499,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003775575817784032,
      "loss": 4.8851,
      "mean_token_accuracy": 0.21160683631896973,
      "num_tokens": 79026185.0,
      "step": 45550
    },
    {
      "entropy": 5.925047445297241,
      "epoch": 3.5443298969072163,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037753291201145375,
      "loss": 5.0153,
      "mean_token_accuracy": 0.20147063881158828,
      "num_tokens": 79034076.0,
      "step": 45555
    },
    {
      "entropy": 5.89366397857666,
      "epoch": 3.544718926278934,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003775082406887305,
      "loss": 4.9861,
      "mean_token_accuracy": 0.19643015414476395,
      "num_tokens": 79042519.0,
      "step": 45560
    },
    {
      "entropy": 5.884097766876221,
      "epoch": 3.5451079556506517,
      "grad_norm": 1.515625,
      "learning_rate": 0.00037748356781060785,
      "loss": 5.0406,
      "mean_token_accuracy": 0.2060250833630562,
      "num_tokens": 79050316.0,
      "step": 45565
    },
    {
      "entropy": 5.84866156578064,
      "epoch": 3.5454969850223694,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003774588933774602,
      "loss": 5.0111,
      "mean_token_accuracy": 0.20687042623758317,
      "num_tokens": 79059378.0,
      "step": 45570
    },
    {
      "entropy": 5.833556175231934,
      "epoch": 3.5458860143940867,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003774342173896619,
      "loss": 4.9674,
      "mean_token_accuracy": 0.2077896162867546,
      "num_tokens": 79068241.0,
      "step": 45575
    },
    {
      "entropy": 5.928825378417969,
      "epoch": 3.5462750437658044,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037740953984758736,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20532742589712144,
      "num_tokens": 79077033.0,
      "step": 45580
    },
    {
      "entropy": 5.8090461730957035,
      "epoch": 3.5466640731375216,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003773848607516111,
      "loss": 4.9549,
      "mean_token_accuracy": 0.2114281475543976,
      "num_tokens": 79086927.0,
      "step": 45585
    },
    {
      "entropy": 5.9639427185058596,
      "epoch": 3.5470531025092393,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003773601801021075,
      "loss": 5.1813,
      "mean_token_accuracy": 0.19972584396600723,
      "num_tokens": 79095593.0,
      "step": 45590
    },
    {
      "entropy": 5.911545944213867,
      "epoch": 3.547442131880957,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003773354978994512,
      "loss": 4.9697,
      "mean_token_accuracy": 0.21421399861574172,
      "num_tokens": 79104173.0,
      "step": 45595
    },
    {
      "entropy": 5.925880002975464,
      "epoch": 3.5478311612526747,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037731081414401653,
      "loss": 5.0099,
      "mean_token_accuracy": 0.2088039994239807,
      "num_tokens": 79112728.0,
      "step": 45600
    },
    {
      "entropy": 5.90098614692688,
      "epoch": 3.548220190624392,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037728612883617806,
      "loss": 5.0307,
      "mean_token_accuracy": 0.20740415900945663,
      "num_tokens": 79121488.0,
      "step": 45605
    },
    {
      "entropy": 5.878537321090699,
      "epoch": 3.5486092199961097,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003772614419763104,
      "loss": 4.9203,
      "mean_token_accuracy": 0.21487898975610734,
      "num_tokens": 79130411.0,
      "step": 45610
    },
    {
      "entropy": 5.8762133598327635,
      "epoch": 3.5489982493678274,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003772367535647882,
      "loss": 4.926,
      "mean_token_accuracy": 0.2152986854314804,
      "num_tokens": 79139943.0,
      "step": 45615
    },
    {
      "entropy": 5.846447610855103,
      "epoch": 3.5493872787395446,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00037721206360198596,
      "loss": 4.858,
      "mean_token_accuracy": 0.21627771556377412,
      "num_tokens": 79148478.0,
      "step": 45620
    },
    {
      "entropy": 5.826213407516479,
      "epoch": 3.5497763081112623,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003771873720882783,
      "loss": 4.9745,
      "mean_token_accuracy": 0.21221626549959183,
      "num_tokens": 79156390.0,
      "step": 45625
    },
    {
      "entropy": 5.800556707382202,
      "epoch": 3.55016533748298,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037716267902403996,
      "loss": 4.9465,
      "mean_token_accuracy": 0.21528827100992204,
      "num_tokens": 79165603.0,
      "step": 45630
    },
    {
      "entropy": 5.860878372192383,
      "epoch": 3.5505543668546977,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003771379844096456,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20813124775886535,
      "num_tokens": 79174529.0,
      "step": 45635
    },
    {
      "entropy": 5.857524490356445,
      "epoch": 3.550943396226415,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003771132882454698,
      "loss": 4.9351,
      "mean_token_accuracy": 0.2114263191819191,
      "num_tokens": 79183127.0,
      "step": 45640
    },
    {
      "entropy": 5.85979700088501,
      "epoch": 3.5513324255981327,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003770885905318874,
      "loss": 4.961,
      "mean_token_accuracy": 0.2104208290576935,
      "num_tokens": 79192057.0,
      "step": 45645
    },
    {
      "entropy": 5.875337743759156,
      "epoch": 3.5517214549698504,
      "grad_norm": 1.5,
      "learning_rate": 0.0003770638912692732,
      "loss": 4.9579,
      "mean_token_accuracy": 0.21083979159593583,
      "num_tokens": 79200354.0,
      "step": 45650
    },
    {
      "entropy": 5.859758472442627,
      "epoch": 3.5521104843415676,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003770391904580019,
      "loss": 4.9819,
      "mean_token_accuracy": 0.2142678201198578,
      "num_tokens": 79209456.0,
      "step": 45655
    },
    {
      "entropy": 5.8424718379974365,
      "epoch": 3.5524995137132853,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003770144880984482,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20960631668567659,
      "num_tokens": 79217737.0,
      "step": 45660
    },
    {
      "entropy": 5.941015529632568,
      "epoch": 3.552888543085003,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037698978419098704,
      "loss": 5.0676,
      "mean_token_accuracy": 0.20829030871391296,
      "num_tokens": 79225501.0,
      "step": 45665
    },
    {
      "entropy": 5.860081958770752,
      "epoch": 3.5532775724567207,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037696507873599336,
      "loss": 5.0178,
      "mean_token_accuracy": 0.21263686269521714,
      "num_tokens": 79235518.0,
      "step": 45670
    },
    {
      "entropy": 5.822968482971191,
      "epoch": 3.553666601828438,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037694037173384177,
      "loss": 4.9859,
      "mean_token_accuracy": 0.21790971457958222,
      "num_tokens": 79244876.0,
      "step": 45675
    },
    {
      "entropy": 5.873045253753662,
      "epoch": 3.5540556312001557,
      "grad_norm": 1.375,
      "learning_rate": 0.0003769156631849074,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20969051122665405,
      "num_tokens": 79252720.0,
      "step": 45680
    },
    {
      "entropy": 5.862482786178589,
      "epoch": 3.554444660571873,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000376890953089565,
      "loss": 4.9536,
      "mean_token_accuracy": 0.21131736934185028,
      "num_tokens": 79260891.0,
      "step": 45685
    },
    {
      "entropy": 5.889886713027954,
      "epoch": 3.5548336899435906,
      "grad_norm": 1.25,
      "learning_rate": 0.0003768662414481895,
      "loss": 5.0564,
      "mean_token_accuracy": 0.21019736230373381,
      "num_tokens": 79269649.0,
      "step": 45690
    },
    {
      "entropy": 5.870569562911987,
      "epoch": 3.5552227193153083,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037684152826115595,
      "loss": 5.0073,
      "mean_token_accuracy": 0.20639795511960984,
      "num_tokens": 79278598.0,
      "step": 45695
    },
    {
      "entropy": 5.8846289157867435,
      "epoch": 3.555611748687026,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003768168135288394,
      "loss": 4.969,
      "mean_token_accuracy": 0.2163334757089615,
      "num_tokens": 79287457.0,
      "step": 45700
    },
    {
      "entropy": 5.867749404907227,
      "epoch": 3.5560007780587437,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003767920972516147,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22301874160766602,
      "num_tokens": 79296157.0,
      "step": 45705
    },
    {
      "entropy": 5.77487154006958,
      "epoch": 3.556389807430461,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037676737942985697,
      "loss": 4.9207,
      "mean_token_accuracy": 0.21886297762393953,
      "num_tokens": 79304784.0,
      "step": 45710
    },
    {
      "entropy": 5.844782924652099,
      "epoch": 3.5567788368021787,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037674266006394123,
      "loss": 5.0513,
      "mean_token_accuracy": 0.2003174051642418,
      "num_tokens": 79313442.0,
      "step": 45715
    },
    {
      "entropy": 5.900986003875732,
      "epoch": 3.557167866173896,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00037671793915424254,
      "loss": 5.0011,
      "mean_token_accuracy": 0.21074083298444748,
      "num_tokens": 79322191.0,
      "step": 45720
    },
    {
      "entropy": 5.922296619415283,
      "epoch": 3.5575568955456136,
      "grad_norm": 1.453125,
      "learning_rate": 0.000376693216701136,
      "loss": 5.0206,
      "mean_token_accuracy": 0.20228171050548555,
      "num_tokens": 79330505.0,
      "step": 45725
    },
    {
      "entropy": 5.911898279190064,
      "epoch": 3.5579459249173313,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003766684927049968,
      "loss": 5.0356,
      "mean_token_accuracy": 0.21892029941082,
      "num_tokens": 79339519.0,
      "step": 45730
    },
    {
      "entropy": 5.820607614517212,
      "epoch": 3.558334954289049,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003766437671662,
      "loss": 4.9757,
      "mean_token_accuracy": 0.20914221853017806,
      "num_tokens": 79348177.0,
      "step": 45735
    },
    {
      "entropy": 5.732741117477417,
      "epoch": 3.5587239836607663,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003766190400851209,
      "loss": 4.888,
      "mean_token_accuracy": 0.21921923905611038,
      "num_tokens": 79357243.0,
      "step": 45740
    },
    {
      "entropy": 5.893880605697632,
      "epoch": 3.559113013032484,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003765943114621345,
      "loss": 4.9873,
      "mean_token_accuracy": 0.20770905464887618,
      "num_tokens": 79365299.0,
      "step": 45745
    },
    {
      "entropy": 5.85505256652832,
      "epoch": 3.5595020424042016,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037656958129761626,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21364941149950029,
      "num_tokens": 79373567.0,
      "step": 45750
    },
    {
      "entropy": 5.797436761856079,
      "epoch": 3.559891071775919,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037654484959194126,
      "loss": 4.9996,
      "mean_token_accuracy": 0.19979981631040572,
      "num_tokens": 79383090.0,
      "step": 45755
    },
    {
      "entropy": 5.8583417415618895,
      "epoch": 3.5602801011476366,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037652011634548466,
      "loss": 4.9353,
      "mean_token_accuracy": 0.2106696382164955,
      "num_tokens": 79391522.0,
      "step": 45760
    },
    {
      "entropy": 5.8792425155639645,
      "epoch": 3.5606691305193543,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000376495381558622,
      "loss": 4.9926,
      "mean_token_accuracy": 0.21427711099386215,
      "num_tokens": 79399644.0,
      "step": 45765
    },
    {
      "entropy": 5.832421207427979,
      "epoch": 3.561058159891072,
      "grad_norm": 1.375,
      "learning_rate": 0.0003764706452317284,
      "loss": 4.9215,
      "mean_token_accuracy": 0.22437880039215088,
      "num_tokens": 79408029.0,
      "step": 45770
    },
    {
      "entropy": 5.723764133453369,
      "epoch": 3.5614471892627892,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037644590736517935,
      "loss": 4.8325,
      "mean_token_accuracy": 0.21959281712770462,
      "num_tokens": 79415917.0,
      "step": 45775
    },
    {
      "entropy": 5.804325008392334,
      "epoch": 3.561836218634507,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003764211679593501,
      "loss": 5.0225,
      "mean_token_accuracy": 0.20635680854320526,
      "num_tokens": 79425122.0,
      "step": 45780
    },
    {
      "entropy": 5.8740379333496096,
      "epoch": 3.562225248006224,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037639642701461603,
      "loss": 4.8946,
      "mean_token_accuracy": 0.21563945412635804,
      "num_tokens": 79433319.0,
      "step": 45785
    },
    {
      "entropy": 5.888499927520752,
      "epoch": 3.562614277377942,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003763716845313526,
      "loss": 4.9937,
      "mean_token_accuracy": 0.2124168172478676,
      "num_tokens": 79442027.0,
      "step": 45790
    },
    {
      "entropy": 5.851191568374634,
      "epoch": 3.5630033067496596,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003763469405099352,
      "loss": 4.9094,
      "mean_token_accuracy": 0.21054278761148454,
      "num_tokens": 79450449.0,
      "step": 45795
    },
    {
      "entropy": 5.78983588218689,
      "epoch": 3.5633923361213773,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003763221949507392,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21200930923223496,
      "num_tokens": 79459416.0,
      "step": 45800
    },
    {
      "entropy": 5.820954513549805,
      "epoch": 3.563781365493095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037629744785414027,
      "loss": 4.9305,
      "mean_token_accuracy": 0.21047134548425675,
      "num_tokens": 79468906.0,
      "step": 45805
    },
    {
      "entropy": 5.904242897033692,
      "epoch": 3.5641703948648122,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037627269922051377,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20674313604831696,
      "num_tokens": 79478472.0,
      "step": 45810
    },
    {
      "entropy": 5.876113224029541,
      "epoch": 3.56455942423653,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037624794905023523,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21913743168115615,
      "num_tokens": 79486418.0,
      "step": 45815
    },
    {
      "entropy": 5.907744121551514,
      "epoch": 3.564948453608247,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003762231973436802,
      "loss": 5.008,
      "mean_token_accuracy": 0.20920095443725586,
      "num_tokens": 79495238.0,
      "step": 45820
    },
    {
      "entropy": 5.8446667194366455,
      "epoch": 3.565337482979965,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003761984441012243,
      "loss": 4.9033,
      "mean_token_accuracy": 0.22427387684583663,
      "num_tokens": 79504136.0,
      "step": 45825
    },
    {
      "entropy": 5.952948522567749,
      "epoch": 3.5657265123516826,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003761736893232431,
      "loss": 5.1032,
      "mean_token_accuracy": 0.20073896646499634,
      "num_tokens": 79513013.0,
      "step": 45830
    },
    {
      "entropy": 5.823034000396729,
      "epoch": 3.5661155417234003,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037614893301011223,
      "loss": 5.0837,
      "mean_token_accuracy": 0.20065441876649856,
      "num_tokens": 79521930.0,
      "step": 45835
    },
    {
      "entropy": 5.836878490447998,
      "epoch": 3.5665045710951175,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003761241751622072,
      "loss": 4.9544,
      "mean_token_accuracy": 0.21930686980485917,
      "num_tokens": 79530596.0,
      "step": 45840
    },
    {
      "entropy": 5.828778696060181,
      "epoch": 3.5668936004668352,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003760994157799038,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21095447689294816,
      "num_tokens": 79539145.0,
      "step": 45845
    },
    {
      "entropy": 5.8811522960662845,
      "epoch": 3.567282629838553,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003760746548635777,
      "loss": 4.9085,
      "mean_token_accuracy": 0.2207944467663765,
      "num_tokens": 79548041.0,
      "step": 45850
    },
    {
      "entropy": 5.881802368164062,
      "epoch": 3.56767165921027,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003760498924136046,
      "loss": 5.0722,
      "mean_token_accuracy": 0.20737486481666564,
      "num_tokens": 79557302.0,
      "step": 45855
    },
    {
      "entropy": 5.791285371780395,
      "epoch": 3.568060688581988,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037602512843036025,
      "loss": 4.9008,
      "mean_token_accuracy": 0.21256273090839387,
      "num_tokens": 79565718.0,
      "step": 45860
    },
    {
      "entropy": 5.823634672164917,
      "epoch": 3.5684497179537056,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003760003629142203,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2139604791998863,
      "num_tokens": 79573561.0,
      "step": 45865
    },
    {
      "entropy": 5.862139081954956,
      "epoch": 3.5688387473254233,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037597559586556063,
      "loss": 4.8538,
      "mean_token_accuracy": 0.22195274382829666,
      "num_tokens": 79582492.0,
      "step": 45870
    },
    {
      "entropy": 5.943076276779175,
      "epoch": 3.5692277766971405,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000375950827284757,
      "loss": 5.0549,
      "mean_token_accuracy": 0.20898479223251343,
      "num_tokens": 79591244.0,
      "step": 45875
    },
    {
      "entropy": 5.896769762039185,
      "epoch": 3.5696168060688582,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037592605717218523,
      "loss": 5.0799,
      "mean_token_accuracy": 0.20008875727653502,
      "num_tokens": 79599718.0,
      "step": 45880
    },
    {
      "entropy": 5.858865737915039,
      "epoch": 3.570005835440576,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037590128552822123,
      "loss": 4.8925,
      "mean_token_accuracy": 0.2194092407822609,
      "num_tokens": 79607408.0,
      "step": 45885
    },
    {
      "entropy": 5.914982843399048,
      "epoch": 3.570394864812293,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003758765123532407,
      "loss": 5.0348,
      "mean_token_accuracy": 0.20701408833265306,
      "num_tokens": 79616033.0,
      "step": 45890
    },
    {
      "entropy": 5.89799485206604,
      "epoch": 3.570783894184011,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003758517376476198,
      "loss": 5.0472,
      "mean_token_accuracy": 0.20917126834392546,
      "num_tokens": 79625512.0,
      "step": 45895
    },
    {
      "entropy": 5.8384157657623295,
      "epoch": 3.5711729235557286,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037582696141173435,
      "loss": 4.9576,
      "mean_token_accuracy": 0.21272702664136886,
      "num_tokens": 79635429.0,
      "step": 45900
    },
    {
      "entropy": 5.859475231170654,
      "epoch": 3.5715619529274463,
      "grad_norm": 1.421875,
      "learning_rate": 0.00037580218364596015,
      "loss": 4.8545,
      "mean_token_accuracy": 0.22064073830842973,
      "num_tokens": 79643778.0,
      "step": 45905
    },
    {
      "entropy": 5.8814873695373535,
      "epoch": 3.5719509822991635,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037577740435067327,
      "loss": 4.9795,
      "mean_token_accuracy": 0.20763864666223525,
      "num_tokens": 79652203.0,
      "step": 45910
    },
    {
      "entropy": 5.882450294494629,
      "epoch": 3.572340011670881,
      "grad_norm": 1.375,
      "learning_rate": 0.00037575262352624965,
      "loss": 4.9784,
      "mean_token_accuracy": 0.20885082334280014,
      "num_tokens": 79660891.0,
      "step": 45915
    },
    {
      "entropy": 5.834826231002808,
      "epoch": 3.5727290410425985,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037572784117306535,
      "loss": 5.006,
      "mean_token_accuracy": 0.2111694797873497,
      "num_tokens": 79669489.0,
      "step": 45920
    },
    {
      "entropy": 5.878077363967895,
      "epoch": 3.573118070414316,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003757030572914963,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2153889462351799,
      "num_tokens": 79678167.0,
      "step": 45925
    },
    {
      "entropy": 5.940540599822998,
      "epoch": 3.573507099786034,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003756782718819188,
      "loss": 5.0094,
      "mean_token_accuracy": 0.2104715123772621,
      "num_tokens": 79686202.0,
      "step": 45930
    },
    {
      "entropy": 5.864874219894409,
      "epoch": 3.5738961291577516,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037565348494470873,
      "loss": 5.0227,
      "mean_token_accuracy": 0.20707044899463653,
      "num_tokens": 79694187.0,
      "step": 45935
    },
    {
      "entropy": 5.7899213314056395,
      "epoch": 3.574285158529469,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003756286964802422,
      "loss": 4.9713,
      "mean_token_accuracy": 0.21531477719545364,
      "num_tokens": 79702898.0,
      "step": 45940
    },
    {
      "entropy": 5.92048921585083,
      "epoch": 3.5746741879011865,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003756039064888954,
      "loss": 5.0563,
      "mean_token_accuracy": 0.19817824214696883,
      "num_tokens": 79712122.0,
      "step": 45945
    },
    {
      "entropy": 5.875470399856567,
      "epoch": 3.575063217272904,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003755791149710444,
      "loss": 4.9565,
      "mean_token_accuracy": 0.2159022495150566,
      "num_tokens": 79720623.0,
      "step": 45950
    },
    {
      "entropy": 5.877656173706055,
      "epoch": 3.5754522466446215,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037555432192706547,
      "loss": 5.0122,
      "mean_token_accuracy": 0.21374149471521378,
      "num_tokens": 79729750.0,
      "step": 45955
    },
    {
      "entropy": 5.866694164276123,
      "epoch": 3.575841276016339,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037552952735733466,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22105976343154907,
      "num_tokens": 79738485.0,
      "step": 45960
    },
    {
      "entropy": 5.944686937332153,
      "epoch": 3.576230305388057,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003755047312622283,
      "loss": 5.1587,
      "mean_token_accuracy": 0.20127062946558,
      "num_tokens": 79747378.0,
      "step": 45965
    },
    {
      "entropy": 5.920019102096558,
      "epoch": 3.5766193347597746,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037547993364212267,
      "loss": 5.0131,
      "mean_token_accuracy": 0.21087877750396727,
      "num_tokens": 79756106.0,
      "step": 45970
    },
    {
      "entropy": 5.891671752929687,
      "epoch": 3.577008364131492,
      "grad_norm": 1.359375,
      "learning_rate": 0.000375455134497394,
      "loss": 4.9735,
      "mean_token_accuracy": 0.21524260342121124,
      "num_tokens": 79764347.0,
      "step": 45975
    },
    {
      "entropy": 5.811753797531128,
      "epoch": 3.5773973935032095,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003754303338284186,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21230557411909104,
      "num_tokens": 79772811.0,
      "step": 45980
    },
    {
      "entropy": 5.864470481872559,
      "epoch": 3.577786422874927,
      "grad_norm": 1.4375,
      "learning_rate": 0.00037540553163557263,
      "loss": 5.0355,
      "mean_token_accuracy": 0.21046366542577744,
      "num_tokens": 79781968.0,
      "step": 45985
    },
    {
      "entropy": 5.923332643508911,
      "epoch": 3.5781754522466445,
      "grad_norm": 1.375,
      "learning_rate": 0.00037538072791923257,
      "loss": 5.0408,
      "mean_token_accuracy": 0.21181668788194657,
      "num_tokens": 79790289.0,
      "step": 45990
    },
    {
      "entropy": 5.898117017745972,
      "epoch": 3.578564481618362,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037535592267977467,
      "loss": 5.0148,
      "mean_token_accuracy": 0.20468965768814087,
      "num_tokens": 79799007.0,
      "step": 45995
    },
    {
      "entropy": 5.865650033950805,
      "epoch": 3.57895351099008,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003753311159175754,
      "loss": 5.0035,
      "mean_token_accuracy": 0.20628161281347274,
      "num_tokens": 79807445.0,
      "step": 46000
    },
    {
      "entropy": 5.880518007278442,
      "epoch": 3.5793425403617976,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037530630763301113,
      "loss": 5.0714,
      "mean_token_accuracy": 0.2052274152636528,
      "num_tokens": 79817169.0,
      "step": 46005
    },
    {
      "entropy": 5.875416564941406,
      "epoch": 3.579731569733515,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003752814978264584,
      "loss": 4.9019,
      "mean_token_accuracy": 0.20770440697669984,
      "num_tokens": 79825491.0,
      "step": 46010
    },
    {
      "entropy": 5.961150979995727,
      "epoch": 3.5801205991052325,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003752566864982934,
      "loss": 5.0189,
      "mean_token_accuracy": 0.20848783254623413,
      "num_tokens": 79833904.0,
      "step": 46015
    },
    {
      "entropy": 5.879149675369263,
      "epoch": 3.5805096284769498,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00037523187364889293,
      "loss": 5.024,
      "mean_token_accuracy": 0.1999804899096489,
      "num_tokens": 79842778.0,
      "step": 46020
    },
    {
      "entropy": 5.868411731719971,
      "epoch": 3.5808986578486675,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003752070592786332,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21574199944734573,
      "num_tokens": 79851175.0,
      "step": 46025
    },
    {
      "entropy": 5.854500770568848,
      "epoch": 3.581287687220385,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037518224338789084,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21392289847135543,
      "num_tokens": 79860509.0,
      "step": 46030
    },
    {
      "entropy": 5.901912689208984,
      "epoch": 3.581676716592103,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003751574259770424,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20559228360652923,
      "num_tokens": 79869405.0,
      "step": 46035
    },
    {
      "entropy": 5.856354427337647,
      "epoch": 3.58206574596382,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003751326070464645,
      "loss": 4.9914,
      "mean_token_accuracy": 0.212415511906147,
      "num_tokens": 79877929.0,
      "step": 46040
    },
    {
      "entropy": 5.941971969604492,
      "epoch": 3.582454775335538,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003751077865965337,
      "loss": 5.0289,
      "mean_token_accuracy": 0.2112569361925125,
      "num_tokens": 79885860.0,
      "step": 46045
    },
    {
      "entropy": 5.895830154418945,
      "epoch": 3.5828438047072555,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037508296462762647,
      "loss": 4.969,
      "mean_token_accuracy": 0.20711284577846528,
      "num_tokens": 79894031.0,
      "step": 46050
    },
    {
      "entropy": 5.829115152359009,
      "epoch": 3.5832328340789728,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037505814114011956,
      "loss": 4.9093,
      "mean_token_accuracy": 0.21497206538915634,
      "num_tokens": 79902228.0,
      "step": 46055
    },
    {
      "entropy": 5.845581293106079,
      "epoch": 3.5836218634506904,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037503331613438965,
      "loss": 4.9639,
      "mean_token_accuracy": 0.21814270168542862,
      "num_tokens": 79911195.0,
      "step": 46060
    },
    {
      "entropy": 5.878824090957641,
      "epoch": 3.584010892822408,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037500848961081334,
      "loss": 4.9715,
      "mean_token_accuracy": 0.20980352610349656,
      "num_tokens": 79919844.0,
      "step": 46065
    },
    {
      "entropy": 5.881328773498535,
      "epoch": 3.584399922194126,
      "grad_norm": 1.453125,
      "learning_rate": 0.00037498366156976746,
      "loss": 4.9134,
      "mean_token_accuracy": 0.21261506527662277,
      "num_tokens": 79927925.0,
      "step": 46070
    },
    {
      "entropy": 5.866397285461426,
      "epoch": 3.584788951565843,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037495883201162854,
      "loss": 5.068,
      "mean_token_accuracy": 0.20412982404232025,
      "num_tokens": 79937517.0,
      "step": 46075
    },
    {
      "entropy": 5.807900381088257,
      "epoch": 3.585177980937561,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003749340009367736,
      "loss": 4.9448,
      "mean_token_accuracy": 0.21405884325504304,
      "num_tokens": 79946203.0,
      "step": 46080
    },
    {
      "entropy": 5.883450174331665,
      "epoch": 3.5855670103092785,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00037490916834557916,
      "loss": 4.9274,
      "mean_token_accuracy": 0.21738828122615814,
      "num_tokens": 79954216.0,
      "step": 46085
    },
    {
      "entropy": 5.85238995552063,
      "epoch": 3.5859560396809957,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003748843342384221,
      "loss": 4.984,
      "mean_token_accuracy": 0.21415503770112992,
      "num_tokens": 79962420.0,
      "step": 46090
    },
    {
      "entropy": 5.875460910797119,
      "epoch": 3.5863450690527134,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037485949861567925,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20908914208412172,
      "num_tokens": 79971650.0,
      "step": 46095
    },
    {
      "entropy": 5.866400909423828,
      "epoch": 3.586734098424431,
      "grad_norm": 1.25,
      "learning_rate": 0.0003748346614777275,
      "loss": 4.9942,
      "mean_token_accuracy": 0.20947646498680114,
      "num_tokens": 79980046.0,
      "step": 46100
    },
    {
      "entropy": 5.857568597793579,
      "epoch": 3.587123127796149,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003748098228249437,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21181931644678115,
      "num_tokens": 79988257.0,
      "step": 46105
    },
    {
      "entropy": 5.849348831176758,
      "epoch": 3.587512157167866,
      "grad_norm": 1.4375,
      "learning_rate": 0.00037478498265770463,
      "loss": 4.9233,
      "mean_token_accuracy": 0.21124979853630066,
      "num_tokens": 79996982.0,
      "step": 46110
    },
    {
      "entropy": 5.8945395946502686,
      "epoch": 3.587901186539584,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003747601409763872,
      "loss": 5.0619,
      "mean_token_accuracy": 0.20623077899217607,
      "num_tokens": 80006309.0,
      "step": 46115
    },
    {
      "entropy": 5.883035802841187,
      "epoch": 3.588290215911301,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037473529778136853,
      "loss": 4.9779,
      "mean_token_accuracy": 0.2090377241373062,
      "num_tokens": 80015082.0,
      "step": 46120
    },
    {
      "entropy": 5.840625858306884,
      "epoch": 3.5886792452830187,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037471045307302544,
      "loss": 4.8835,
      "mean_token_accuracy": 0.2141177996993065,
      "num_tokens": 80023469.0,
      "step": 46125
    },
    {
      "entropy": 5.88462266921997,
      "epoch": 3.5890682746547364,
      "grad_norm": 1.5,
      "learning_rate": 0.00037468560685173497,
      "loss": 4.983,
      "mean_token_accuracy": 0.21068092286586762,
      "num_tokens": 80033359.0,
      "step": 46130
    },
    {
      "entropy": 5.808923053741455,
      "epoch": 3.589457304026454,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000374660759117874,
      "loss": 4.8563,
      "mean_token_accuracy": 0.2218080699443817,
      "num_tokens": 80042182.0,
      "step": 46135
    },
    {
      "entropy": 5.828692245483398,
      "epoch": 3.589846333398172,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037463590987181966,
      "loss": 4.9878,
      "mean_token_accuracy": 0.21008822917938233,
      "num_tokens": 80050372.0,
      "step": 46140
    },
    {
      "entropy": 5.808371639251709,
      "epoch": 3.590235362769889,
      "grad_norm": 1.484375,
      "learning_rate": 0.00037461105911394904,
      "loss": 4.8537,
      "mean_token_accuracy": 0.2209500104188919,
      "num_tokens": 80058001.0,
      "step": 46145
    },
    {
      "entropy": 5.822599935531616,
      "epoch": 3.590624392141607,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037458620684463907,
      "loss": 4.9506,
      "mean_token_accuracy": 0.21376388370990754,
      "num_tokens": 80066549.0,
      "step": 46150
    },
    {
      "entropy": 5.795543670654297,
      "epoch": 3.591013421513324,
      "grad_norm": 1.25,
      "learning_rate": 0.00037456135306426693,
      "loss": 4.8379,
      "mean_token_accuracy": 0.21953493356704712,
      "num_tokens": 80075316.0,
      "step": 46155
    },
    {
      "entropy": 5.898145627975464,
      "epoch": 3.5914024508850417,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003745364977732097,
      "loss": 5.0464,
      "mean_token_accuracy": 0.2096621200442314,
      "num_tokens": 80083644.0,
      "step": 46160
    },
    {
      "entropy": 5.875305271148681,
      "epoch": 3.5917914802567594,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003745116409718446,
      "loss": 4.9276,
      "mean_token_accuracy": 0.21191998124122619,
      "num_tokens": 80092266.0,
      "step": 46165
    },
    {
      "entropy": 5.925926351547242,
      "epoch": 3.592180509628477,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003744867826605488,
      "loss": 4.9692,
      "mean_token_accuracy": 0.21589736342430116,
      "num_tokens": 80101039.0,
      "step": 46170
    },
    {
      "entropy": 5.827014970779419,
      "epoch": 3.5925695390001944,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003744619228396993,
      "loss": 4.952,
      "mean_token_accuracy": 0.2111702173948288,
      "num_tokens": 80109210.0,
      "step": 46175
    },
    {
      "entropy": 5.8238622665405275,
      "epoch": 3.592958568371912,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003744370615096734,
      "loss": 4.9237,
      "mean_token_accuracy": 0.21691021621227263,
      "num_tokens": 80118438.0,
      "step": 46180
    },
    {
      "entropy": 5.901769971847534,
      "epoch": 3.5933475977436298,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003744121986708485,
      "loss": 4.958,
      "mean_token_accuracy": 0.2069777473807335,
      "num_tokens": 80126363.0,
      "step": 46185
    },
    {
      "entropy": 5.906550884246826,
      "epoch": 3.593736627115347,
      "grad_norm": 1.375,
      "learning_rate": 0.00037438733432360163,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21050717681646347,
      "num_tokens": 80134186.0,
      "step": 46190
    },
    {
      "entropy": 5.849613046646118,
      "epoch": 3.5941256564870647,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003743624684683102,
      "loss": 4.994,
      "mean_token_accuracy": 0.2044980674982071,
      "num_tokens": 80141881.0,
      "step": 46195
    },
    {
      "entropy": 5.857237911224365,
      "epoch": 3.5945146858587824,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003743376011053514,
      "loss": 5.0656,
      "mean_token_accuracy": 0.20415599197149276,
      "num_tokens": 80150047.0,
      "step": 46200
    },
    {
      "entropy": 5.8955518245697025,
      "epoch": 3.5949037152305,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00037431273223510265,
      "loss": 4.9892,
      "mean_token_accuracy": 0.20903924554586412,
      "num_tokens": 80158126.0,
      "step": 46205
    },
    {
      "entropy": 5.886825323104858,
      "epoch": 3.5952927446022174,
      "grad_norm": 1.25,
      "learning_rate": 0.0003742878618579413,
      "loss": 4.9813,
      "mean_token_accuracy": 0.20976168513298035,
      "num_tokens": 80167046.0,
      "step": 46210
    },
    {
      "entropy": 5.928079223632812,
      "epoch": 3.595681773973935,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003742629899742446,
      "loss": 5.0578,
      "mean_token_accuracy": 0.19952477663755416,
      "num_tokens": 80176296.0,
      "step": 46215
    },
    {
      "entropy": 5.869991683959961,
      "epoch": 3.5960708033456523,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037423811658439014,
      "loss": 4.935,
      "mean_token_accuracy": 0.21350838840007783,
      "num_tokens": 80184472.0,
      "step": 46220
    },
    {
      "entropy": 5.876732921600341,
      "epoch": 3.59645983271737,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037421324168875503,
      "loss": 5.0123,
      "mean_token_accuracy": 0.2097334608435631,
      "num_tokens": 80193334.0,
      "step": 46225
    },
    {
      "entropy": 5.8576991081237795,
      "epoch": 3.5968488620890877,
      "grad_norm": 1.3125,
      "learning_rate": 0.000374188365287717,
      "loss": 5.0398,
      "mean_token_accuracy": 0.20874162912368774,
      "num_tokens": 80202633.0,
      "step": 46230
    },
    {
      "entropy": 5.880957984924317,
      "epoch": 3.5972378914608054,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003741634873816534,
      "loss": 5.0024,
      "mean_token_accuracy": 0.20587234497070311,
      "num_tokens": 80211454.0,
      "step": 46235
    },
    {
      "entropy": 5.810548305511475,
      "epoch": 3.597626920832523,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037413860797094175,
      "loss": 4.9428,
      "mean_token_accuracy": 0.21535708457231523,
      "num_tokens": 80219748.0,
      "step": 46240
    },
    {
      "entropy": 5.86266713142395,
      "epoch": 3.5980159502042404,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037411372705595947,
      "loss": 5.0881,
      "mean_token_accuracy": 0.20693282186985015,
      "num_tokens": 80229503.0,
      "step": 46245
    },
    {
      "entropy": 5.909901094436646,
      "epoch": 3.598404979575958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003740888446370842,
      "loss": 4.9001,
      "mean_token_accuracy": 0.22200883477926253,
      "num_tokens": 80238432.0,
      "step": 46250
    },
    {
      "entropy": 5.887056589126587,
      "epoch": 3.5987940089476753,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003740639607146933,
      "loss": 4.9253,
      "mean_token_accuracy": 0.21928769201040268,
      "num_tokens": 80247478.0,
      "step": 46255
    },
    {
      "entropy": 5.898938417434692,
      "epoch": 3.599183038319393,
      "grad_norm": 1.46875,
      "learning_rate": 0.00037403907528916455,
      "loss": 5.1106,
      "mean_token_accuracy": 0.20271747559309006,
      "num_tokens": 80256643.0,
      "step": 46260
    },
    {
      "entropy": 5.894898366928101,
      "epoch": 3.5995720676911107,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003740141883608755,
      "loss": 5.003,
      "mean_token_accuracy": 0.2050217017531395,
      "num_tokens": 80265351.0,
      "step": 46265
    },
    {
      "entropy": 5.858670043945312,
      "epoch": 3.5999610970628284,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003739892999302037,
      "loss": 4.9019,
      "mean_token_accuracy": 0.22504570484161376,
      "num_tokens": 80274698.0,
      "step": 46270
    },
    {
      "entropy": 5.849434852600098,
      "epoch": 3.6003501264345457,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003739644099975269,
      "loss": 4.9037,
      "mean_token_accuracy": 0.20673559606075287,
      "num_tokens": 80282763.0,
      "step": 46275
    },
    {
      "entropy": 5.879590749740601,
      "epoch": 3.6007391558062634,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037393951856322263,
      "loss": 5.0503,
      "mean_token_accuracy": 0.20671216696500777,
      "num_tokens": 80292095.0,
      "step": 46280
    },
    {
      "entropy": 5.862115716934204,
      "epoch": 3.601128185177981,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003739146256276686,
      "loss": 4.938,
      "mean_token_accuracy": 0.21387362033128737,
      "num_tokens": 80299897.0,
      "step": 46285
    },
    {
      "entropy": 5.823745059967041,
      "epoch": 3.6015172145496983,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037388973119124267,
      "loss": 5.0015,
      "mean_token_accuracy": 0.20010467022657394,
      "num_tokens": 80309371.0,
      "step": 46290
    },
    {
      "entropy": 5.828321075439453,
      "epoch": 3.601906243921416,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003738648352543224,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21129647493362427,
      "num_tokens": 80319177.0,
      "step": 46295
    },
    {
      "entropy": 5.871084547042846,
      "epoch": 3.6022952732931337,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003738399378172856,
      "loss": 4.9475,
      "mean_token_accuracy": 0.2121367871761322,
      "num_tokens": 80327618.0,
      "step": 46300
    },
    {
      "entropy": 5.876247692108154,
      "epoch": 3.6026843026648514,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003738150388805101,
      "loss": 5.023,
      "mean_token_accuracy": 0.20406688749790192,
      "num_tokens": 80336576.0,
      "step": 46305
    },
    {
      "entropy": 5.883008861541748,
      "epoch": 3.6030733320365687,
      "grad_norm": 1.515625,
      "learning_rate": 0.00037379013844437373,
      "loss": 4.9531,
      "mean_token_accuracy": 0.21754887998104094,
      "num_tokens": 80344614.0,
      "step": 46310
    },
    {
      "entropy": 5.92507905960083,
      "epoch": 3.6034623614082864,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037376523650925415,
      "loss": 4.9815,
      "mean_token_accuracy": 0.20936556458473204,
      "num_tokens": 80353414.0,
      "step": 46315
    },
    {
      "entropy": 5.830273771286011,
      "epoch": 3.603851390780004,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003737403330755293,
      "loss": 5.0077,
      "mean_token_accuracy": 0.2048140600323677,
      "num_tokens": 80362472.0,
      "step": 46320
    },
    {
      "entropy": 5.893815517425537,
      "epoch": 3.6042404201517213,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003737154281435771,
      "loss": 4.9847,
      "mean_token_accuracy": 0.21396642327308654,
      "num_tokens": 80370995.0,
      "step": 46325
    },
    {
      "entropy": 5.882711505889892,
      "epoch": 3.604629449523439,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037369052171377535,
      "loss": 5.0793,
      "mean_token_accuracy": 0.20336281210184098,
      "num_tokens": 80379725.0,
      "step": 46330
    },
    {
      "entropy": 5.875619840621948,
      "epoch": 3.6050184788951567,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003736656137865021,
      "loss": 4.8821,
      "mean_token_accuracy": 0.21493092477321624,
      "num_tokens": 80388289.0,
      "step": 46335
    },
    {
      "entropy": 5.87822904586792,
      "epoch": 3.6054075082668744,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037364070436213514,
      "loss": 4.9462,
      "mean_token_accuracy": 0.22077955305576324,
      "num_tokens": 80396293.0,
      "step": 46340
    },
    {
      "entropy": 5.85198130607605,
      "epoch": 3.6057965376385916,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003736157934410525,
      "loss": 4.9908,
      "mean_token_accuracy": 0.21193979382514955,
      "num_tokens": 80405258.0,
      "step": 46345
    },
    {
      "entropy": 5.820558071136475,
      "epoch": 3.6061855670103093,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003735908810236322,
      "loss": 4.8683,
      "mean_token_accuracy": 0.21868887096643447,
      "num_tokens": 80413198.0,
      "step": 46350
    },
    {
      "entropy": 5.848029518127442,
      "epoch": 3.6065745963820266,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037356596711025214,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2101803109049797,
      "num_tokens": 80421707.0,
      "step": 46355
    },
    {
      "entropy": 5.893310546875,
      "epoch": 3.6069636257537443,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003735410517012905,
      "loss": 5.0016,
      "mean_token_accuracy": 0.20872990489006044,
      "num_tokens": 80430207.0,
      "step": 46360
    },
    {
      "entropy": 5.9498326778411865,
      "epoch": 3.607352655125462,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003735161347971252,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20334067940711975,
      "num_tokens": 80439255.0,
      "step": 46365
    },
    {
      "entropy": 5.87770709991455,
      "epoch": 3.6077416844971797,
      "grad_norm": 1.375,
      "learning_rate": 0.0003734912163981344,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21765171736478806,
      "num_tokens": 80447468.0,
      "step": 46370
    },
    {
      "entropy": 5.848084354400635,
      "epoch": 3.608130713868897,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037346629650469605,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20012205988168716,
      "num_tokens": 80456496.0,
      "step": 46375
    },
    {
      "entropy": 5.91798620223999,
      "epoch": 3.6085197432406146,
      "grad_norm": 1.375,
      "learning_rate": 0.00037344137511718847,
      "loss": 5.0636,
      "mean_token_accuracy": 0.204497130215168,
      "num_tokens": 80465078.0,
      "step": 46380
    },
    {
      "entropy": 5.825898218154907,
      "epoch": 3.6089087726123323,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037341645223598965,
      "loss": 4.9661,
      "mean_token_accuracy": 0.21278919875621796,
      "num_tokens": 80473241.0,
      "step": 46385
    },
    {
      "entropy": 5.855575037002564,
      "epoch": 3.6092978019840496,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037339152786147777,
      "loss": 4.927,
      "mean_token_accuracy": 0.21428704559803008,
      "num_tokens": 80481487.0,
      "step": 46390
    },
    {
      "entropy": 5.87964129447937,
      "epoch": 3.6096868313557673,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037336660199403115,
      "loss": 4.975,
      "mean_token_accuracy": 0.21076726913452148,
      "num_tokens": 80489327.0,
      "step": 46395
    },
    {
      "entropy": 5.835446500778199,
      "epoch": 3.610075860727485,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003733416746340279,
      "loss": 5.0499,
      "mean_token_accuracy": 0.20607276111841202,
      "num_tokens": 80497597.0,
      "step": 46400
    },
    {
      "entropy": 5.914765882492065,
      "epoch": 3.6104648900992027,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003733167457818463,
      "loss": 4.944,
      "mean_token_accuracy": 0.2136630743741989,
      "num_tokens": 80505734.0,
      "step": 46405
    },
    {
      "entropy": 5.87595944404602,
      "epoch": 3.61085391947092,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003732918154378645,
      "loss": 4.8539,
      "mean_token_accuracy": 0.21776183247566222,
      "num_tokens": 80514080.0,
      "step": 46410
    },
    {
      "entropy": 5.855459070205688,
      "epoch": 3.6112429488426376,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00037326688360246087,
      "loss": 4.954,
      "mean_token_accuracy": 0.21237864196300507,
      "num_tokens": 80522748.0,
      "step": 46415
    },
    {
      "entropy": 5.90538592338562,
      "epoch": 3.6116319782143553,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003732419502760137,
      "loss": 4.9146,
      "mean_token_accuracy": 0.21069543659687043,
      "num_tokens": 80531060.0,
      "step": 46420
    },
    {
      "entropy": 5.818411874771118,
      "epoch": 3.6120210075860726,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037321701545890126,
      "loss": 4.9841,
      "mean_token_accuracy": 0.21415457725524903,
      "num_tokens": 80539317.0,
      "step": 46425
    },
    {
      "entropy": 5.898630428314209,
      "epoch": 3.6124100369577903,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003731920791515019,
      "loss": 5.0122,
      "mean_token_accuracy": 0.20522041022777557,
      "num_tokens": 80548382.0,
      "step": 46430
    },
    {
      "entropy": 5.906782531738282,
      "epoch": 3.612799066329508,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003731671413541942,
      "loss": 4.9986,
      "mean_token_accuracy": 0.20882537662982942,
      "num_tokens": 80556957.0,
      "step": 46435
    },
    {
      "entropy": 5.854633045196533,
      "epoch": 3.6131880957012257,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037314220206735626,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2204595535993576,
      "num_tokens": 80565102.0,
      "step": 46440
    },
    {
      "entropy": 5.824913549423218,
      "epoch": 3.613577125072943,
      "grad_norm": 1.453125,
      "learning_rate": 0.00037311726129136665,
      "loss": 4.878,
      "mean_token_accuracy": 0.21720949709415435,
      "num_tokens": 80573774.0,
      "step": 46445
    },
    {
      "entropy": 5.8358931064605715,
      "epoch": 3.6139661544446606,
      "grad_norm": 1.25,
      "learning_rate": 0.0003730923190266038,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20975459665060042,
      "num_tokens": 80582217.0,
      "step": 46450
    },
    {
      "entropy": 5.867620897293091,
      "epoch": 3.614355183816378,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000373067375273446,
      "loss": 5.0227,
      "mean_token_accuracy": 0.20011592358350755,
      "num_tokens": 80589978.0,
      "step": 46455
    },
    {
      "entropy": 5.96382851600647,
      "epoch": 3.6147442131880956,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037304243003227197,
      "loss": 5.1716,
      "mean_token_accuracy": 0.2022392377257347,
      "num_tokens": 80598333.0,
      "step": 46460
    },
    {
      "entropy": 5.891906213760376,
      "epoch": 3.6151332425598133,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003730174833034601,
      "loss": 4.9898,
      "mean_token_accuracy": 0.20581740736961365,
      "num_tokens": 80607442.0,
      "step": 46465
    },
    {
      "entropy": 5.90625696182251,
      "epoch": 3.615522271931531,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037299253508738894,
      "loss": 4.9191,
      "mean_token_accuracy": 0.21384364515542983,
      "num_tokens": 80616002.0,
      "step": 46470
    },
    {
      "entropy": 5.953362131118775,
      "epoch": 3.6159113013032487,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037296758538443703,
      "loss": 5.0941,
      "mean_token_accuracy": 0.19955772459506987,
      "num_tokens": 80625066.0,
      "step": 46475
    },
    {
      "entropy": 5.924220323562622,
      "epoch": 3.616300330674966,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037294263419498295,
      "loss": 4.978,
      "mean_token_accuracy": 0.21444753408432007,
      "num_tokens": 80633809.0,
      "step": 46480
    },
    {
      "entropy": 5.882300662994385,
      "epoch": 3.6166893600466836,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003729176815194053,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2062678411602974,
      "num_tokens": 80641761.0,
      "step": 46485
    },
    {
      "entropy": 5.832078456878662,
      "epoch": 3.617078389418401,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037289272735808267,
      "loss": 4.9422,
      "mean_token_accuracy": 0.22402411848306655,
      "num_tokens": 80649977.0,
      "step": 46490
    },
    {
      "entropy": 5.825156021118164,
      "epoch": 3.6174674187901186,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003728677717113936,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20727197527885438,
      "num_tokens": 80658851.0,
      "step": 46495
    },
    {
      "entropy": 5.77953987121582,
      "epoch": 3.6178564481618363,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037284281457971707,
      "loss": 4.8931,
      "mean_token_accuracy": 0.20774994790554047,
      "num_tokens": 80667758.0,
      "step": 46500
    },
    {
      "entropy": 5.860615825653076,
      "epoch": 3.618245477533554,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003728178559634315,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21559597849845885,
      "num_tokens": 80676309.0,
      "step": 46505
    },
    {
      "entropy": 5.856022834777832,
      "epoch": 3.618634506905271,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003727928958629156,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21178776770830154,
      "num_tokens": 80683966.0,
      "step": 46510
    },
    {
      "entropy": 5.891255569458008,
      "epoch": 3.619023536276989,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037276793427854817,
      "loss": 5.0823,
      "mean_token_accuracy": 0.2026694357395172,
      "num_tokens": 80692850.0,
      "step": 46515
    },
    {
      "entropy": 5.903160095214844,
      "epoch": 3.6194125656487066,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037274297121070796,
      "loss": 4.9774,
      "mean_token_accuracy": 0.20803457796573638,
      "num_tokens": 80701404.0,
      "step": 46520
    },
    {
      "entropy": 5.881598424911499,
      "epoch": 3.619801595020424,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003727180066597737,
      "loss": 5.0241,
      "mean_token_accuracy": 0.20474547445774077,
      "num_tokens": 80710264.0,
      "step": 46525
    },
    {
      "entropy": 5.9047668933868405,
      "epoch": 3.6201906243921416,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003726930406261243,
      "loss": 5.0004,
      "mean_token_accuracy": 0.20578745901584625,
      "num_tokens": 80718946.0,
      "step": 46530
    },
    {
      "entropy": 5.85740327835083,
      "epoch": 3.6205796537638593,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003726680731101385,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21857842057943344,
      "num_tokens": 80726918.0,
      "step": 46535
    },
    {
      "entropy": 5.868860149383545,
      "epoch": 3.620968683135577,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003726431041121951,
      "loss": 4.9587,
      "mean_token_accuracy": 0.2118617445230484,
      "num_tokens": 80735024.0,
      "step": 46540
    },
    {
      "entropy": 5.876454639434814,
      "epoch": 3.621357712507294,
      "grad_norm": 1.34375,
      "learning_rate": 0.000372618133632673,
      "loss": 5.0011,
      "mean_token_accuracy": 0.20349937081336975,
      "num_tokens": 80743393.0,
      "step": 46545
    },
    {
      "entropy": 5.7621081352233885,
      "epoch": 3.621746741879012,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003725931616719511,
      "loss": 4.9028,
      "mean_token_accuracy": 0.21248666793107987,
      "num_tokens": 80753058.0,
      "step": 46550
    },
    {
      "entropy": 5.936996793746948,
      "epoch": 3.622135771250729,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003725681882304084,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20232336670160295,
      "num_tokens": 80761883.0,
      "step": 46555
    },
    {
      "entropy": 5.9229692935943605,
      "epoch": 3.622524800622447,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003725432133084237,
      "loss": 4.969,
      "mean_token_accuracy": 0.21203252375125886,
      "num_tokens": 80770665.0,
      "step": 46560
    },
    {
      "entropy": 6.032285404205322,
      "epoch": 3.6229138299941646,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003725182369063759,
      "loss": 5.1275,
      "mean_token_accuracy": 0.19932887256145476,
      "num_tokens": 80779123.0,
      "step": 46565
    },
    {
      "entropy": 5.865744590759277,
      "epoch": 3.6233028593658823,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003724932590246441,
      "loss": 4.9338,
      "mean_token_accuracy": 0.2123592883348465,
      "num_tokens": 80787643.0,
      "step": 46570
    },
    {
      "entropy": 5.881605720520019,
      "epoch": 3.6236918887376,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003724682796636073,
      "loss": 5.0152,
      "mean_token_accuracy": 0.2087026759982109,
      "num_tokens": 80797457.0,
      "step": 46575
    },
    {
      "entropy": 5.883164119720459,
      "epoch": 3.624080918109317,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003724432988236445,
      "loss": 5.0391,
      "mean_token_accuracy": 0.20505425035953523,
      "num_tokens": 80806632.0,
      "step": 46580
    },
    {
      "entropy": 5.935055446624756,
      "epoch": 3.624469947481035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003724183165051347,
      "loss": 5.0675,
      "mean_token_accuracy": 0.20734118521213532,
      "num_tokens": 80815625.0,
      "step": 46585
    },
    {
      "entropy": 5.92406735420227,
      "epoch": 3.624858976852752,
      "grad_norm": 1.453125,
      "learning_rate": 0.00037239333270845714,
      "loss": 4.913,
      "mean_token_accuracy": 0.21828123033046723,
      "num_tokens": 80823808.0,
      "step": 46590
    },
    {
      "entropy": 5.906607532501221,
      "epoch": 3.62524800622447,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003723683474339907,
      "loss": 5.0092,
      "mean_token_accuracy": 0.20444276034832,
      "num_tokens": 80832093.0,
      "step": 46595
    },
    {
      "entropy": 5.849700975418091,
      "epoch": 3.6256370355961876,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003723433606821145,
      "loss": 4.9682,
      "mean_token_accuracy": 0.2150287002325058,
      "num_tokens": 80840411.0,
      "step": 46600
    },
    {
      "entropy": 5.890695190429687,
      "epoch": 3.6260260649679052,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037231837245320777,
      "loss": 4.9539,
      "mean_token_accuracy": 0.2094782754778862,
      "num_tokens": 80848923.0,
      "step": 46605
    },
    {
      "entropy": 5.913061618804932,
      "epoch": 3.6264150943396225,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037229338274764965,
      "loss": 5.0189,
      "mean_token_accuracy": 0.209695440530777,
      "num_tokens": 80858459.0,
      "step": 46610
    },
    {
      "entropy": 5.970396566390991,
      "epoch": 3.62680412371134,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003722683915658193,
      "loss": 5.0661,
      "mean_token_accuracy": 0.20314272195100785,
      "num_tokens": 80867559.0,
      "step": 46615
    },
    {
      "entropy": 5.8923484802246096,
      "epoch": 3.627193153083058,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037224339890809587,
      "loss": 4.9623,
      "mean_token_accuracy": 0.2098035469651222,
      "num_tokens": 80875285.0,
      "step": 46620
    },
    {
      "entropy": 5.7909900665283205,
      "epoch": 3.627582182454775,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003722184047748587,
      "loss": 4.9318,
      "mean_token_accuracy": 0.2115820527076721,
      "num_tokens": 80883805.0,
      "step": 46625
    },
    {
      "entropy": 5.90681939125061,
      "epoch": 3.627971211826493,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00037219340916648703,
      "loss": 5.0132,
      "mean_token_accuracy": 0.21286354959011078,
      "num_tokens": 80892496.0,
      "step": 46630
    },
    {
      "entropy": 5.832587862014771,
      "epoch": 3.6283602411982105,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037216841208336,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21940865516662597,
      "num_tokens": 80901364.0,
      "step": 46635
    },
    {
      "entropy": 5.824172163009644,
      "epoch": 3.6287492705699282,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037214341352585697,
      "loss": 4.9323,
      "mean_token_accuracy": 0.21472516804933547,
      "num_tokens": 80910563.0,
      "step": 46640
    },
    {
      "entropy": 5.858545923233033,
      "epoch": 3.6291382999416455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003721184134943573,
      "loss": 4.9362,
      "mean_token_accuracy": 0.2087385579943657,
      "num_tokens": 80918939.0,
      "step": 46645
    },
    {
      "entropy": 5.920136260986328,
      "epoch": 3.629527329313363,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037209341198924024,
      "loss": 5.066,
      "mean_token_accuracy": 0.20816862732172012,
      "num_tokens": 80927896.0,
      "step": 46650
    },
    {
      "entropy": 5.877099847793579,
      "epoch": 3.6299163586850804,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003720684090108852,
      "loss": 4.9963,
      "mean_token_accuracy": 0.2129910632967949,
      "num_tokens": 80936961.0,
      "step": 46655
    },
    {
      "entropy": 5.891387271881103,
      "epoch": 3.630305388056798,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003720434045596716,
      "loss": 5.0122,
      "mean_token_accuracy": 0.2051403343677521,
      "num_tokens": 80947240.0,
      "step": 46660
    },
    {
      "entropy": 5.87029390335083,
      "epoch": 3.630694417428516,
      "grad_norm": 1.59375,
      "learning_rate": 0.0003720183986359788,
      "loss": 4.8988,
      "mean_token_accuracy": 0.22364370673894882,
      "num_tokens": 80955722.0,
      "step": 46665
    },
    {
      "entropy": 5.825017213821411,
      "epoch": 3.6310834468002335,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003719933912401862,
      "loss": 4.9201,
      "mean_token_accuracy": 0.21871040612459183,
      "num_tokens": 80964224.0,
      "step": 46670
    },
    {
      "entropy": 5.858215284347534,
      "epoch": 3.6314724761719512,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003719683823726732,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20352280437946318,
      "num_tokens": 80972379.0,
      "step": 46675
    },
    {
      "entropy": 5.769119358062744,
      "epoch": 3.6318615055436685,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037194337203381945,
      "loss": 4.8192,
      "mean_token_accuracy": 0.22052616477012635,
      "num_tokens": 80980365.0,
      "step": 46680
    },
    {
      "entropy": 5.8707255840301515,
      "epoch": 3.632250534915386,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00037191836022400427,
      "loss": 5.0681,
      "mean_token_accuracy": 0.20385424345731734,
      "num_tokens": 80989938.0,
      "step": 46685
    },
    {
      "entropy": 5.93416428565979,
      "epoch": 3.6326395642871034,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037189334694360724,
      "loss": 5.0234,
      "mean_token_accuracy": 0.2064856395125389,
      "num_tokens": 80998300.0,
      "step": 46690
    },
    {
      "entropy": 5.987253046035766,
      "epoch": 3.633028593658821,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003718683321930079,
      "loss": 5.1411,
      "mean_token_accuracy": 0.203850919008255,
      "num_tokens": 81007802.0,
      "step": 46695
    },
    {
      "entropy": 5.88831353187561,
      "epoch": 3.633417623030539,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003718433159725857,
      "loss": 5.0221,
      "mean_token_accuracy": 0.2063385874032974,
      "num_tokens": 81016668.0,
      "step": 46700
    },
    {
      "entropy": 5.8856714248657225,
      "epoch": 3.6338066524022565,
      "grad_norm": 1.5,
      "learning_rate": 0.0003718182982827205,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21400780230760574,
      "num_tokens": 81024492.0,
      "step": 46705
    },
    {
      "entropy": 5.840031671524048,
      "epoch": 3.634195681773974,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037179327912379166,
      "loss": 5.03,
      "mean_token_accuracy": 0.20880710929632187,
      "num_tokens": 81033787.0,
      "step": 46710
    },
    {
      "entropy": 5.880104017257691,
      "epoch": 3.6345847111456915,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037176825849617886,
      "loss": 5.017,
      "mean_token_accuracy": 0.20855602324008943,
      "num_tokens": 81043098.0,
      "step": 46715
    },
    {
      "entropy": 5.854564476013183,
      "epoch": 3.634973740517409,
      "grad_norm": 1.25,
      "learning_rate": 0.00037174323640026173,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21144081503152848,
      "num_tokens": 81052638.0,
      "step": 46720
    },
    {
      "entropy": 5.885738468170166,
      "epoch": 3.6353627698891264,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037171821283642003,
      "loss": 4.9412,
      "mean_token_accuracy": 0.21109349131584168,
      "num_tokens": 81061350.0,
      "step": 46725
    },
    {
      "entropy": 5.792341041564941,
      "epoch": 3.635751799260844,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003716931878050334,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22055627703666686,
      "num_tokens": 81071013.0,
      "step": 46730
    },
    {
      "entropy": 5.875975894927978,
      "epoch": 3.636140828632562,
      "grad_norm": 1.40625,
      "learning_rate": 0.00037166816130648143,
      "loss": 5.0428,
      "mean_token_accuracy": 0.20620951801538467,
      "num_tokens": 81080705.0,
      "step": 46735
    },
    {
      "entropy": 5.915296459197998,
      "epoch": 3.6365298580042795,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037164313334114407,
      "loss": 4.9944,
      "mean_token_accuracy": 0.21317459642887115,
      "num_tokens": 81089422.0,
      "step": 46740
    },
    {
      "entropy": 5.902558040618897,
      "epoch": 3.636918887375997,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000371618103909401,
      "loss": 4.9912,
      "mean_token_accuracy": 0.20946327298879625,
      "num_tokens": 81097348.0,
      "step": 46745
    },
    {
      "entropy": 5.870756149291992,
      "epoch": 3.6373079167477145,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000371593073011632,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21996910721063614,
      "num_tokens": 81106083.0,
      "step": 46750
    },
    {
      "entropy": 5.878364133834839,
      "epoch": 3.637696946119432,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037156804064821674,
      "loss": 5.0326,
      "mean_token_accuracy": 0.20342165231704712,
      "num_tokens": 81114822.0,
      "step": 46755
    },
    {
      "entropy": 5.9075604438781735,
      "epoch": 3.6380859754911494,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037154300681953523,
      "loss": 4.991,
      "mean_token_accuracy": 0.20620027333498,
      "num_tokens": 81124037.0,
      "step": 46760
    },
    {
      "entropy": 5.93548173904419,
      "epoch": 3.638475004862867,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003715179715259673,
      "loss": 5.0085,
      "mean_token_accuracy": 0.2102886527776718,
      "num_tokens": 81133578.0,
      "step": 46765
    },
    {
      "entropy": 5.882705163955689,
      "epoch": 3.638864034234585,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037149293476789277,
      "loss": 4.9121,
      "mean_token_accuracy": 0.21679313182830812,
      "num_tokens": 81142073.0,
      "step": 46770
    },
    {
      "entropy": 5.90130877494812,
      "epoch": 3.6392530636063025,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003714678965456915,
      "loss": 5.0353,
      "mean_token_accuracy": 0.20645336657762528,
      "num_tokens": 81150271.0,
      "step": 46775
    },
    {
      "entropy": 5.929980134963989,
      "epoch": 3.6396420929780198,
      "grad_norm": 1.46875,
      "learning_rate": 0.00037144285685974346,
      "loss": 5.0399,
      "mean_token_accuracy": 0.21021473556756973,
      "num_tokens": 81158355.0,
      "step": 46780
    },
    {
      "entropy": 5.89545521736145,
      "epoch": 3.6400311223497375,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037141781571042854,
      "loss": 5.0231,
      "mean_token_accuracy": 0.2101237490773201,
      "num_tokens": 81167573.0,
      "step": 46785
    },
    {
      "entropy": 5.865622043609619,
      "epoch": 3.6404201517214547,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037139277309812676,
      "loss": 4.9176,
      "mean_token_accuracy": 0.21763512045145034,
      "num_tokens": 81176756.0,
      "step": 46790
    },
    {
      "entropy": 5.825727653503418,
      "epoch": 3.6408091810931724,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003713677290232181,
      "loss": 4.9011,
      "mean_token_accuracy": 0.21397103369235992,
      "num_tokens": 81185589.0,
      "step": 46795
    },
    {
      "entropy": 5.806036567687988,
      "epoch": 3.64119821046489,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037134268348608255,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21238091588020325,
      "num_tokens": 81194381.0,
      "step": 46800
    },
    {
      "entropy": 5.9399206161499025,
      "epoch": 3.641587239836608,
      "grad_norm": 1.390625,
      "learning_rate": 0.00037131763648710003,
      "loss": 5.0303,
      "mean_token_accuracy": 0.2019777625799179,
      "num_tokens": 81202441.0,
      "step": 46805
    },
    {
      "entropy": 5.92555832862854,
      "epoch": 3.641976269208325,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037129258802665076,
      "loss": 4.9915,
      "mean_token_accuracy": 0.20542169213294983,
      "num_tokens": 81211008.0,
      "step": 46810
    },
    {
      "entropy": 5.8925525665283205,
      "epoch": 3.6423652985800428,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003712675381051147,
      "loss": 4.9918,
      "mean_token_accuracy": 0.21255370676517488,
      "num_tokens": 81220585.0,
      "step": 46815
    },
    {
      "entropy": 5.8374334335327145,
      "epoch": 3.6427543279517605,
      "grad_norm": 1.46875,
      "learning_rate": 0.000371242486722872,
      "loss": 4.9701,
      "mean_token_accuracy": 0.2102351129055023,
      "num_tokens": 81229086.0,
      "step": 46820
    },
    {
      "entropy": 5.89926552772522,
      "epoch": 3.6431433573234777,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003712174338803027,
      "loss": 4.9858,
      "mean_token_accuracy": 0.20744353532791138,
      "num_tokens": 81237755.0,
      "step": 46825
    },
    {
      "entropy": 5.923137092590332,
      "epoch": 3.6435323866951954,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000371192379577787,
      "loss": 5.0667,
      "mean_token_accuracy": 0.20846085250377655,
      "num_tokens": 81247098.0,
      "step": 46830
    },
    {
      "entropy": 5.851921510696411,
      "epoch": 3.643921416066913,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003711673238157051,
      "loss": 4.9945,
      "mean_token_accuracy": 0.21187085658311844,
      "num_tokens": 81256103.0,
      "step": 46835
    },
    {
      "entropy": 5.833305978775025,
      "epoch": 3.644310445438631,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037114226659443704,
      "loss": 4.9337,
      "mean_token_accuracy": 0.20982952713966369,
      "num_tokens": 81264375.0,
      "step": 46840
    },
    {
      "entropy": 5.860716724395752,
      "epoch": 3.644699474810348,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003711172079143632,
      "loss": 4.8529,
      "mean_token_accuracy": 0.21892087161540985,
      "num_tokens": 81272589.0,
      "step": 46845
    },
    {
      "entropy": 5.813661432266235,
      "epoch": 3.6450885041820658,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003710921477758637,
      "loss": 4.929,
      "mean_token_accuracy": 0.2200906589627266,
      "num_tokens": 81280777.0,
      "step": 46850
    },
    {
      "entropy": 5.922341966629029,
      "epoch": 3.6454775335537835,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003710670861793187,
      "loss": 5.0729,
      "mean_token_accuracy": 0.20805466175079346,
      "num_tokens": 81290451.0,
      "step": 46855
    },
    {
      "entropy": 5.871334886550903,
      "epoch": 3.6458665629255007,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003710420231251087,
      "loss": 4.9238,
      "mean_token_accuracy": 0.21820848137140275,
      "num_tokens": 81299450.0,
      "step": 46860
    },
    {
      "entropy": 5.810213375091553,
      "epoch": 3.6462555922972184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037101695861361384,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21733849197626115,
      "num_tokens": 81308400.0,
      "step": 46865
    },
    {
      "entropy": 5.81704044342041,
      "epoch": 3.646644621668936,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00037099189264521446,
      "loss": 4.9319,
      "mean_token_accuracy": 0.21933533549308776,
      "num_tokens": 81316322.0,
      "step": 46870
    },
    {
      "entropy": 5.765547561645508,
      "epoch": 3.647033651040654,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003709668252202909,
      "loss": 4.8525,
      "mean_token_accuracy": 0.21882947832345961,
      "num_tokens": 81325444.0,
      "step": 46875
    },
    {
      "entropy": 5.848865652084351,
      "epoch": 3.647422680412371,
      "grad_norm": 1.375,
      "learning_rate": 0.0003709417563392235,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22343226820230483,
      "num_tokens": 81334279.0,
      "step": 46880
    },
    {
      "entropy": 5.882297468185425,
      "epoch": 3.6478117097840888,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003709166860023926,
      "loss": 4.8896,
      "mean_token_accuracy": 0.2170726016163826,
      "num_tokens": 81342641.0,
      "step": 46885
    },
    {
      "entropy": 5.814369440078735,
      "epoch": 3.648200739155806,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037089161421017874,
      "loss": 4.9267,
      "mean_token_accuracy": 0.21857561320066451,
      "num_tokens": 81352029.0,
      "step": 46890
    },
    {
      "entropy": 5.8552083492279055,
      "epoch": 3.6485897685275237,
      "grad_norm": 1.578125,
      "learning_rate": 0.00037086654096296217,
      "loss": 4.9718,
      "mean_token_accuracy": 0.21150198131799697,
      "num_tokens": 81361084.0,
      "step": 46895
    },
    {
      "entropy": 5.795525121688843,
      "epoch": 3.6489787978992414,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003708414662611235,
      "loss": 4.8842,
      "mean_token_accuracy": 0.2188178554177284,
      "num_tokens": 81369945.0,
      "step": 46900
    },
    {
      "entropy": 5.8677267074584964,
      "epoch": 3.649367827270959,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003708163901050432,
      "loss": 4.9971,
      "mean_token_accuracy": 0.21284661293029786,
      "num_tokens": 81378918.0,
      "step": 46905
    },
    {
      "entropy": 5.856452512741089,
      "epoch": 3.649756856642677,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037079131249510155,
      "loss": 4.9698,
      "mean_token_accuracy": 0.21594124436378478,
      "num_tokens": 81386937.0,
      "step": 46910
    },
    {
      "entropy": 5.889850044250489,
      "epoch": 3.650145886014394,
      "grad_norm": 1.375,
      "learning_rate": 0.0003707662334316793,
      "loss": 5.0358,
      "mean_token_accuracy": 0.21108922511339187,
      "num_tokens": 81396042.0,
      "step": 46915
    },
    {
      "entropy": 5.9429543018341064,
      "epoch": 3.6505349153861117,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003707411529151568,
      "loss": 5.0138,
      "mean_token_accuracy": 0.20987160950899125,
      "num_tokens": 81404737.0,
      "step": 46920
    },
    {
      "entropy": 5.910168695449829,
      "epoch": 3.650923944757829,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003707160709459147,
      "loss": 5.1012,
      "mean_token_accuracy": 0.19917007088661193,
      "num_tokens": 81414197.0,
      "step": 46925
    },
    {
      "entropy": 5.824350452423095,
      "epoch": 3.6513129741295467,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003706909875243336,
      "loss": 4.9486,
      "mean_token_accuracy": 0.21237266808748245,
      "num_tokens": 81423033.0,
      "step": 46930
    },
    {
      "entropy": 5.835911750793457,
      "epoch": 3.6517020035012644,
      "grad_norm": 1.390625,
      "learning_rate": 0.000370665902650794,
      "loss": 4.9471,
      "mean_token_accuracy": 0.21599343717098235,
      "num_tokens": 81431450.0,
      "step": 46935
    },
    {
      "entropy": 5.844062614440918,
      "epoch": 3.652091032872982,
      "grad_norm": 1.40625,
      "learning_rate": 0.00037064081632567675,
      "loss": 5.0143,
      "mean_token_accuracy": 0.20958954691886902,
      "num_tokens": 81440928.0,
      "step": 46940
    },
    {
      "entropy": 5.943365955352784,
      "epoch": 3.6524800622446993,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037061572854936225,
      "loss": 5.0391,
      "mean_token_accuracy": 0.2096440836787224,
      "num_tokens": 81448740.0,
      "step": 46945
    },
    {
      "entropy": 5.874339866638183,
      "epoch": 3.652869091616417,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003705906393222313,
      "loss": 4.9822,
      "mean_token_accuracy": 0.21593944132328033,
      "num_tokens": 81456992.0,
      "step": 46950
    },
    {
      "entropy": 5.806954765319825,
      "epoch": 3.6532581209881347,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003705655486446645,
      "loss": 4.9321,
      "mean_token_accuracy": 0.216282220184803,
      "num_tokens": 81465711.0,
      "step": 46955
    },
    {
      "entropy": 5.908067464828491,
      "epoch": 3.653647150359852,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003705404565170426,
      "loss": 5.0022,
      "mean_token_accuracy": 0.2064366713166237,
      "num_tokens": 81473469.0,
      "step": 46960
    },
    {
      "entropy": 5.933704376220703,
      "epoch": 3.6540361797315697,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003705153629397463,
      "loss": 5.0388,
      "mean_token_accuracy": 0.2017107844352722,
      "num_tokens": 81482984.0,
      "step": 46965
    },
    {
      "entropy": 5.878258800506591,
      "epoch": 3.6544252091032874,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037049026791315646,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21487851142883302,
      "num_tokens": 81491172.0,
      "step": 46970
    },
    {
      "entropy": 5.926511287689209,
      "epoch": 3.654814238475005,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003704651714376538,
      "loss": 5.0114,
      "mean_token_accuracy": 0.2088474377989769,
      "num_tokens": 81499712.0,
      "step": 46975
    },
    {
      "entropy": 5.898883199691772,
      "epoch": 3.6552032678467223,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003704400735136191,
      "loss": 5.0853,
      "mean_token_accuracy": 0.19693916738033296,
      "num_tokens": 81508937.0,
      "step": 46980
    },
    {
      "entropy": 5.84464430809021,
      "epoch": 3.65559229721844,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003704149741414332,
      "loss": 4.9559,
      "mean_token_accuracy": 0.214585842192173,
      "num_tokens": 81517755.0,
      "step": 46985
    },
    {
      "entropy": 5.859436559677124,
      "epoch": 3.6559813265901573,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003703898733214769,
      "loss": 4.9538,
      "mean_token_accuracy": 0.2163550451397896,
      "num_tokens": 81525894.0,
      "step": 46990
    },
    {
      "entropy": 5.9042294979095455,
      "epoch": 3.656370355961875,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003703647710541311,
      "loss": 4.9346,
      "mean_token_accuracy": 0.21389556378126146,
      "num_tokens": 81533907.0,
      "step": 46995
    },
    {
      "entropy": 5.8482812404632565,
      "epoch": 3.6567593853335927,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003703396673397768,
      "loss": 5.0335,
      "mean_token_accuracy": 0.2111420676112175,
      "num_tokens": 81542329.0,
      "step": 47000
    },
    {
      "entropy": 5.778375673294067,
      "epoch": 3.6571484147053104,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003703145621787947,
      "loss": 4.8866,
      "mean_token_accuracy": 0.21735000759363174,
      "num_tokens": 81551746.0,
      "step": 47005
    },
    {
      "entropy": 5.9086973667144775,
      "epoch": 3.657537444077028,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00037028945557156587,
      "loss": 5.001,
      "mean_token_accuracy": 0.21235192716121673,
      "num_tokens": 81561064.0,
      "step": 47010
    },
    {
      "entropy": 5.829688310623169,
      "epoch": 3.6579264734487453,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037026434751847117,
      "loss": 4.893,
      "mean_token_accuracy": 0.223715740442276,
      "num_tokens": 81570808.0,
      "step": 47015
    },
    {
      "entropy": 5.865843439102173,
      "epoch": 3.658315502820463,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037023923801989163,
      "loss": 5.038,
      "mean_token_accuracy": 0.20480707585811614,
      "num_tokens": 81579871.0,
      "step": 47020
    },
    {
      "entropy": 5.962576866149902,
      "epoch": 3.6587045321921803,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003702141270762083,
      "loss": 5.1474,
      "mean_token_accuracy": 0.19772750735282899,
      "num_tokens": 81588422.0,
      "step": 47025
    },
    {
      "entropy": 5.955078840255737,
      "epoch": 3.659093561563898,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037018901468780207,
      "loss": 5.0754,
      "mean_token_accuracy": 0.20600162893533708,
      "num_tokens": 81597026.0,
      "step": 47030
    },
    {
      "entropy": 5.811412000656128,
      "epoch": 3.6594825909356157,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003701639008550541,
      "loss": 4.9574,
      "mean_token_accuracy": 0.21007381826639177,
      "num_tokens": 81605480.0,
      "step": 47035
    },
    {
      "entropy": 5.837076425552368,
      "epoch": 3.6598716203073334,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003701387855783454,
      "loss": 4.892,
      "mean_token_accuracy": 0.21660759299993515,
      "num_tokens": 81613953.0,
      "step": 47040
    },
    {
      "entropy": 5.88428897857666,
      "epoch": 3.6602606496790506,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000370113668858057,
      "loss": 5.0315,
      "mean_token_accuracy": 0.21145493686199188,
      "num_tokens": 81623660.0,
      "step": 47045
    },
    {
      "entropy": 5.865754318237305,
      "epoch": 3.6606496790507683,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003700885506945701,
      "loss": 4.9518,
      "mean_token_accuracy": 0.21455022394657136,
      "num_tokens": 81631920.0,
      "step": 47050
    },
    {
      "entropy": 5.896390819549561,
      "epoch": 3.661038708422486,
      "grad_norm": 1.25,
      "learning_rate": 0.0003700634310882658,
      "loss": 4.9212,
      "mean_token_accuracy": 0.21557882726192473,
      "num_tokens": 81640501.0,
      "step": 47055
    },
    {
      "entropy": 5.85871262550354,
      "epoch": 3.6614277377942033,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037003831003952525,
      "loss": 5.0188,
      "mean_token_accuracy": 0.2048247531056404,
      "num_tokens": 81648494.0,
      "step": 47060
    },
    {
      "entropy": 5.836420679092408,
      "epoch": 3.661816767165921,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00037001318754872957,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20911206901073456,
      "num_tokens": 81657611.0,
      "step": 47065
    },
    {
      "entropy": 5.893690347671509,
      "epoch": 3.6622057965376387,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036998806361626,
      "loss": 4.9582,
      "mean_token_accuracy": 0.21384949684143068,
      "num_tokens": 81666116.0,
      "step": 47070
    },
    {
      "entropy": 5.865044832229614,
      "epoch": 3.6625948259093564,
      "grad_norm": 1.5,
      "learning_rate": 0.0003699629382424979,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21660293340682985,
      "num_tokens": 81674720.0,
      "step": 47075
    },
    {
      "entropy": 5.786578559875489,
      "epoch": 3.6629838552810736,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00036993781142782416,
      "loss": 4.8445,
      "mean_token_accuracy": 0.22837084233760835,
      "num_tokens": 81683086.0,
      "step": 47080
    },
    {
      "entropy": 5.821144533157349,
      "epoch": 3.6633728846527913,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036991268317262033,
      "loss": 4.9559,
      "mean_token_accuracy": 0.21503710150718688,
      "num_tokens": 81691769.0,
      "step": 47085
    },
    {
      "entropy": 5.815629291534424,
      "epoch": 3.663761914024509,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036988755347726766,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2131766065955162,
      "num_tokens": 81700036.0,
      "step": 47090
    },
    {
      "entropy": 5.935163259506226,
      "epoch": 3.6641509433962263,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036986242234214736,
      "loss": 5.0852,
      "mean_token_accuracy": 0.2060905024409294,
      "num_tokens": 81708652.0,
      "step": 47095
    },
    {
      "entropy": 5.865155363082886,
      "epoch": 3.664539972767944,
      "grad_norm": 1.375,
      "learning_rate": 0.00036983728976764074,
      "loss": 4.9257,
      "mean_token_accuracy": 0.21482794731855392,
      "num_tokens": 81717272.0,
      "step": 47100
    },
    {
      "entropy": 5.871768474578857,
      "epoch": 3.6649290021396617,
      "grad_norm": 1.5,
      "learning_rate": 0.00036981215575412923,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22212135493755342,
      "num_tokens": 81726085.0,
      "step": 47105
    },
    {
      "entropy": 5.866903066635132,
      "epoch": 3.6653180315113794,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003697870203019942,
      "loss": 5.0457,
      "mean_token_accuracy": 0.20276347249746324,
      "num_tokens": 81734757.0,
      "step": 47110
    },
    {
      "entropy": 5.873139905929565,
      "epoch": 3.6657070608830966,
      "grad_norm": 1.4375,
      "learning_rate": 0.00036976188341161694,
      "loss": 4.9634,
      "mean_token_accuracy": 0.21564424335956572,
      "num_tokens": 81743419.0,
      "step": 47115
    },
    {
      "entropy": 5.918490076065064,
      "epoch": 3.6660960902548143,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003697367450833789,
      "loss": 5.0695,
      "mean_token_accuracy": 0.20339560210704805,
      "num_tokens": 81752199.0,
      "step": 47120
    },
    {
      "entropy": 5.945154523849487,
      "epoch": 3.6664851196265316,
      "grad_norm": 1.390625,
      "learning_rate": 0.00036971160531766155,
      "loss": 5.0552,
      "mean_token_accuracy": 0.2047775536775589,
      "num_tokens": 81760722.0,
      "step": 47125
    },
    {
      "entropy": 5.9280845642089846,
      "epoch": 3.6668741489982493,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003696864641148465,
      "loss": 5.0851,
      "mean_token_accuracy": 0.2049752652645111,
      "num_tokens": 81769826.0,
      "step": 47130
    },
    {
      "entropy": 5.8825586318969725,
      "epoch": 3.667263178369967,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036966132147531487,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22377275228500365,
      "num_tokens": 81778074.0,
      "step": 47135
    },
    {
      "entropy": 5.91159200668335,
      "epoch": 3.6676522077416847,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036963617739944844,
      "loss": 5.0697,
      "mean_token_accuracy": 0.20271925777196884,
      "num_tokens": 81786621.0,
      "step": 47140
    },
    {
      "entropy": 5.862623977661133,
      "epoch": 3.668041237113402,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036961103188762866,
      "loss": 4.9848,
      "mean_token_accuracy": 0.2099633187055588,
      "num_tokens": 81795660.0,
      "step": 47145
    },
    {
      "entropy": 5.75711874961853,
      "epoch": 3.6684302664851196,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036958588494023705,
      "loss": 4.848,
      "mean_token_accuracy": 0.2137064129114151,
      "num_tokens": 81803850.0,
      "step": 47150
    },
    {
      "entropy": 5.868940353393555,
      "epoch": 3.6688192958568373,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003695607365576551,
      "loss": 4.9531,
      "mean_token_accuracy": 0.21491249054670333,
      "num_tokens": 81813210.0,
      "step": 47155
    },
    {
      "entropy": 5.870437097549439,
      "epoch": 3.6692083252285546,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003695355867402646,
      "loss": 4.987,
      "mean_token_accuracy": 0.21909279525279998,
      "num_tokens": 81821892.0,
      "step": 47160
    },
    {
      "entropy": 5.824654197692871,
      "epoch": 3.6695973546002723,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000369510435488447,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21735740303993226,
      "num_tokens": 81830516.0,
      "step": 47165
    },
    {
      "entropy": 5.862827110290527,
      "epoch": 3.66998638397199,
      "grad_norm": 1.234375,
      "learning_rate": 0.000369485282802584,
      "loss": 5.0382,
      "mean_token_accuracy": 0.20220848321914672,
      "num_tokens": 81839613.0,
      "step": 47170
    },
    {
      "entropy": 5.884251546859741,
      "epoch": 3.6703754133437076,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036946012868305716,
      "loss": 5.0053,
      "mean_token_accuracy": 0.21161291897296905,
      "num_tokens": 81847693.0,
      "step": 47175
    },
    {
      "entropy": 5.875396060943603,
      "epoch": 3.670764442715425,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003694349731302483,
      "loss": 4.9236,
      "mean_token_accuracy": 0.22211882174015046,
      "num_tokens": 81856137.0,
      "step": 47180
    },
    {
      "entropy": 5.835538721084594,
      "epoch": 3.6711534720871426,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003694098161445389,
      "loss": 4.8805,
      "mean_token_accuracy": 0.21363172680139542,
      "num_tokens": 81865097.0,
      "step": 47185
    },
    {
      "entropy": 5.809325408935547,
      "epoch": 3.6715425014588603,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003693846577263109,
      "loss": 4.898,
      "mean_token_accuracy": 0.20695205479860307,
      "num_tokens": 81873923.0,
      "step": 47190
    },
    {
      "entropy": 5.878504848480224,
      "epoch": 3.6719315308305776,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003693594978759459,
      "loss": 4.9892,
      "mean_token_accuracy": 0.20688411295413972,
      "num_tokens": 81882253.0,
      "step": 47195
    },
    {
      "entropy": 5.823017168045044,
      "epoch": 3.6723205602022952,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036933433659382575,
      "loss": 4.91,
      "mean_token_accuracy": 0.20734661370515822,
      "num_tokens": 81890614.0,
      "step": 47200
    },
    {
      "entropy": 5.844015789031983,
      "epoch": 3.672709589574013,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003693091738803322,
      "loss": 4.9345,
      "mean_token_accuracy": 0.2156512752175331,
      "num_tokens": 81899101.0,
      "step": 47205
    },
    {
      "entropy": 5.934416627883911,
      "epoch": 3.6730986189457306,
      "grad_norm": 1.6875,
      "learning_rate": 0.000369284009735847,
      "loss": 5.017,
      "mean_token_accuracy": 0.21558071970939635,
      "num_tokens": 81908454.0,
      "step": 47210
    },
    {
      "entropy": 5.863157081604004,
      "epoch": 3.673487648317448,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003692588441607521,
      "loss": 4.9518,
      "mean_token_accuracy": 0.21059620082378389,
      "num_tokens": 81916963.0,
      "step": 47215
    },
    {
      "entropy": 5.857166051864624,
      "epoch": 3.6738766776891656,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036923367715542923,
      "loss": 5.0397,
      "mean_token_accuracy": 0.20788625478744507,
      "num_tokens": 81925292.0,
      "step": 47220
    },
    {
      "entropy": 5.919726753234864,
      "epoch": 3.674265707060883,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036920850872026025,
      "loss": 5.0039,
      "mean_token_accuracy": 0.21209299117326735,
      "num_tokens": 81934088.0,
      "step": 47225
    },
    {
      "entropy": 5.879449653625488,
      "epoch": 3.6746547364326005,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003691833388556272,
      "loss": 4.9131,
      "mean_token_accuracy": 0.2174004778265953,
      "num_tokens": 81942656.0,
      "step": 47230
    },
    {
      "entropy": 5.9062388896942135,
      "epoch": 3.6750437658043182,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003691581675619118,
      "loss": 5.0339,
      "mean_token_accuracy": 0.21296038031578063,
      "num_tokens": 81951493.0,
      "step": 47235
    },
    {
      "entropy": 5.8033582210540775,
      "epoch": 3.675432795176036,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003691329948394961,
      "loss": 4.8867,
      "mean_token_accuracy": 0.22258873283863068,
      "num_tokens": 81960149.0,
      "step": 47240
    },
    {
      "entropy": 5.965799331665039,
      "epoch": 3.675821824547753,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036910782068876204,
      "loss": 5.0164,
      "mean_token_accuracy": 0.2026522308588028,
      "num_tokens": 81968315.0,
      "step": 47245
    },
    {
      "entropy": 5.874694871902466,
      "epoch": 3.676210853919471,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036908264511009155,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21032766848802567,
      "num_tokens": 81976941.0,
      "step": 47250
    },
    {
      "entropy": 5.887384843826294,
      "epoch": 3.6765998832911886,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003690574681038667,
      "loss": 5.0105,
      "mean_token_accuracy": 0.2090612143278122,
      "num_tokens": 81985763.0,
      "step": 47255
    },
    {
      "entropy": 5.787668371200562,
      "epoch": 3.676988912662906,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036903228967046946,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21922686100006103,
      "num_tokens": 81994551.0,
      "step": 47260
    },
    {
      "entropy": 5.898846578598023,
      "epoch": 3.6773779420346235,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003690071098102819,
      "loss": 5.0024,
      "mean_token_accuracy": 0.21343599408864974,
      "num_tokens": 82002889.0,
      "step": 47265
    },
    {
      "entropy": 5.903883218765259,
      "epoch": 3.6777669714063412,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003689819285236861,
      "loss": 5.0342,
      "mean_token_accuracy": 0.20295630842447282,
      "num_tokens": 82011137.0,
      "step": 47270
    },
    {
      "entropy": 5.9146342277526855,
      "epoch": 3.678156000778059,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036895674581106415,
      "loss": 4.9825,
      "mean_token_accuracy": 0.2116169735789299,
      "num_tokens": 82019857.0,
      "step": 47275
    },
    {
      "entropy": 5.772842121124268,
      "epoch": 3.678545030149776,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036893156167279803,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2138437047600746,
      "num_tokens": 82029407.0,
      "step": 47280
    },
    {
      "entropy": 5.821469449996949,
      "epoch": 3.678934059521494,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003689063761092701,
      "loss": 4.9667,
      "mean_token_accuracy": 0.21591445654630662,
      "num_tokens": 82038086.0,
      "step": 47285
    },
    {
      "entropy": 5.855459451675415,
      "epoch": 3.6793230888932116,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036888118912086225,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22176550030708314,
      "num_tokens": 82046696.0,
      "step": 47290
    },
    {
      "entropy": 5.871962404251098,
      "epoch": 3.679712118264929,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003688560007079569,
      "loss": 5.0101,
      "mean_token_accuracy": 0.2026022955775261,
      "num_tokens": 82056082.0,
      "step": 47295
    },
    {
      "entropy": 5.876474809646607,
      "epoch": 3.6801011476366465,
      "grad_norm": 1.296875,
      "learning_rate": 0.000368830810870936,
      "loss": 4.9616,
      "mean_token_accuracy": 0.2131582498550415,
      "num_tokens": 82065008.0,
      "step": 47300
    },
    {
      "entropy": 5.818769693374634,
      "epoch": 3.6804901770083642,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000368805619610182,
      "loss": 4.9,
      "mean_token_accuracy": 0.21206071525812148,
      "num_tokens": 82073722.0,
      "step": 47305
    },
    {
      "entropy": 5.7744122505187985,
      "epoch": 3.680879206380082,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036878042692607693,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21160048842430115,
      "num_tokens": 82082240.0,
      "step": 47310
    },
    {
      "entropy": 5.857927370071411,
      "epoch": 3.681268235751799,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036875523281900324,
      "loss": 4.9264,
      "mean_token_accuracy": 0.21341080367565154,
      "num_tokens": 82090954.0,
      "step": 47315
    },
    {
      "entropy": 5.903256559371949,
      "epoch": 3.681657265123517,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003687300372893431,
      "loss": 5.0487,
      "mean_token_accuracy": 0.2102670818567276,
      "num_tokens": 82099426.0,
      "step": 47320
    },
    {
      "entropy": 5.957647228240967,
      "epoch": 3.682046294495234,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036870484033747883,
      "loss": 5.0247,
      "mean_token_accuracy": 0.20761010497808458,
      "num_tokens": 82107240.0,
      "step": 47325
    },
    {
      "entropy": 5.946564054489135,
      "epoch": 3.682435323866952,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036867964196379274,
      "loss": 5.0442,
      "mean_token_accuracy": 0.2088195025920868,
      "num_tokens": 82115985.0,
      "step": 47330
    },
    {
      "entropy": 5.900092697143554,
      "epoch": 3.6828243532386695,
      "grad_norm": 1.375,
      "learning_rate": 0.0003686544421686672,
      "loss": 5.0001,
      "mean_token_accuracy": 0.20620049089193343,
      "num_tokens": 82124820.0,
      "step": 47335
    },
    {
      "entropy": 5.820423316955567,
      "epoch": 3.683213382610387,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036862924095248457,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22637704461812974,
      "num_tokens": 82133124.0,
      "step": 47340
    },
    {
      "entropy": 5.917011260986328,
      "epoch": 3.683602411982105,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00036860403831562717,
      "loss": 5.0873,
      "mean_token_accuracy": 0.2039943218231201,
      "num_tokens": 82141524.0,
      "step": 47345
    },
    {
      "entropy": 5.951389026641846,
      "epoch": 3.683991441353822,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003685788342584775,
      "loss": 5.0325,
      "mean_token_accuracy": 0.20540373027324677,
      "num_tokens": 82150068.0,
      "step": 47350
    },
    {
      "entropy": 5.865971231460572,
      "epoch": 3.68438047072554,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000368553628781418,
      "loss": 4.9903,
      "mean_token_accuracy": 0.20933420062065125,
      "num_tokens": 82159642.0,
      "step": 47355
    },
    {
      "entropy": 5.8826268196105955,
      "epoch": 3.684769500097257,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003685284218848311,
      "loss": 4.8981,
      "mean_token_accuracy": 0.21715060770511627,
      "num_tokens": 82168501.0,
      "step": 47360
    },
    {
      "entropy": 5.91005859375,
      "epoch": 3.685158529468975,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036850321356909926,
      "loss": 5.063,
      "mean_token_accuracy": 0.2033192679286003,
      "num_tokens": 82178517.0,
      "step": 47365
    },
    {
      "entropy": 5.885046672821045,
      "epoch": 3.6855475588406925,
      "grad_norm": 1.515625,
      "learning_rate": 0.00036847800383460493,
      "loss": 5.0349,
      "mean_token_accuracy": 0.2118054062128067,
      "num_tokens": 82186271.0,
      "step": 47370
    },
    {
      "entropy": 5.931359386444091,
      "epoch": 3.68593658821241,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036845279268173076,
      "loss": 5.0371,
      "mean_token_accuracy": 0.21168748736381532,
      "num_tokens": 82195731.0,
      "step": 47375
    },
    {
      "entropy": 5.768580770492553,
      "epoch": 3.6863256175841275,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036842758011085914,
      "loss": 4.8757,
      "mean_token_accuracy": 0.21822210252285004,
      "num_tokens": 82204179.0,
      "step": 47380
    },
    {
      "entropy": 5.836758184432983,
      "epoch": 3.686714646955845,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003684023661223727,
      "loss": 4.9807,
      "mean_token_accuracy": 0.21092383116483687,
      "num_tokens": 82212549.0,
      "step": 47385
    },
    {
      "entropy": 5.854185009002686,
      "epoch": 3.687103676327563,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036837715071665403,
      "loss": 4.9212,
      "mean_token_accuracy": 0.21795668601989746,
      "num_tokens": 82221434.0,
      "step": 47390
    },
    {
      "entropy": 5.894934368133545,
      "epoch": 3.68749270569928,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003683519338940857,
      "loss": 4.9621,
      "mean_token_accuracy": 0.2134670913219452,
      "num_tokens": 82230869.0,
      "step": 47395
    },
    {
      "entropy": 5.929989242553711,
      "epoch": 3.687881735070998,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036832671565505045,
      "loss": 4.9918,
      "mean_token_accuracy": 0.2142057552933693,
      "num_tokens": 82239238.0,
      "step": 47400
    },
    {
      "entropy": 5.904805278778076,
      "epoch": 3.6882707644427155,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003683014959999307,
      "loss": 5.0615,
      "mean_token_accuracy": 0.20968099385499955,
      "num_tokens": 82247588.0,
      "step": 47405
    },
    {
      "entropy": 5.900343179702759,
      "epoch": 3.688659793814433,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003682762749291094,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21503131687641144,
      "num_tokens": 82255372.0,
      "step": 47410
    },
    {
      "entropy": 5.895303630828858,
      "epoch": 3.6890488231861505,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000368251052442969,
      "loss": 5.0113,
      "mean_token_accuracy": 0.20724520683288575,
      "num_tokens": 82263399.0,
      "step": 47415
    },
    {
      "entropy": 5.87931489944458,
      "epoch": 3.689437852557868,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036822582854189234,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21244235336780548,
      "num_tokens": 82272394.0,
      "step": 47420
    },
    {
      "entropy": 5.831026411056518,
      "epoch": 3.6898268819295854,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000368200603226262,
      "loss": 4.9457,
      "mean_token_accuracy": 0.21571694910526276,
      "num_tokens": 82282342.0,
      "step": 47425
    },
    {
      "entropy": 5.945184659957886,
      "epoch": 3.690215911301303,
      "grad_norm": 1.28125,
      "learning_rate": 0.000368175376496461,
      "loss": 5.0425,
      "mean_token_accuracy": 0.20774406045675278,
      "num_tokens": 82291727.0,
      "step": 47430
    },
    {
      "entropy": 5.929994201660156,
      "epoch": 3.690604940673021,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003681501483528719,
      "loss": 4.9638,
      "mean_token_accuracy": 0.20847769528627397,
      "num_tokens": 82300606.0,
      "step": 47435
    },
    {
      "entropy": 5.929299592971802,
      "epoch": 3.6909939700447385,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003681249187958776,
      "loss": 4.9924,
      "mean_token_accuracy": 0.21280840188264846,
      "num_tokens": 82309034.0,
      "step": 47440
    },
    {
      "entropy": 5.844507074356079,
      "epoch": 3.691382999416456,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036809968782586087,
      "loss": 4.9232,
      "mean_token_accuracy": 0.21318621784448624,
      "num_tokens": 82317850.0,
      "step": 47445
    },
    {
      "entropy": 5.821482133865357,
      "epoch": 3.6917720287881735,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003680744554432045,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21581053882837295,
      "num_tokens": 82326471.0,
      "step": 47450
    },
    {
      "entropy": 5.9228750705719,
      "epoch": 3.692161058159891,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036804922164829153,
      "loss": 5.0178,
      "mean_token_accuracy": 0.2060135453939438,
      "num_tokens": 82335138.0,
      "step": 47455
    },
    {
      "entropy": 5.893397045135498,
      "epoch": 3.6925500875316084,
      "grad_norm": 1.375,
      "learning_rate": 0.0003680239864415046,
      "loss": 5.011,
      "mean_token_accuracy": 0.21087074428796768,
      "num_tokens": 82344208.0,
      "step": 47460
    },
    {
      "entropy": 5.878235673904419,
      "epoch": 3.692939116903326,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003679987498232268,
      "loss": 4.9948,
      "mean_token_accuracy": 0.21751565337181092,
      "num_tokens": 82353778.0,
      "step": 47465
    },
    {
      "entropy": 5.859918546676636,
      "epoch": 3.693328146275044,
      "grad_norm": 1.375,
      "learning_rate": 0.000367973511793841,
      "loss": 4.9201,
      "mean_token_accuracy": 0.21796105653047562,
      "num_tokens": 82362198.0,
      "step": 47470
    },
    {
      "entropy": 5.916689157485962,
      "epoch": 3.6937171756467615,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036794827235373006,
      "loss": 5.0067,
      "mean_token_accuracy": 0.21473181694746019,
      "num_tokens": 82371274.0,
      "step": 47475
    },
    {
      "entropy": 5.87775673866272,
      "epoch": 3.6941062050184788,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003679230315032772,
      "loss": 5.0131,
      "mean_token_accuracy": 0.2055780529975891,
      "num_tokens": 82380107.0,
      "step": 47480
    },
    {
      "entropy": 5.825700330734253,
      "epoch": 3.6944952343901964,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003678977892428651,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2171643406152725,
      "num_tokens": 82388874.0,
      "step": 47485
    },
    {
      "entropy": 5.948813152313233,
      "epoch": 3.694884263761914,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003678725455728769,
      "loss": 5.0736,
      "mean_token_accuracy": 0.20502327531576156,
      "num_tokens": 82397015.0,
      "step": 47490
    },
    {
      "entropy": 5.866436958312988,
      "epoch": 3.6952732931336314,
      "grad_norm": 1.390625,
      "learning_rate": 0.00036784730049369577,
      "loss": 4.944,
      "mean_token_accuracy": 0.21051363945007323,
      "num_tokens": 82405870.0,
      "step": 47495
    },
    {
      "entropy": 5.898384141921997,
      "epoch": 3.695662322505349,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003678220540057045,
      "loss": 4.9933,
      "mean_token_accuracy": 0.20920102894306183,
      "num_tokens": 82415120.0,
      "step": 47500
    },
    {
      "entropy": 5.821006965637207,
      "epoch": 3.696051351877067,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036779680610928635,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21709294021129608,
      "num_tokens": 82423082.0,
      "step": 47505
    },
    {
      "entropy": 5.841439867019654,
      "epoch": 3.6964403812487845,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003677715568048244,
      "loss": 4.9711,
      "mean_token_accuracy": 0.21107546091079712,
      "num_tokens": 82431250.0,
      "step": 47510
    },
    {
      "entropy": 5.8742344856262205,
      "epoch": 3.6968294106205017,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036774630609270165,
      "loss": 5.0624,
      "mean_token_accuracy": 0.20867359936237334,
      "num_tokens": 82439731.0,
      "step": 47515
    },
    {
      "entropy": 5.928369188308716,
      "epoch": 3.6972184399922194,
      "grad_norm": 1.4375,
      "learning_rate": 0.00036772105397330147,
      "loss": 5.1073,
      "mean_token_accuracy": 0.21032053381204605,
      "num_tokens": 82447808.0,
      "step": 47520
    },
    {
      "entropy": 5.912308311462402,
      "epoch": 3.697607469363937,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003676958004470067,
      "loss": 4.9694,
      "mean_token_accuracy": 0.2006336823105812,
      "num_tokens": 82456030.0,
      "step": 47525
    },
    {
      "entropy": 5.9012779712677,
      "epoch": 3.6979964987356544,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003676705455142009,
      "loss": 4.9431,
      "mean_token_accuracy": 0.20773434787988662,
      "num_tokens": 82465368.0,
      "step": 47530
    },
    {
      "entropy": 5.929051542282105,
      "epoch": 3.698385528107372,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003676452891752669,
      "loss": 5.0192,
      "mean_token_accuracy": 0.21657595336437224,
      "num_tokens": 82473615.0,
      "step": 47535
    },
    {
      "entropy": 5.928440523147583,
      "epoch": 3.69877455747909,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003676200314305882,
      "loss": 5.0005,
      "mean_token_accuracy": 0.21071732342243193,
      "num_tokens": 82482603.0,
      "step": 47540
    },
    {
      "entropy": 5.861611557006836,
      "epoch": 3.6991635868508075,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003675947722805479,
      "loss": 4.98,
      "mean_token_accuracy": 0.2089359000325203,
      "num_tokens": 82491807.0,
      "step": 47545
    },
    {
      "entropy": 5.88281283378601,
      "epoch": 3.6995526162225247,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003675695117255293,
      "loss": 4.9581,
      "mean_token_accuracy": 0.21827838122844695,
      "num_tokens": 82500288.0,
      "step": 47550
    },
    {
      "entropy": 5.8631415367126465,
      "epoch": 3.6999416455942424,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003675442497659157,
      "loss": 4.8998,
      "mean_token_accuracy": 0.21825513988733292,
      "num_tokens": 82509272.0,
      "step": 47555
    },
    {
      "entropy": 5.848896121978759,
      "epoch": 3.7003306749659597,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003675189864020905,
      "loss": 4.8554,
      "mean_token_accuracy": 0.219254207611084,
      "num_tokens": 82518121.0,
      "step": 47560
    },
    {
      "entropy": 5.873148679733276,
      "epoch": 3.7007197043376774,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003674937216344368,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21447642147541046,
      "num_tokens": 82527110.0,
      "step": 47565
    },
    {
      "entropy": 5.851773691177368,
      "epoch": 3.701108733709395,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003674684554633382,
      "loss": 5.0215,
      "mean_token_accuracy": 0.20834087580442429,
      "num_tokens": 82536493.0,
      "step": 47570
    },
    {
      "entropy": 5.89495964050293,
      "epoch": 3.701497763081113,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003674431878891779,
      "loss": 4.9203,
      "mean_token_accuracy": 0.21307066828012466,
      "num_tokens": 82545214.0,
      "step": 47575
    },
    {
      "entropy": 5.891555643081665,
      "epoch": 3.70188679245283,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0003674179189123393,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20723798274993896,
      "num_tokens": 82552947.0,
      "step": 47580
    },
    {
      "entropy": 5.889625453948975,
      "epoch": 3.7022758218245477,
      "grad_norm": 1.4375,
      "learning_rate": 0.00036739264853320606,
      "loss": 4.9951,
      "mean_token_accuracy": 0.21966756582260133,
      "num_tokens": 82560812.0,
      "step": 47585
    },
    {
      "entropy": 5.803754663467407,
      "epoch": 3.7026648511962654,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036736737675216127,
      "loss": 4.9272,
      "mean_token_accuracy": 0.21546991914510727,
      "num_tokens": 82569494.0,
      "step": 47590
    },
    {
      "entropy": 5.88056788444519,
      "epoch": 3.7030538805679827,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003673421035695887,
      "loss": 4.8955,
      "mean_token_accuracy": 0.20866105258464812,
      "num_tokens": 82577194.0,
      "step": 47595
    },
    {
      "entropy": 5.892652273178101,
      "epoch": 3.7034429099397004,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003673168289858716,
      "loss": 4.978,
      "mean_token_accuracy": 0.20659514814615249,
      "num_tokens": 82585652.0,
      "step": 47600
    },
    {
      "entropy": 5.845430660247803,
      "epoch": 3.703831939311418,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003672915530013936,
      "loss": 5.0027,
      "mean_token_accuracy": 0.21350591331720353,
      "num_tokens": 82595044.0,
      "step": 47605
    },
    {
      "entropy": 5.897797060012818,
      "epoch": 3.7042209686831358,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036726627561653815,
      "loss": 5.0273,
      "mean_token_accuracy": 0.20445096492767334,
      "num_tokens": 82603483.0,
      "step": 47610
    },
    {
      "entropy": 5.952613258361817,
      "epoch": 3.704609998054853,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003672409968316888,
      "loss": 5.0802,
      "mean_token_accuracy": 0.20293897539377212,
      "num_tokens": 82611632.0,
      "step": 47615
    },
    {
      "entropy": 5.868357372283936,
      "epoch": 3.7049990274265707,
      "grad_norm": 1.375,
      "learning_rate": 0.00036721571664722914,
      "loss": 4.944,
      "mean_token_accuracy": 0.21880002468824386,
      "num_tokens": 82620595.0,
      "step": 47620
    },
    {
      "entropy": 5.914335441589356,
      "epoch": 3.7053880567982884,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003671904350635428,
      "loss": 5.0635,
      "mean_token_accuracy": 0.20641884803771973,
      "num_tokens": 82630140.0,
      "step": 47625
    },
    {
      "entropy": 5.8944004535675045,
      "epoch": 3.7057770861700057,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036716515208101334,
      "loss": 5.0077,
      "mean_token_accuracy": 0.20012824535369872,
      "num_tokens": 82639361.0,
      "step": 47630
    },
    {
      "entropy": 5.9419232368469235,
      "epoch": 3.7061661155417234,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003671398677000244,
      "loss": 5.0266,
      "mean_token_accuracy": 0.2084161639213562,
      "num_tokens": 82647248.0,
      "step": 47635
    },
    {
      "entropy": 5.913771533966065,
      "epoch": 3.706555144913441,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003671145819209596,
      "loss": 4.9237,
      "mean_token_accuracy": 0.21914731115102767,
      "num_tokens": 82655428.0,
      "step": 47640
    },
    {
      "entropy": 5.8054060459136965,
      "epoch": 3.7069441742851588,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003670892947442026,
      "loss": 4.9082,
      "mean_token_accuracy": 0.21999040395021438,
      "num_tokens": 82664079.0,
      "step": 47645
    },
    {
      "entropy": 5.856548833847046,
      "epoch": 3.707333203656876,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003670640061701371,
      "loss": 5.005,
      "mean_token_accuracy": 0.20887198448181152,
      "num_tokens": 82672557.0,
      "step": 47650
    },
    {
      "entropy": 5.888843679428101,
      "epoch": 3.7077222330285937,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003670387161991469,
      "loss": 4.9831,
      "mean_token_accuracy": 0.21221133470535278,
      "num_tokens": 82681247.0,
      "step": 47655
    },
    {
      "entropy": 5.8657567501068115,
      "epoch": 3.708111262400311,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003670134248316157,
      "loss": 4.8815,
      "mean_token_accuracy": 0.22012775242328644,
      "num_tokens": 82690631.0,
      "step": 47660
    },
    {
      "entropy": 5.934181070327758,
      "epoch": 3.7085002917720287,
      "grad_norm": 1.46875,
      "learning_rate": 0.00036698813206792724,
      "loss": 5.0575,
      "mean_token_accuracy": 0.2045275166630745,
      "num_tokens": 82699244.0,
      "step": 47665
    },
    {
      "entropy": 5.909397745132447,
      "epoch": 3.7088893211437464,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036696283790846524,
      "loss": 4.982,
      "mean_token_accuracy": 0.20795222371816635,
      "num_tokens": 82707630.0,
      "step": 47670
    },
    {
      "entropy": 5.928687810897827,
      "epoch": 3.709278350515464,
      "grad_norm": 1.53125,
      "learning_rate": 0.00036693754235361364,
      "loss": 5.0228,
      "mean_token_accuracy": 0.20884073227643968,
      "num_tokens": 82716853.0,
      "step": 47675
    },
    {
      "entropy": 5.935403108596802,
      "epoch": 3.7096673798871818,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003669122454037561,
      "loss": 5.0022,
      "mean_token_accuracy": 0.20814314633607864,
      "num_tokens": 82725350.0,
      "step": 47680
    },
    {
      "entropy": 5.841417741775513,
      "epoch": 3.710056409258899,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003668869470592765,
      "loss": 4.9889,
      "mean_token_accuracy": 0.22059328407049178,
      "num_tokens": 82734021.0,
      "step": 47685
    },
    {
      "entropy": 5.831270599365235,
      "epoch": 3.7104454386306167,
      "grad_norm": 1.4375,
      "learning_rate": 0.00036686164732055876,
      "loss": 4.9146,
      "mean_token_accuracy": 0.21359071731567383,
      "num_tokens": 82742086.0,
      "step": 47690
    },
    {
      "entropy": 5.816365051269531,
      "epoch": 3.710834468002334,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036683634618798675,
      "loss": 4.8824,
      "mean_token_accuracy": 0.21874499022960664,
      "num_tokens": 82750382.0,
      "step": 47695
    },
    {
      "entropy": 5.905793523788452,
      "epoch": 3.7112234973740517,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003668110436619443,
      "loss": 5.0801,
      "mean_token_accuracy": 0.20578447729349136,
      "num_tokens": 82759118.0,
      "step": 47700
    },
    {
      "entropy": 5.92527494430542,
      "epoch": 3.7116125267457694,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036678573974281546,
      "loss": 5.0646,
      "mean_token_accuracy": 0.2041286736726761,
      "num_tokens": 82767742.0,
      "step": 47705
    },
    {
      "entropy": 5.926243019104004,
      "epoch": 3.712001556117487,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003667604344309841,
      "loss": 5.0432,
      "mean_token_accuracy": 0.20635371506214142,
      "num_tokens": 82776608.0,
      "step": 47710
    },
    {
      "entropy": 5.875054883956909,
      "epoch": 3.7123905854892043,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036673512772683424,
      "loss": 4.8911,
      "mean_token_accuracy": 0.21911421567201614,
      "num_tokens": 82785090.0,
      "step": 47715
    },
    {
      "entropy": 5.915210819244384,
      "epoch": 3.712779614860922,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003667098196307498,
      "loss": 5.0783,
      "mean_token_accuracy": 0.20266338139772416,
      "num_tokens": 82793222.0,
      "step": 47720
    },
    {
      "entropy": 5.966049528121948,
      "epoch": 3.7131686442326397,
      "grad_norm": 1.4375,
      "learning_rate": 0.00036668451014311475,
      "loss": 5.0771,
      "mean_token_accuracy": 0.20689714401960374,
      "num_tokens": 82801991.0,
      "step": 47725
    },
    {
      "entropy": 5.902362632751465,
      "epoch": 3.713557673604357,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00036665919926431325,
      "loss": 4.9645,
      "mean_token_accuracy": 0.2159545212984085,
      "num_tokens": 82809972.0,
      "step": 47730
    },
    {
      "entropy": 5.8952607154846195,
      "epoch": 3.7139467029760747,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003666338869947293,
      "loss": 5.0197,
      "mean_token_accuracy": 0.21077741235494613,
      "num_tokens": 82819244.0,
      "step": 47735
    },
    {
      "entropy": 5.882529306411743,
      "epoch": 3.7143357323477924,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003666085733347469,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21556648313999177,
      "num_tokens": 82827725.0,
      "step": 47740
    },
    {
      "entropy": 5.796825551986695,
      "epoch": 3.71472476171951,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036658325828475033,
      "loss": 4.9029,
      "mean_token_accuracy": 0.22155662029981613,
      "num_tokens": 82835599.0,
      "step": 47745
    },
    {
      "entropy": 5.835478639602661,
      "epoch": 3.7151137910912273,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003665579418451235,
      "loss": 4.907,
      "mean_token_accuracy": 0.2176805019378662,
      "num_tokens": 82844592.0,
      "step": 47750
    },
    {
      "entropy": 5.892356300354004,
      "epoch": 3.715502820462945,
      "grad_norm": 1.25,
      "learning_rate": 0.0003665326240162506,
      "loss": 5.034,
      "mean_token_accuracy": 0.21091899573802947,
      "num_tokens": 82853648.0,
      "step": 47755
    },
    {
      "entropy": 5.952601623535156,
      "epoch": 3.7158918498346623,
      "grad_norm": 2.21875,
      "learning_rate": 0.00036650730479851593,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21267052441835405,
      "num_tokens": 82862328.0,
      "step": 47760
    },
    {
      "entropy": 5.912382888793945,
      "epoch": 3.71628087920638,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036648198419230344,
      "loss": 5.0296,
      "mean_token_accuracy": 0.21655944138765335,
      "num_tokens": 82870870.0,
      "step": 47765
    },
    {
      "entropy": 5.8687700748443605,
      "epoch": 3.7166699085780976,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036645666219799753,
      "loss": 4.9685,
      "mean_token_accuracy": 0.20914117693901063,
      "num_tokens": 82879913.0,
      "step": 47770
    },
    {
      "entropy": 5.921257543563843,
      "epoch": 3.7170589379498153,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003664313388159823,
      "loss": 4.9868,
      "mean_token_accuracy": 0.20879445374011993,
      "num_tokens": 82888821.0,
      "step": 47775
    },
    {
      "entropy": 5.848430204391479,
      "epoch": 3.717447967321533,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000366406014046642,
      "loss": 4.9003,
      "mean_token_accuracy": 0.21648698002099992,
      "num_tokens": 82897539.0,
      "step": 47780
    },
    {
      "entropy": 5.828088760375977,
      "epoch": 3.7178369966932503,
      "grad_norm": 1.390625,
      "learning_rate": 0.000366380687890361,
      "loss": 4.912,
      "mean_token_accuracy": 0.2196787491440773,
      "num_tokens": 82905875.0,
      "step": 47785
    },
    {
      "entropy": 5.757492637634277,
      "epoch": 3.718226026064968,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003663553603475235,
      "loss": 4.8842,
      "mean_token_accuracy": 0.2143015593290329,
      "num_tokens": 82913781.0,
      "step": 47790
    },
    {
      "entropy": 5.840970039367676,
      "epoch": 3.7186150554366852,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036633003141851373,
      "loss": 4.9065,
      "mean_token_accuracy": 0.21258942037820816,
      "num_tokens": 82922688.0,
      "step": 47795
    },
    {
      "entropy": 5.896210765838623,
      "epoch": 3.719004084808403,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003663047011037162,
      "loss": 5.0413,
      "mean_token_accuracy": 0.2105635419487953,
      "num_tokens": 82931682.0,
      "step": 47800
    },
    {
      "entropy": 5.934612846374511,
      "epoch": 3.7193931141801206,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003662793694035151,
      "loss": 5.0093,
      "mean_token_accuracy": 0.20816944986581803,
      "num_tokens": 82939838.0,
      "step": 47805
    },
    {
      "entropy": 5.930421829223633,
      "epoch": 3.7197821435518383,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003662540363182949,
      "loss": 5.0685,
      "mean_token_accuracy": 0.20340142548084258,
      "num_tokens": 82948655.0,
      "step": 47810
    },
    {
      "entropy": 5.9347755908966064,
      "epoch": 3.7201711729235556,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003662287018484398,
      "loss": 5.0467,
      "mean_token_accuracy": 0.2129087671637535,
      "num_tokens": 82957884.0,
      "step": 47815
    },
    {
      "entropy": 5.88123517036438,
      "epoch": 3.7205602022952733,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003662033659943345,
      "loss": 5.0007,
      "mean_token_accuracy": 0.2183639660477638,
      "num_tokens": 82966164.0,
      "step": 47820
    },
    {
      "entropy": 5.899237871170044,
      "epoch": 3.720949231666991,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036617802875636333,
      "loss": 4.9434,
      "mean_token_accuracy": 0.2165793001651764,
      "num_tokens": 82975010.0,
      "step": 47825
    },
    {
      "entropy": 5.931487035751343,
      "epoch": 3.7213382610387082,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003661526901349106,
      "loss": 5.0362,
      "mean_token_accuracy": 0.21135224103927613,
      "num_tokens": 82984302.0,
      "step": 47830
    },
    {
      "entropy": 5.826360082626342,
      "epoch": 3.721727290410426,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036612735013036085,
      "loss": 4.9826,
      "mean_token_accuracy": 0.20848394930362701,
      "num_tokens": 82992886.0,
      "step": 47835
    },
    {
      "entropy": 5.838512182235718,
      "epoch": 3.7221163197821436,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003661020087430987,
      "loss": 4.9641,
      "mean_token_accuracy": 0.21535406708717347,
      "num_tokens": 83000840.0,
      "step": 47840
    },
    {
      "entropy": 5.920995044708252,
      "epoch": 3.7225053491538613,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003660766659735086,
      "loss": 4.9405,
      "mean_token_accuracy": 0.21438690423965454,
      "num_tokens": 83009410.0,
      "step": 47845
    },
    {
      "entropy": 5.84325475692749,
      "epoch": 3.7228943785255786,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003660513218219749,
      "loss": 4.9601,
      "mean_token_accuracy": 0.22165871262550355,
      "num_tokens": 83018738.0,
      "step": 47850
    },
    {
      "entropy": 5.8683864116668705,
      "epoch": 3.7232834078972963,
      "grad_norm": 1.375,
      "learning_rate": 0.0003660259762888824,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20600922554731368,
      "num_tokens": 83027234.0,
      "step": 47855
    },
    {
      "entropy": 5.870380449295044,
      "epoch": 3.7236724372690135,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003660006293746156,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21544401496648788,
      "num_tokens": 83036180.0,
      "step": 47860
    },
    {
      "entropy": 5.947040319442749,
      "epoch": 3.7240614666407312,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036597528107955914,
      "loss": 5.0481,
      "mean_token_accuracy": 0.20773148238658906,
      "num_tokens": 83044301.0,
      "step": 47865
    },
    {
      "entropy": 5.85324182510376,
      "epoch": 3.724450496012449,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036594993140409753,
      "loss": 4.8294,
      "mean_token_accuracy": 0.2251978784799576,
      "num_tokens": 83052814.0,
      "step": 47870
    },
    {
      "entropy": 5.852813816070556,
      "epoch": 3.7248395253841666,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003659245803486155,
      "loss": 5.018,
      "mean_token_accuracy": 0.21338242590427398,
      "num_tokens": 83061695.0,
      "step": 47875
    },
    {
      "entropy": 5.871242427825928,
      "epoch": 3.7252285547558843,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0003658992279134977,
      "loss": 5.1003,
      "mean_token_accuracy": 0.20656393021345137,
      "num_tokens": 83070568.0,
      "step": 47880
    },
    {
      "entropy": 5.91375584602356,
      "epoch": 3.7256175841276016,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003658738740991287,
      "loss": 5.0195,
      "mean_token_accuracy": 0.20435935407876968,
      "num_tokens": 83079254.0,
      "step": 47885
    },
    {
      "entropy": 5.87984299659729,
      "epoch": 3.7260066134993193,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036584851890589335,
      "loss": 4.8735,
      "mean_token_accuracy": 0.21962738186120986,
      "num_tokens": 83087192.0,
      "step": 47890
    },
    {
      "entropy": 5.851999568939209,
      "epoch": 3.7263956428710365,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036582316233417636,
      "loss": 4.9317,
      "mean_token_accuracy": 0.2050440415740013,
      "num_tokens": 83095446.0,
      "step": 47895
    },
    {
      "entropy": 5.890657901763916,
      "epoch": 3.7267846722427542,
      "grad_norm": 1.515625,
      "learning_rate": 0.00036579780438436246,
      "loss": 5.0134,
      "mean_token_accuracy": 0.21194821298122407,
      "num_tokens": 83103242.0,
      "step": 47900
    },
    {
      "entropy": 5.792805528640747,
      "epoch": 3.727173701614472,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003657724450568363,
      "loss": 4.9234,
      "mean_token_accuracy": 0.21826227903366088,
      "num_tokens": 83112691.0,
      "step": 47905
    },
    {
      "entropy": 5.890968179702758,
      "epoch": 3.7275627309861896,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003657470843519829,
      "loss": 4.9975,
      "mean_token_accuracy": 0.21080065667629241,
      "num_tokens": 83120680.0,
      "step": 47910
    },
    {
      "entropy": 5.82760066986084,
      "epoch": 3.727951760357907,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003657217222701868,
      "loss": 4.9614,
      "mean_token_accuracy": 0.21651224195957183,
      "num_tokens": 83128777.0,
      "step": 47915
    },
    {
      "entropy": 5.853410148620606,
      "epoch": 3.7283407897296246,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000365696358811833,
      "loss": 4.983,
      "mean_token_accuracy": 0.2093761071562767,
      "num_tokens": 83136632.0,
      "step": 47920
    },
    {
      "entropy": 5.896419858932495,
      "epoch": 3.7287298191013423,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036567099397730635,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20074321925640107,
      "num_tokens": 83145069.0,
      "step": 47925
    },
    {
      "entropy": 5.852449655532837,
      "epoch": 3.7291188484730595,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003656456277669917,
      "loss": 4.8675,
      "mean_token_accuracy": 0.2201487809419632,
      "num_tokens": 83154484.0,
      "step": 47930
    },
    {
      "entropy": 5.930552625656128,
      "epoch": 3.729507877844777,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036562026018127386,
      "loss": 5.0069,
      "mean_token_accuracy": 0.21395441442728041,
      "num_tokens": 83162899.0,
      "step": 47935
    },
    {
      "entropy": 5.845112037658692,
      "epoch": 3.729896907216495,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036559489122053786,
      "loss": 4.939,
      "mean_token_accuracy": 0.22011471837759017,
      "num_tokens": 83171171.0,
      "step": 47940
    },
    {
      "entropy": 5.826121997833252,
      "epoch": 3.7302859365882126,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003655695208851686,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21325213313102723,
      "num_tokens": 83179139.0,
      "step": 47945
    },
    {
      "entropy": 5.9026415824890135,
      "epoch": 3.73067496595993,
      "grad_norm": 1.375,
      "learning_rate": 0.00036554414917555096,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20919049382209778,
      "num_tokens": 83187729.0,
      "step": 47950
    },
    {
      "entropy": 5.928280019760132,
      "epoch": 3.7310639953316476,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036551877609207,
      "loss": 5.0385,
      "mean_token_accuracy": 0.20397934019565583,
      "num_tokens": 83195930.0,
      "step": 47955
    },
    {
      "entropy": 5.773700904846192,
      "epoch": 3.7314530247033653,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003654934016351107,
      "loss": 4.8877,
      "mean_token_accuracy": 0.21167474389076232,
      "num_tokens": 83203926.0,
      "step": 47960
    },
    {
      "entropy": 5.851754093170166,
      "epoch": 3.7318420540750825,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036546802580505805,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21901919245719909,
      "num_tokens": 83212385.0,
      "step": 47965
    },
    {
      "entropy": 5.857619047164917,
      "epoch": 3.7322310834468,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003654426486022971,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2167314112186432,
      "num_tokens": 83221011.0,
      "step": 47970
    },
    {
      "entropy": 5.883874464035034,
      "epoch": 3.732620112818518,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003654172700272129,
      "loss": 4.9297,
      "mean_token_accuracy": 0.22126694321632384,
      "num_tokens": 83230554.0,
      "step": 47975
    },
    {
      "entropy": 5.856896591186524,
      "epoch": 3.7330091421902356,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003653918900801905,
      "loss": 4.9704,
      "mean_token_accuracy": 0.21048498302698135,
      "num_tokens": 83239959.0,
      "step": 47980
    },
    {
      "entropy": 5.9156165599823,
      "epoch": 3.733398171561953,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036536650876161514,
      "loss": 5.0862,
      "mean_token_accuracy": 0.19858671128749847,
      "num_tokens": 83248345.0,
      "step": 47985
    },
    {
      "entropy": 5.8642051219940186,
      "epoch": 3.7337872009336706,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003653411260718718,
      "loss": 4.899,
      "mean_token_accuracy": 0.2160438358783722,
      "num_tokens": 83256285.0,
      "step": 47990
    },
    {
      "entropy": 5.854768896102906,
      "epoch": 3.734176230305388,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003653157420113457,
      "loss": 4.9405,
      "mean_token_accuracy": 0.2179851844906807,
      "num_tokens": 83264685.0,
      "step": 47995
    },
    {
      "entropy": 5.862247896194458,
      "epoch": 3.7345652596771055,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003652903565804219,
      "loss": 5.0126,
      "mean_token_accuracy": 0.2081531971693039,
      "num_tokens": 83273352.0,
      "step": 48000
    },
    {
      "epoch": 3.7345652596771055,
      "eval_entropy": 5.639032107620313,
      "eval_loss": 5.06563663482666,
      "eval_mean_token_accuracy": 0.2170177726187995,
      "eval_num_tokens": 83273352.0,
      "eval_runtime": 21.8025,
      "eval_samples_per_second": 1906.109,
      "eval_steps_per_second": 238.275,
      "step": 48000
    },
    {
      "entropy": 5.854826259613037,
      "epoch": 3.734954289048823,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036526496977948574,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2165261000394821,
      "num_tokens": 83281787.0,
      "step": 48005
    },
    {
      "entropy": 5.826662063598633,
      "epoch": 3.735343318420541,
      "grad_norm": 1.390625,
      "learning_rate": 0.00036523958160892227,
      "loss": 4.9827,
      "mean_token_accuracy": 0.20684851557016373,
      "num_tokens": 83290367.0,
      "step": 48010
    },
    {
      "entropy": 5.897960090637207,
      "epoch": 3.735732347792258,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036521419206911684,
      "loss": 4.9963,
      "mean_token_accuracy": 0.20624033808708192,
      "num_tokens": 83299094.0,
      "step": 48015
    },
    {
      "entropy": 5.857441091537476,
      "epoch": 3.736121377163976,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003651888011604545,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22256296575069429,
      "num_tokens": 83308270.0,
      "step": 48020
    },
    {
      "entropy": 5.911079597473145,
      "epoch": 3.7365104065356936,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00036516340888332076,
      "loss": 5.0409,
      "mean_token_accuracy": 0.20911362767219543,
      "num_tokens": 83316697.0,
      "step": 48025
    },
    {
      "entropy": 5.875934886932373,
      "epoch": 3.736899435907411,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036513801523810075,
      "loss": 4.9417,
      "mean_token_accuracy": 0.21533966511487962,
      "num_tokens": 83325281.0,
      "step": 48030
    },
    {
      "entropy": 5.908315420150757,
      "epoch": 3.7372884652791285,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003651126202251799,
      "loss": 5.009,
      "mean_token_accuracy": 0.2048704907298088,
      "num_tokens": 83333759.0,
      "step": 48035
    },
    {
      "entropy": 5.883016872406006,
      "epoch": 3.737677494650846,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003650872238449434,
      "loss": 5.0132,
      "mean_token_accuracy": 0.2131052076816559,
      "num_tokens": 83342703.0,
      "step": 48040
    },
    {
      "entropy": 5.921681833267212,
      "epoch": 3.738066524022564,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003650618260977767,
      "loss": 4.9993,
      "mean_token_accuracy": 0.2101897895336151,
      "num_tokens": 83351521.0,
      "step": 48045
    },
    {
      "entropy": 5.933659648895263,
      "epoch": 3.738455553394281,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003650364269840651,
      "loss": 5.0179,
      "mean_token_accuracy": 0.2100882723927498,
      "num_tokens": 83359702.0,
      "step": 48050
    },
    {
      "entropy": 5.913753318786621,
      "epoch": 3.738844582765999,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003650110265041941,
      "loss": 5.0499,
      "mean_token_accuracy": 0.2055484414100647,
      "num_tokens": 83368121.0,
      "step": 48055
    },
    {
      "entropy": 5.877788543701172,
      "epoch": 3.7392336121377165,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036498562465854897,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21223054379224776,
      "num_tokens": 83376742.0,
      "step": 48060
    },
    {
      "entropy": 5.8548308372497555,
      "epoch": 3.739622641509434,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003649602214475152,
      "loss": 4.899,
      "mean_token_accuracy": 0.21868574023246765,
      "num_tokens": 83385825.0,
      "step": 48065
    },
    {
      "entropy": 5.90470323562622,
      "epoch": 3.7400116708811515,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036493481687147836,
      "loss": 5.0024,
      "mean_token_accuracy": 0.21145954579114914,
      "num_tokens": 83395339.0,
      "step": 48070
    },
    {
      "entropy": 5.865393161773682,
      "epoch": 3.740400700252869,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003649094109308237,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20764390975236893,
      "num_tokens": 83404318.0,
      "step": 48075
    },
    {
      "entropy": 5.861928081512451,
      "epoch": 3.740789729624587,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00036488400362593695,
      "loss": 4.9343,
      "mean_token_accuracy": 0.21131043285131454,
      "num_tokens": 83412511.0,
      "step": 48080
    },
    {
      "entropy": 5.843631744384766,
      "epoch": 3.741178758996304,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036485859495720337,
      "loss": 5.0269,
      "mean_token_accuracy": 0.21402496695518494,
      "num_tokens": 83421628.0,
      "step": 48085
    },
    {
      "entropy": 5.886945390701294,
      "epoch": 3.741567788368022,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036483318492500875,
      "loss": 4.9839,
      "mean_token_accuracy": 0.2043114572763443,
      "num_tokens": 83430427.0,
      "step": 48090
    },
    {
      "entropy": 5.874609327316284,
      "epoch": 3.741956817739739,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036480777352973856,
      "loss": 4.961,
      "mean_token_accuracy": 0.21462851464748384,
      "num_tokens": 83440025.0,
      "step": 48095
    },
    {
      "entropy": 5.948618650436401,
      "epoch": 3.742345847111457,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036478236077177825,
      "loss": 5.0157,
      "mean_token_accuracy": 0.20820534080266953,
      "num_tokens": 83448457.0,
      "step": 48100
    },
    {
      "entropy": 5.931408452987671,
      "epoch": 3.7427348764831745,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036475694665151357,
      "loss": 5.0106,
      "mean_token_accuracy": 0.20751720070838928,
      "num_tokens": 83457544.0,
      "step": 48105
    },
    {
      "entropy": 5.854187488555908,
      "epoch": 3.743123905854892,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036473153116933007,
      "loss": 4.8941,
      "mean_token_accuracy": 0.21558720767498016,
      "num_tokens": 83466069.0,
      "step": 48110
    },
    {
      "entropy": 5.87074990272522,
      "epoch": 3.74351293522661,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00036470611432561334,
      "loss": 4.9198,
      "mean_token_accuracy": 0.21550531983375548,
      "num_tokens": 83474847.0,
      "step": 48115
    },
    {
      "entropy": 5.782390689849853,
      "epoch": 3.743901964598327,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003646806961207492,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21232952922582626,
      "num_tokens": 83483337.0,
      "step": 48120
    },
    {
      "entropy": 5.922655248641968,
      "epoch": 3.744290993970045,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036465527655512323,
      "loss": 5.0735,
      "mean_token_accuracy": 0.19859220832586288,
      "num_tokens": 83492403.0,
      "step": 48125
    },
    {
      "entropy": 5.912694072723388,
      "epoch": 3.744680023341762,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003646298556291211,
      "loss": 4.9943,
      "mean_token_accuracy": 0.21157421469688414,
      "num_tokens": 83501555.0,
      "step": 48130
    },
    {
      "entropy": 5.91883487701416,
      "epoch": 3.74506905271348,
      "grad_norm": 1.375,
      "learning_rate": 0.00036460443334312867,
      "loss": 5.0139,
      "mean_token_accuracy": 0.21132491379976273,
      "num_tokens": 83509886.0,
      "step": 48135
    },
    {
      "entropy": 5.9509519100189205,
      "epoch": 3.7454580820851975,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003645790096975315,
      "loss": 5.0074,
      "mean_token_accuracy": 0.21166204959154128,
      "num_tokens": 83518576.0,
      "step": 48140
    },
    {
      "entropy": 5.939409303665161,
      "epoch": 3.745847111456915,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036455358469271546,
      "loss": 5.014,
      "mean_token_accuracy": 0.21124004274606706,
      "num_tokens": 83526757.0,
      "step": 48145
    },
    {
      "entropy": 5.878424024581909,
      "epoch": 3.7462361408286324,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036452815832906625,
      "loss": 4.9774,
      "mean_token_accuracy": 0.2164441466331482,
      "num_tokens": 83536346.0,
      "step": 48150
    },
    {
      "entropy": 5.848426103591919,
      "epoch": 3.74662517020035,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00036450273060696984,
      "loss": 4.8823,
      "mean_token_accuracy": 0.22519754767417907,
      "num_tokens": 83544368.0,
      "step": 48155
    },
    {
      "entropy": 5.871972370147705,
      "epoch": 3.747014199572068,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003644773015268118,
      "loss": 5.0114,
      "mean_token_accuracy": 0.2104874312877655,
      "num_tokens": 83553719.0,
      "step": 48160
    },
    {
      "entropy": 5.926652956008911,
      "epoch": 3.747403228943785,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00036445187108897825,
      "loss": 4.9526,
      "mean_token_accuracy": 0.21617908030748367,
      "num_tokens": 83562180.0,
      "step": 48165
    },
    {
      "entropy": 5.923716115951538,
      "epoch": 3.747792258315503,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003644264392938549,
      "loss": 5.0316,
      "mean_token_accuracy": 0.21308171153068542,
      "num_tokens": 83571255.0,
      "step": 48170
    },
    {
      "entropy": 5.872839498519897,
      "epoch": 3.7481812876872205,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036440100614182775,
      "loss": 5.0,
      "mean_token_accuracy": 0.20843035578727723,
      "num_tokens": 83580352.0,
      "step": 48175
    },
    {
      "entropy": 5.882680892944336,
      "epoch": 3.748570317058938,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036437557163328255,
      "loss": 5.0019,
      "mean_token_accuracy": 0.2109435096383095,
      "num_tokens": 83589132.0,
      "step": 48180
    },
    {
      "entropy": 5.888722610473633,
      "epoch": 3.7489593464306554,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003643501357686053,
      "loss": 5.0064,
      "mean_token_accuracy": 0.21606704145669936,
      "num_tokens": 83597863.0,
      "step": 48185
    },
    {
      "entropy": 5.878688001632691,
      "epoch": 3.749348375802373,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036432469854818194,
      "loss": 4.9431,
      "mean_token_accuracy": 0.20874632596969606,
      "num_tokens": 83606705.0,
      "step": 48190
    },
    {
      "entropy": 5.869173002243042,
      "epoch": 3.7497374051740904,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003642992599723985,
      "loss": 4.9789,
      "mean_token_accuracy": 0.2084087014198303,
      "num_tokens": 83615164.0,
      "step": 48195
    },
    {
      "entropy": 5.837147331237793,
      "epoch": 3.750126434545808,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003642738200416409,
      "loss": 4.9571,
      "mean_token_accuracy": 0.2182650610804558,
      "num_tokens": 83624015.0,
      "step": 48200
    },
    {
      "entropy": 5.912715291976928,
      "epoch": 3.7505154639175258,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036424837875629517,
      "loss": 4.9913,
      "mean_token_accuracy": 0.20875738859176635,
      "num_tokens": 83631946.0,
      "step": 48205
    },
    {
      "entropy": 5.847022867202758,
      "epoch": 3.7509044932892435,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036422293611674736,
      "loss": 4.918,
      "mean_token_accuracy": 0.2164605438709259,
      "num_tokens": 83640279.0,
      "step": 48210
    },
    {
      "entropy": 5.875805330276489,
      "epoch": 3.751293522660961,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036419749212338346,
      "loss": 5.0737,
      "mean_token_accuracy": 0.20834892094135285,
      "num_tokens": 83649275.0,
      "step": 48215
    },
    {
      "entropy": 5.921933889389038,
      "epoch": 3.7516825520326784,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003641720467765897,
      "loss": 4.9661,
      "mean_token_accuracy": 0.21490372866392135,
      "num_tokens": 83657381.0,
      "step": 48220
    },
    {
      "entropy": 5.830944585800171,
      "epoch": 3.752071581404396,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000364146600076752,
      "loss": 4.8729,
      "mean_token_accuracy": 0.21288878917694093,
      "num_tokens": 83665778.0,
      "step": 48225
    },
    {
      "entropy": 5.908101511001587,
      "epoch": 3.7524606107761134,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003641211520242565,
      "loss": 5.0651,
      "mean_token_accuracy": 0.19733507931232452,
      "num_tokens": 83674644.0,
      "step": 48230
    },
    {
      "entropy": 5.982935237884521,
      "epoch": 3.752849640147831,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036409570261948943,
      "loss": 5.0865,
      "mean_token_accuracy": 0.19708358645439147,
      "num_tokens": 83682207.0,
      "step": 48235
    },
    {
      "entropy": 5.867715692520141,
      "epoch": 3.7532386695195488,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003640702518628369,
      "loss": 4.9625,
      "mean_token_accuracy": 0.21439034938812257,
      "num_tokens": 83691176.0,
      "step": 48240
    },
    {
      "entropy": 5.997307968139649,
      "epoch": 3.7536276988912665,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000364044799754685,
      "loss": 5.1168,
      "mean_token_accuracy": 0.19908307492733002,
      "num_tokens": 83699646.0,
      "step": 48245
    },
    {
      "entropy": 5.886437129974365,
      "epoch": 3.7540167282629837,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036401934629542,
      "loss": 4.9902,
      "mean_token_accuracy": 0.21033261418342591,
      "num_tokens": 83708287.0,
      "step": 48250
    },
    {
      "entropy": 5.945775318145752,
      "epoch": 3.7544057576347014,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036399389148542813,
      "loss": 5.105,
      "mean_token_accuracy": 0.1981591612100601,
      "num_tokens": 83716842.0,
      "step": 48255
    },
    {
      "entropy": 5.859292936325073,
      "epoch": 3.754794787006419,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036396843532509563,
      "loss": 4.9659,
      "mean_token_accuracy": 0.21460585445165634,
      "num_tokens": 83725464.0,
      "step": 48260
    },
    {
      "entropy": 5.90838098526001,
      "epoch": 3.7551838163781364,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003639429778148087,
      "loss": 4.9542,
      "mean_token_accuracy": 0.20946005135774612,
      "num_tokens": 83734478.0,
      "step": 48265
    },
    {
      "entropy": 5.884678840637207,
      "epoch": 3.755572845749854,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003639175189549536,
      "loss": 4.9473,
      "mean_token_accuracy": 0.2202048718929291,
      "num_tokens": 83743513.0,
      "step": 48270
    },
    {
      "entropy": 5.914224863052368,
      "epoch": 3.7559618751215718,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036389205874591676,
      "loss": 5.0002,
      "mean_token_accuracy": 0.21169825047254562,
      "num_tokens": 83751483.0,
      "step": 48275
    },
    {
      "entropy": 5.79984073638916,
      "epoch": 3.7563509044932895,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036386659718808437,
      "loss": 4.8206,
      "mean_token_accuracy": 0.2220975711941719,
      "num_tokens": 83759781.0,
      "step": 48280
    },
    {
      "entropy": 5.864936971664429,
      "epoch": 3.7567399338650067,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036384113428184286,
      "loss": 4.9353,
      "mean_token_accuracy": 0.20898450314998626,
      "num_tokens": 83767717.0,
      "step": 48285
    },
    {
      "entropy": 5.8957971096038815,
      "epoch": 3.7571289632367244,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003638156700275784,
      "loss": 4.9942,
      "mean_token_accuracy": 0.21222014129161834,
      "num_tokens": 83776160.0,
      "step": 48290
    },
    {
      "entropy": 5.870040416717529,
      "epoch": 3.757517992608442,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003637902044256777,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21886495798826217,
      "num_tokens": 83784885.0,
      "step": 48295
    },
    {
      "entropy": 5.798266458511352,
      "epoch": 3.7579070219801594,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003637647374765269,
      "loss": 4.9572,
      "mean_token_accuracy": 0.22388071268796922,
      "num_tokens": 83793746.0,
      "step": 48300
    },
    {
      "entropy": 5.858450126647949,
      "epoch": 3.758296051351877,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036373926918051243,
      "loss": 4.9165,
      "mean_token_accuracy": 0.2135024756193161,
      "num_tokens": 83803221.0,
      "step": 48305
    },
    {
      "entropy": 5.910942602157593,
      "epoch": 3.7586850807235948,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036371379953802085,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21292421221733093,
      "num_tokens": 83812973.0,
      "step": 48310
    },
    {
      "entropy": 5.862515783309936,
      "epoch": 3.7590741100953124,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036368832854943856,
      "loss": 4.9346,
      "mean_token_accuracy": 0.2146860510110855,
      "num_tokens": 83821560.0,
      "step": 48315
    },
    {
      "entropy": 5.891303777694702,
      "epoch": 3.7594631394670297,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036366285621515206,
      "loss": 5.0186,
      "mean_token_accuracy": 0.2018510013818741,
      "num_tokens": 83830743.0,
      "step": 48320
    },
    {
      "entropy": 5.936154699325561,
      "epoch": 3.7598521688387474,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003636373825355478,
      "loss": 4.9632,
      "mean_token_accuracy": 0.20942680090665816,
      "num_tokens": 83839400.0,
      "step": 48325
    },
    {
      "entropy": 5.9548423290252686,
      "epoch": 3.7602411982104647,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003636119075110124,
      "loss": 5.1158,
      "mean_token_accuracy": 0.20146631300449372,
      "num_tokens": 83848074.0,
      "step": 48330
    },
    {
      "entropy": 5.843993663787842,
      "epoch": 3.7606302275821823,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003635864311419323,
      "loss": 4.9528,
      "mean_token_accuracy": 0.21656300574541093,
      "num_tokens": 83856471.0,
      "step": 48335
    },
    {
      "entropy": 5.813853979110718,
      "epoch": 3.7610192569539,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036356095342869414,
      "loss": 4.8909,
      "mean_token_accuracy": 0.21711220741271972,
      "num_tokens": 83865300.0,
      "step": 48340
    },
    {
      "entropy": 5.8402234554290775,
      "epoch": 3.7614082863256177,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003635354743716844,
      "loss": 4.9365,
      "mean_token_accuracy": 0.21621186137199402,
      "num_tokens": 83873797.0,
      "step": 48345
    },
    {
      "entropy": 5.860577583312988,
      "epoch": 3.761797315697335,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003635099939712898,
      "loss": 4.9791,
      "mean_token_accuracy": 0.21004964858293534,
      "num_tokens": 83882595.0,
      "step": 48350
    },
    {
      "entropy": 5.88159761428833,
      "epoch": 3.7621863450690527,
      "grad_norm": 1.375,
      "learning_rate": 0.00036348451222789694,
      "loss": 4.8922,
      "mean_token_accuracy": 0.22143671065568923,
      "num_tokens": 83891323.0,
      "step": 48355
    },
    {
      "entropy": 5.889911746978759,
      "epoch": 3.7625753744407704,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036345902914189246,
      "loss": 4.9352,
      "mean_token_accuracy": 0.21332240849733353,
      "num_tokens": 83899761.0,
      "step": 48360
    },
    {
      "entropy": 5.809483385086059,
      "epoch": 3.7629644038124876,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000363433544713663,
      "loss": 4.8776,
      "mean_token_accuracy": 0.2168743446469307,
      "num_tokens": 83908870.0,
      "step": 48365
    },
    {
      "entropy": 5.807506418228149,
      "epoch": 3.7633534331842053,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036340805894359524,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21271784901618956,
      "num_tokens": 83916961.0,
      "step": 48370
    },
    {
      "entropy": 5.896142387390137,
      "epoch": 3.763742462555923,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036338257183207587,
      "loss": 4.9705,
      "mean_token_accuracy": 0.2061152219772339,
      "num_tokens": 83926098.0,
      "step": 48375
    },
    {
      "entropy": 5.808431482315063,
      "epoch": 3.7641314919276407,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036335708337949165,
      "loss": 4.9087,
      "mean_token_accuracy": 0.21584024429321289,
      "num_tokens": 83934903.0,
      "step": 48380
    },
    {
      "entropy": 5.899194812774658,
      "epoch": 3.764520521299358,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003633315935862293,
      "loss": 5.0699,
      "mean_token_accuracy": 0.20630761086940766,
      "num_tokens": 83943743.0,
      "step": 48385
    },
    {
      "entropy": 5.926990509033203,
      "epoch": 3.7649095506710757,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003633061024526757,
      "loss": 5.0544,
      "mean_token_accuracy": 0.2073652297258377,
      "num_tokens": 83952892.0,
      "step": 48390
    },
    {
      "entropy": 5.845371866226197,
      "epoch": 3.7652985800427934,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003632806099792175,
      "loss": 4.9622,
      "mean_token_accuracy": 0.20980707705020904,
      "num_tokens": 83961820.0,
      "step": 48395
    },
    {
      "entropy": 5.862010955810547,
      "epoch": 3.7656876094145106,
      "grad_norm": 1.5,
      "learning_rate": 0.00036325511616624155,
      "loss": 4.9087,
      "mean_token_accuracy": 0.21851401031017303,
      "num_tokens": 83969730.0,
      "step": 48400
    },
    {
      "entropy": 5.898706388473511,
      "epoch": 3.7660766387862283,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003632296210141348,
      "loss": 4.9545,
      "mean_token_accuracy": 0.20416849553585054,
      "num_tokens": 83977740.0,
      "step": 48405
    },
    {
      "entropy": 5.8004320621490475,
      "epoch": 3.766465668157946,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036320412452328383,
      "loss": 4.8517,
      "mean_token_accuracy": 0.22398382872343064,
      "num_tokens": 83986165.0,
      "step": 48410
    },
    {
      "entropy": 5.888893413543701,
      "epoch": 3.7668546975296637,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003631786266940757,
      "loss": 4.9894,
      "mean_token_accuracy": 0.209625805914402,
      "num_tokens": 83994796.0,
      "step": 48415
    },
    {
      "entropy": 5.93151216506958,
      "epoch": 3.767243726901381,
      "grad_norm": 1.609375,
      "learning_rate": 0.00036315312752689726,
      "loss": 4.9589,
      "mean_token_accuracy": 0.20985649526119232,
      "num_tokens": 84003865.0,
      "step": 48420
    },
    {
      "entropy": 5.95916600227356,
      "epoch": 3.7676327562730987,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003631276270221354,
      "loss": 5.0662,
      "mean_token_accuracy": 0.2057712718844414,
      "num_tokens": 84012872.0,
      "step": 48425
    },
    {
      "entropy": 5.8916198253631595,
      "epoch": 3.768021785644816,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036310212518017717,
      "loss": 4.9549,
      "mean_token_accuracy": 0.21782825589179994,
      "num_tokens": 84022162.0,
      "step": 48430
    },
    {
      "entropy": 5.901426029205322,
      "epoch": 3.7684108150165336,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036307662200140935,
      "loss": 4.9989,
      "mean_token_accuracy": 0.20590577870607377,
      "num_tokens": 84030327.0,
      "step": 48435
    },
    {
      "entropy": 5.920516586303711,
      "epoch": 3.7687998443882513,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003630511174862189,
      "loss": 5.0223,
      "mean_token_accuracy": 0.20111789405345917,
      "num_tokens": 84039613.0,
      "step": 48440
    },
    {
      "entropy": 5.935374689102173,
      "epoch": 3.769188873759969,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003630256116349931,
      "loss": 5.0644,
      "mean_token_accuracy": 0.20172394216060638,
      "num_tokens": 84049216.0,
      "step": 48445
    },
    {
      "entropy": 5.875127553939819,
      "epoch": 3.7695779031316863,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003630001044481186,
      "loss": 4.9698,
      "mean_token_accuracy": 0.20724075436592101,
      "num_tokens": 84057734.0,
      "step": 48450
    },
    {
      "entropy": 5.933534717559814,
      "epoch": 3.769966932503404,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00036297459592598256,
      "loss": 5.0358,
      "mean_token_accuracy": 0.2036649242043495,
      "num_tokens": 84065468.0,
      "step": 48455
    },
    {
      "entropy": 5.899046182632446,
      "epoch": 3.7703559618751217,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003629490860689721,
      "loss": 4.9788,
      "mean_token_accuracy": 0.21242628544569014,
      "num_tokens": 84074078.0,
      "step": 48460
    },
    {
      "entropy": 5.86505708694458,
      "epoch": 3.770744991246839,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036292357487747426,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21210132539272308,
      "num_tokens": 84082489.0,
      "step": 48465
    },
    {
      "entropy": 5.869071578979492,
      "epoch": 3.7711340206185566,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036289806235187606,
      "loss": 4.9906,
      "mean_token_accuracy": 0.20853897631168367,
      "num_tokens": 84090993.0,
      "step": 48470
    },
    {
      "entropy": 5.815415382385254,
      "epoch": 3.7715230499902743,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003628725484925647,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21623148620128632,
      "num_tokens": 84099256.0,
      "step": 48475
    },
    {
      "entropy": 5.891503477096558,
      "epoch": 3.771912079361992,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036284703329992734,
      "loss": 4.9992,
      "mean_token_accuracy": 0.21152681857347488,
      "num_tokens": 84108599.0,
      "step": 48480
    },
    {
      "entropy": 5.848405265808106,
      "epoch": 3.7723011087337093,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000362821516774351,
      "loss": 4.8984,
      "mean_token_accuracy": 0.21647050678730012,
      "num_tokens": 84116802.0,
      "step": 48485
    },
    {
      "entropy": 5.867918539047241,
      "epoch": 3.772690138105427,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036279599891622297,
      "loss": 5.0213,
      "mean_token_accuracy": 0.21045306473970413,
      "num_tokens": 84124626.0,
      "step": 48490
    },
    {
      "entropy": 5.872149801254272,
      "epoch": 3.7730791674771447,
      "grad_norm": 1.390625,
      "learning_rate": 0.00036277047972593036,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21770405322313308,
      "num_tokens": 84133795.0,
      "step": 48495
    },
    {
      "entropy": 5.848955345153809,
      "epoch": 3.773468196848862,
      "grad_norm": 1.375,
      "learning_rate": 0.0003627449592038604,
      "loss": 4.9531,
      "mean_token_accuracy": 0.21135829240083695,
      "num_tokens": 84142026.0,
      "step": 48500
    },
    {
      "entropy": 5.907657814025879,
      "epoch": 3.7738572262205796,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003627194373504004,
      "loss": 5.0098,
      "mean_token_accuracy": 0.2095989093184471,
      "num_tokens": 84150398.0,
      "step": 48505
    },
    {
      "entropy": 5.94256820678711,
      "epoch": 3.7742462555922973,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00036269391416593753,
      "loss": 4.9535,
      "mean_token_accuracy": 0.2101661041378975,
      "num_tokens": 84159060.0,
      "step": 48510
    },
    {
      "entropy": 5.905809450149536,
      "epoch": 3.774635284964015,
      "grad_norm": 1.25,
      "learning_rate": 0.0003626683896508591,
      "loss": 4.9622,
      "mean_token_accuracy": 0.213009113073349,
      "num_tokens": 84168577.0,
      "step": 48515
    },
    {
      "entropy": 5.858496379852295,
      "epoch": 3.7750243143357323,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003626428638055524,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21128497570753096,
      "num_tokens": 84177052.0,
      "step": 48520
    },
    {
      "entropy": 5.90947790145874,
      "epoch": 3.77541334370745,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036261733663040473,
      "loss": 5.0355,
      "mean_token_accuracy": 0.20872995704412461,
      "num_tokens": 84185731.0,
      "step": 48525
    },
    {
      "entropy": 5.87296051979065,
      "epoch": 3.775802373079167,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036259180812580347,
      "loss": 4.9802,
      "mean_token_accuracy": 0.2118247225880623,
      "num_tokens": 84194809.0,
      "step": 48530
    },
    {
      "entropy": 5.948032522201538,
      "epoch": 3.776191402450885,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036256627829213585,
      "loss": 5.0489,
      "mean_token_accuracy": 0.20938842743635178,
      "num_tokens": 84203643.0,
      "step": 48535
    },
    {
      "entropy": 5.872265577316284,
      "epoch": 3.7765804318226026,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036254074712978947,
      "loss": 4.8974,
      "mean_token_accuracy": 0.2135225296020508,
      "num_tokens": 84212115.0,
      "step": 48540
    },
    {
      "entropy": 5.89473819732666,
      "epoch": 3.7769694611943203,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00036251521463915144,
      "loss": 4.9518,
      "mean_token_accuracy": 0.2124090701341629,
      "num_tokens": 84220092.0,
      "step": 48545
    },
    {
      "entropy": 5.881883716583252,
      "epoch": 3.777358490566038,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003624896808206093,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21594312638044358,
      "num_tokens": 84228142.0,
      "step": 48550
    },
    {
      "entropy": 5.868704462051392,
      "epoch": 3.7777475199377553,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003624641456745507,
      "loss": 4.9488,
      "mean_token_accuracy": 0.2166791081428528,
      "num_tokens": 84237017.0,
      "step": 48555
    },
    {
      "entropy": 5.9591591358184814,
      "epoch": 3.778136549309473,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036243860920136283,
      "loss": 5.0444,
      "mean_token_accuracy": 0.212117762863636,
      "num_tokens": 84245464.0,
      "step": 48560
    },
    {
      "entropy": 5.903661394119263,
      "epoch": 3.77852557868119,
      "grad_norm": 1.40625,
      "learning_rate": 0.00036241307140143323,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20549351423978807,
      "num_tokens": 84253781.0,
      "step": 48565
    },
    {
      "entropy": 5.895534038543701,
      "epoch": 3.778914608052908,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036238753227514934,
      "loss": 4.9692,
      "mean_token_accuracy": 0.2121538683772087,
      "num_tokens": 84262632.0,
      "step": 48570
    },
    {
      "entropy": 5.861011505126953,
      "epoch": 3.7793036374246256,
      "grad_norm": 1.390625,
      "learning_rate": 0.00036236199182289883,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21718286126852035,
      "num_tokens": 84271244.0,
      "step": 48575
    },
    {
      "entropy": 5.919312906265259,
      "epoch": 3.7796926667963433,
      "grad_norm": 1.46875,
      "learning_rate": 0.000362336450045069,
      "loss": 5.1125,
      "mean_token_accuracy": 0.2056755930185318,
      "num_tokens": 84279343.0,
      "step": 48580
    },
    {
      "entropy": 5.842883586883545,
      "epoch": 3.7800816961680606,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036231090694204763,
      "loss": 4.9333,
      "mean_token_accuracy": 0.22065501660108566,
      "num_tokens": 84288304.0,
      "step": 48585
    },
    {
      "entropy": 5.8491288185119625,
      "epoch": 3.7804707255397783,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036228536251422226,
      "loss": 4.9758,
      "mean_token_accuracy": 0.21005798131227493,
      "num_tokens": 84296844.0,
      "step": 48590
    },
    {
      "entropy": 5.8907819271087645,
      "epoch": 3.780859754911496,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036225981676198045,
      "loss": 4.9356,
      "mean_token_accuracy": 0.21562714129686356,
      "num_tokens": 84305521.0,
      "step": 48595
    },
    {
      "entropy": 5.961500263214111,
      "epoch": 3.781248784283213,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003622342696857098,
      "loss": 5.0857,
      "mean_token_accuracy": 0.20564025938510894,
      "num_tokens": 84314072.0,
      "step": 48600
    },
    {
      "entropy": 5.838432502746582,
      "epoch": 3.781637813654931,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003622087212857979,
      "loss": 4.9782,
      "mean_token_accuracy": 0.20770912766456603,
      "num_tokens": 84323198.0,
      "step": 48605
    },
    {
      "entropy": 5.924706172943115,
      "epoch": 3.7820268430266486,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003621831715626325,
      "loss": 5.0353,
      "mean_token_accuracy": 0.2094405934214592,
      "num_tokens": 84331618.0,
      "step": 48610
    },
    {
      "entropy": 5.803772592544556,
      "epoch": 3.7824158723983663,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036215762051660115,
      "loss": 4.9059,
      "mean_token_accuracy": 0.22132749408483504,
      "num_tokens": 84340043.0,
      "step": 48615
    },
    {
      "entropy": 5.853991889953614,
      "epoch": 3.7828049017700835,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036213206814809175,
      "loss": 4.9635,
      "mean_token_accuracy": 0.21315771639347075,
      "num_tokens": 84348326.0,
      "step": 48620
    },
    {
      "entropy": 5.948745155334473,
      "epoch": 3.7831939311418012,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003621065144574919,
      "loss": 5.044,
      "mean_token_accuracy": 0.20476083606481552,
      "num_tokens": 84357213.0,
      "step": 48625
    },
    {
      "entropy": 5.974864387512207,
      "epoch": 3.7835829605135185,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003620809594451893,
      "loss": 5.0803,
      "mean_token_accuracy": 0.20737089067697526,
      "num_tokens": 84366431.0,
      "step": 48630
    },
    {
      "entropy": 6.006722831726075,
      "epoch": 3.783971989885236,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003620554031115718,
      "loss": 5.0173,
      "mean_token_accuracy": 0.21659037321805955,
      "num_tokens": 84374955.0,
      "step": 48635
    },
    {
      "entropy": 5.886042785644531,
      "epoch": 3.784361019256954,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003620298454570271,
      "loss": 4.9799,
      "mean_token_accuracy": 0.2117963030934334,
      "num_tokens": 84384103.0,
      "step": 48640
    },
    {
      "entropy": 5.9071355819702145,
      "epoch": 3.7847500486286716,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003620042864819429,
      "loss": 5.0636,
      "mean_token_accuracy": 0.20445960462093354,
      "num_tokens": 84393419.0,
      "step": 48645
    },
    {
      "entropy": 5.850815391540527,
      "epoch": 3.7851390780003893,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003619787261867072,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21762387901544572,
      "num_tokens": 84402321.0,
      "step": 48650
    },
    {
      "entropy": 5.885432147979737,
      "epoch": 3.7855281073721065,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003619531645717078,
      "loss": 5.0629,
      "mean_token_accuracy": 0.20222173929214476,
      "num_tokens": 84410943.0,
      "step": 48655
    },
    {
      "entropy": 5.953517103195191,
      "epoch": 3.7859171367438242,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003619276016373325,
      "loss": 4.977,
      "mean_token_accuracy": 0.20801522582769394,
      "num_tokens": 84419256.0,
      "step": 48660
    },
    {
      "entropy": 5.89577808380127,
      "epoch": 3.7863061661155415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003619020373839693,
      "loss": 5.0393,
      "mean_token_accuracy": 0.20747072398662567,
      "num_tokens": 84427511.0,
      "step": 48665
    },
    {
      "entropy": 5.837087202072143,
      "epoch": 3.786695195487259,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003618764718120059,
      "loss": 4.9342,
      "mean_token_accuracy": 0.21127746105194092,
      "num_tokens": 84436491.0,
      "step": 48670
    },
    {
      "entropy": 5.833677577972412,
      "epoch": 3.787084224858977,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003618509049218304,
      "loss": 4.8953,
      "mean_token_accuracy": 0.22256937325000764,
      "num_tokens": 84445152.0,
      "step": 48675
    },
    {
      "entropy": 5.952721071243286,
      "epoch": 3.7874732542306946,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00036182533671383056,
      "loss": 5.0807,
      "mean_token_accuracy": 0.21602991968393326,
      "num_tokens": 84453766.0,
      "step": 48680
    },
    {
      "entropy": 5.788374137878418,
      "epoch": 3.787862283602412,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003617997671883944,
      "loss": 4.8153,
      "mean_token_accuracy": 0.22304548919200898,
      "num_tokens": 84462754.0,
      "step": 48685
    },
    {
      "entropy": 5.863444852828979,
      "epoch": 3.7882513129741295,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00036177419634591,
      "loss": 4.9832,
      "mean_token_accuracy": 0.207102170586586,
      "num_tokens": 84470576.0,
      "step": 48690
    },
    {
      "entropy": 5.848718309402466,
      "epoch": 3.7886403423458472,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003617486241867654,
      "loss": 4.8804,
      "mean_token_accuracy": 0.2252107635140419,
      "num_tokens": 84478616.0,
      "step": 48695
    },
    {
      "entropy": 5.902966117858886,
      "epoch": 3.7890293717175645,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003617230507113483,
      "loss": 5.0024,
      "mean_token_accuracy": 0.20299587100744249,
      "num_tokens": 84487070.0,
      "step": 48700
    },
    {
      "entropy": 5.832431983947754,
      "epoch": 3.789418401089282,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00036169747592004704,
      "loss": 4.9218,
      "mean_token_accuracy": 0.22317652851343156,
      "num_tokens": 84495570.0,
      "step": 48705
    },
    {
      "entropy": 5.910794401168824,
      "epoch": 3.789807430461,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036167189981324956,
      "loss": 5.0342,
      "mean_token_accuracy": 0.2025470644235611,
      "num_tokens": 84505321.0,
      "step": 48710
    },
    {
      "entropy": 5.93018069267273,
      "epoch": 3.7901964598327176,
      "grad_norm": 1.359375,
      "learning_rate": 0.000361646322391344,
      "loss": 4.9952,
      "mean_token_accuracy": 0.216048002243042,
      "num_tokens": 84514107.0,
      "step": 48715
    },
    {
      "entropy": 5.887635135650635,
      "epoch": 3.790585489204435,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003616207436547183,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21245805025100709,
      "num_tokens": 84523394.0,
      "step": 48720
    },
    {
      "entropy": 5.93947114944458,
      "epoch": 3.7909745185761525,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036159516360376084,
      "loss": 5.0252,
      "mean_token_accuracy": 0.2092490091919899,
      "num_tokens": 84532237.0,
      "step": 48725
    },
    {
      "entropy": 5.881946420669555,
      "epoch": 3.7913635479478702,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036156958223885945,
      "loss": 4.9911,
      "mean_token_accuracy": 0.21167093515396118,
      "num_tokens": 84540618.0,
      "step": 48730
    },
    {
      "entropy": 5.865811347961426,
      "epoch": 3.7917525773195875,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003615439995604025,
      "loss": 4.9676,
      "mean_token_accuracy": 0.21578311771154404,
      "num_tokens": 84548951.0,
      "step": 48735
    },
    {
      "entropy": 5.782876014709473,
      "epoch": 3.792141606691305,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003615184155687781,
      "loss": 4.8785,
      "mean_token_accuracy": 0.21204535514116288,
      "num_tokens": 84557631.0,
      "step": 48740
    },
    {
      "entropy": 5.865848875045776,
      "epoch": 3.792530636063023,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036149283026437444,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21726535856723786,
      "num_tokens": 84567188.0,
      "step": 48745
    },
    {
      "entropy": 5.945171451568603,
      "epoch": 3.7929196654347406,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036146724364757984,
      "loss": 4.9767,
      "mean_token_accuracy": 0.21601348519325256,
      "num_tokens": 84575730.0,
      "step": 48750
    },
    {
      "entropy": 5.749672555923462,
      "epoch": 3.793308694806458,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00036144165571878233,
      "loss": 4.7962,
      "mean_token_accuracy": 0.22848126888275147,
      "num_tokens": 84583813.0,
      "step": 48755
    },
    {
      "entropy": 5.747557783126831,
      "epoch": 3.7936977241781755,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003614160664783703,
      "loss": 4.9042,
      "mean_token_accuracy": 0.2184164583683014,
      "num_tokens": 84592868.0,
      "step": 48760
    },
    {
      "entropy": 5.841302013397216,
      "epoch": 3.7940867535498928,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000361390475926732,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21587400138378143,
      "num_tokens": 84601693.0,
      "step": 48765
    },
    {
      "entropy": 5.907858657836914,
      "epoch": 3.7944757829216105,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003613648840642558,
      "loss": 4.9405,
      "mean_token_accuracy": 0.20912210494279862,
      "num_tokens": 84610173.0,
      "step": 48770
    },
    {
      "entropy": 5.866399478912354,
      "epoch": 3.794864812293328,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003613392908913299,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20805558562278748,
      "num_tokens": 84618041.0,
      "step": 48775
    },
    {
      "entropy": 5.876042652130127,
      "epoch": 3.795253841665046,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003613136964083427,
      "loss": 4.9747,
      "mean_token_accuracy": 0.20783669501543045,
      "num_tokens": 84626624.0,
      "step": 48780
    },
    {
      "entropy": 5.808679628372192,
      "epoch": 3.795642871036763,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003612881006156826,
      "loss": 4.8247,
      "mean_token_accuracy": 0.22786216139793397,
      "num_tokens": 84634746.0,
      "step": 48785
    },
    {
      "entropy": 5.896240711212158,
      "epoch": 3.796031900408481,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036126250351373786,
      "loss": 5.0248,
      "mean_token_accuracy": 0.210909503698349,
      "num_tokens": 84643695.0,
      "step": 48790
    },
    {
      "entropy": 5.863849258422851,
      "epoch": 3.7964209297801985,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036123690510289684,
      "loss": 4.9782,
      "mean_token_accuracy": 0.20545587837696075,
      "num_tokens": 84652517.0,
      "step": 48795
    },
    {
      "entropy": 5.9930016040802006,
      "epoch": 3.7968099591519158,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003612113053835482,
      "loss": 5.0768,
      "mean_token_accuracy": 0.19964774698019028,
      "num_tokens": 84660987.0,
      "step": 48800
    },
    {
      "entropy": 5.9639365673065186,
      "epoch": 3.7971989885236335,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003611857043560801,
      "loss": 4.9994,
      "mean_token_accuracy": 0.20921349674463272,
      "num_tokens": 84669776.0,
      "step": 48805
    },
    {
      "entropy": 5.939868783950805,
      "epoch": 3.797588017895351,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003611601020208812,
      "loss": 4.9798,
      "mean_token_accuracy": 0.21237122416496276,
      "num_tokens": 84678456.0,
      "step": 48810
    },
    {
      "entropy": 5.883872985839844,
      "epoch": 3.797977047267069,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003611344983783398,
      "loss": 5.0176,
      "mean_token_accuracy": 0.20332980453968047,
      "num_tokens": 84686493.0,
      "step": 48815
    },
    {
      "entropy": 5.9042047500610355,
      "epoch": 3.798366076638786,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003611088934288445,
      "loss": 4.9786,
      "mean_token_accuracy": 0.21438491493463516,
      "num_tokens": 84695152.0,
      "step": 48820
    },
    {
      "entropy": 5.88040566444397,
      "epoch": 3.798755106010504,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036108328717278387,
      "loss": 5.0049,
      "mean_token_accuracy": 0.21014520823955535,
      "num_tokens": 84704511.0,
      "step": 48825
    },
    {
      "entropy": 5.900492191314697,
      "epoch": 3.7991441353822215,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036105767961054636,
      "loss": 5.035,
      "mean_token_accuracy": 0.2104365274310112,
      "num_tokens": 84713102.0,
      "step": 48830
    },
    {
      "entropy": 5.932271432876587,
      "epoch": 3.7995331647539388,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003610320707425205,
      "loss": 4.9998,
      "mean_token_accuracy": 0.21525168418884277,
      "num_tokens": 84721496.0,
      "step": 48835
    },
    {
      "entropy": 5.945984029769898,
      "epoch": 3.7999221941256565,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000361006460569095,
      "loss": 5.0395,
      "mean_token_accuracy": 0.20839597284793854,
      "num_tokens": 84730069.0,
      "step": 48840
    },
    {
      "entropy": 5.9227526664733885,
      "epoch": 3.800311223497374,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036098084909065814,
      "loss": 4.9372,
      "mean_token_accuracy": 0.2165498897433281,
      "num_tokens": 84738434.0,
      "step": 48845
    },
    {
      "entropy": 5.900686120986938,
      "epoch": 3.800700252869092,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003609552363075989,
      "loss": 4.9304,
      "mean_token_accuracy": 0.2197186291217804,
      "num_tokens": 84747381.0,
      "step": 48850
    },
    {
      "entropy": 5.81136417388916,
      "epoch": 3.801089282240809,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036092962222030573,
      "loss": 4.9164,
      "mean_token_accuracy": 0.21362605690956116,
      "num_tokens": 84756763.0,
      "step": 48855
    },
    {
      "entropy": 5.902033901214599,
      "epoch": 3.801478311612527,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003609040068291673,
      "loss": 5.0193,
      "mean_token_accuracy": 0.2166093945503235,
      "num_tokens": 84765128.0,
      "step": 48860
    },
    {
      "entropy": 5.829157924652099,
      "epoch": 3.801867340984244,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003608783901345724,
      "loss": 4.9252,
      "mean_token_accuracy": 0.20910997688770294,
      "num_tokens": 84773730.0,
      "step": 48865
    },
    {
      "entropy": 5.862717580795288,
      "epoch": 3.8022563703559618,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036085277213690947,
      "loss": 4.8821,
      "mean_token_accuracy": 0.21603417843580247,
      "num_tokens": 84782014.0,
      "step": 48870
    },
    {
      "entropy": 5.928815746307373,
      "epoch": 3.8026453997276795,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003608271528365675,
      "loss": 4.9403,
      "mean_token_accuracy": 0.20917174816131592,
      "num_tokens": 84790789.0,
      "step": 48875
    },
    {
      "entropy": 5.942600393295288,
      "epoch": 3.803034429099397,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000360801532233935,
      "loss": 5.0466,
      "mean_token_accuracy": 0.2093389630317688,
      "num_tokens": 84799765.0,
      "step": 48880
    },
    {
      "entropy": 5.816871404647827,
      "epoch": 3.8034234584711144,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003607759103294008,
      "loss": 4.8837,
      "mean_token_accuracy": 0.21778505742549897,
      "num_tokens": 84808319.0,
      "step": 48885
    },
    {
      "entropy": 5.835119676589966,
      "epoch": 3.803812487842832,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003607502871233538,
      "loss": 4.9475,
      "mean_token_accuracy": 0.21441784650087356,
      "num_tokens": 84817995.0,
      "step": 48890
    },
    {
      "entropy": 5.8853302001953125,
      "epoch": 3.80420151721455,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003607246626161826,
      "loss": 5.0324,
      "mean_token_accuracy": 0.20639730244874954,
      "num_tokens": 84826791.0,
      "step": 48895
    },
    {
      "entropy": 5.912590169906617,
      "epoch": 3.804590546586267,
      "grad_norm": 1.40625,
      "learning_rate": 0.000360699036808276,
      "loss": 4.9204,
      "mean_token_accuracy": 0.20942153930664062,
      "num_tokens": 84835395.0,
      "step": 48900
    },
    {
      "entropy": 5.883716058731079,
      "epoch": 3.8049795759579847,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003606734097000231,
      "loss": 5.0026,
      "mean_token_accuracy": 0.20871864557266234,
      "num_tokens": 84844137.0,
      "step": 48905
    },
    {
      "entropy": 5.873722267150879,
      "epoch": 3.8053686053297024,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003606477812918125,
      "loss": 4.9923,
      "mean_token_accuracy": 0.2078575983643532,
      "num_tokens": 84852773.0,
      "step": 48910
    },
    {
      "entropy": 5.958504724502563,
      "epoch": 3.80575763470142,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003606221515840331,
      "loss": 5.0686,
      "mean_token_accuracy": 0.21206944733858107,
      "num_tokens": 84861437.0,
      "step": 48915
    },
    {
      "entropy": 5.940086078643799,
      "epoch": 3.8061466640731374,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036059652057707394,
      "loss": 5.0214,
      "mean_token_accuracy": 0.21235112398862838,
      "num_tokens": 84870185.0,
      "step": 48920
    },
    {
      "entropy": 5.903927230834961,
      "epoch": 3.806535693444855,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003605708882713237,
      "loss": 5.0065,
      "mean_token_accuracy": 0.21221843510866165,
      "num_tokens": 84878701.0,
      "step": 48925
    },
    {
      "entropy": 5.825904798507691,
      "epoch": 3.806924722816573,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036054525466717145,
      "loss": 4.9359,
      "mean_token_accuracy": 0.21322670727968215,
      "num_tokens": 84887329.0,
      "step": 48930
    },
    {
      "entropy": 5.853927946090698,
      "epoch": 3.80731375218829,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003605196197650062,
      "loss": 4.9294,
      "mean_token_accuracy": 0.21351401805877684,
      "num_tokens": 84895283.0,
      "step": 48935
    },
    {
      "entropy": 5.891686582565308,
      "epoch": 3.8077027815600077,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003604939835652167,
      "loss": 4.9679,
      "mean_token_accuracy": 0.2145231157541275,
      "num_tokens": 84903975.0,
      "step": 48940
    },
    {
      "entropy": 5.882328176498413,
      "epoch": 3.8080918109317254,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036046834606819224,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22422532886266708,
      "num_tokens": 84912244.0,
      "step": 48945
    },
    {
      "entropy": 5.834018850326538,
      "epoch": 3.808480840303443,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003604427072743216,
      "loss": 4.9509,
      "mean_token_accuracy": 0.2149347871541977,
      "num_tokens": 84920824.0,
      "step": 48950
    },
    {
      "entropy": 5.79565658569336,
      "epoch": 3.8088698696751604,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036041706718399386,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20446009188890457,
      "num_tokens": 84929661.0,
      "step": 48955
    },
    {
      "entropy": 5.941824197769165,
      "epoch": 3.809258899046878,
      "grad_norm": 1.46875,
      "learning_rate": 0.000360391425797598,
      "loss": 4.9946,
      "mean_token_accuracy": 0.21400182396173478,
      "num_tokens": 84938368.0,
      "step": 48960
    },
    {
      "entropy": 5.902072620391846,
      "epoch": 3.8096479284185953,
      "grad_norm": 1.375,
      "learning_rate": 0.0003603657831155233,
      "loss": 4.8998,
      "mean_token_accuracy": 0.2152867242693901,
      "num_tokens": 84946826.0,
      "step": 48965
    },
    {
      "entropy": 5.866127920150757,
      "epoch": 3.810036957790313,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003603401391381587,
      "loss": 4.9866,
      "mean_token_accuracy": 0.21702080368995666,
      "num_tokens": 84955164.0,
      "step": 48970
    },
    {
      "entropy": 5.85193190574646,
      "epoch": 3.8104259871620307,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036031449386589326,
      "loss": 4.9725,
      "mean_token_accuracy": 0.21098556816577912,
      "num_tokens": 84963857.0,
      "step": 48975
    },
    {
      "entropy": 5.905687379837036,
      "epoch": 3.8108150165337484,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003602888472991162,
      "loss": 4.9877,
      "mean_token_accuracy": 0.2038487359881401,
      "num_tokens": 84973216.0,
      "step": 48980
    },
    {
      "entropy": 5.91627082824707,
      "epoch": 3.811204045905466,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036026319943821664,
      "loss": 4.9642,
      "mean_token_accuracy": 0.20825836956501007,
      "num_tokens": 84982104.0,
      "step": 48985
    },
    {
      "entropy": 5.9184494972229,
      "epoch": 3.8115930752771834,
      "grad_norm": 1.375,
      "learning_rate": 0.0003602375502835837,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20243929624557494,
      "num_tokens": 84991499.0,
      "step": 48990
    },
    {
      "entropy": 5.897100448608398,
      "epoch": 3.811982104648901,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036021189983560667,
      "loss": 4.9655,
      "mean_token_accuracy": 0.2152772918343544,
      "num_tokens": 85000084.0,
      "step": 48995
    },
    {
      "entropy": 5.904660892486572,
      "epoch": 3.8123711340206183,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036018624809467466,
      "loss": 4.9897,
      "mean_token_accuracy": 0.20659076124429704,
      "num_tokens": 85008936.0,
      "step": 49000
    },
    {
      "entropy": 5.889039707183838,
      "epoch": 3.812760163392336,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036016059506117687,
      "loss": 5.0483,
      "mean_token_accuracy": 0.20299322456121444,
      "num_tokens": 85017483.0,
      "step": 49005
    },
    {
      "entropy": 5.868708705902099,
      "epoch": 3.8131491927640537,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003601349407355027,
      "loss": 4.9389,
      "mean_token_accuracy": 0.2181716874241829,
      "num_tokens": 85026040.0,
      "step": 49010
    },
    {
      "entropy": 5.94707727432251,
      "epoch": 3.8135382221357714,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003601092851180412,
      "loss": 5.0203,
      "mean_token_accuracy": 0.21625827699899675,
      "num_tokens": 85034104.0,
      "step": 49015
    },
    {
      "entropy": 5.850922346115112,
      "epoch": 3.8139272515074887,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036008362820918186,
      "loss": 4.9098,
      "mean_token_accuracy": 0.21770786494016647,
      "num_tokens": 85041399.0,
      "step": 49020
    },
    {
      "entropy": 5.8732133388519285,
      "epoch": 3.8143162808792064,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003600579700093138,
      "loss": 5.0153,
      "mean_token_accuracy": 0.20944169908761978,
      "num_tokens": 85050077.0,
      "step": 49025
    },
    {
      "entropy": 5.862433671951294,
      "epoch": 3.814705310250924,
      "grad_norm": 1.46875,
      "learning_rate": 0.00036003231051882646,
      "loss": 4.9532,
      "mean_token_accuracy": 0.2173406586050987,
      "num_tokens": 85058548.0,
      "step": 49030
    },
    {
      "entropy": 5.994568586349487,
      "epoch": 3.8150943396226413,
      "grad_norm": 1.53125,
      "learning_rate": 0.00036000664973810914,
      "loss": 5.0003,
      "mean_token_accuracy": 0.2112899601459503,
      "num_tokens": 85067290.0,
      "step": 49035
    },
    {
      "entropy": 5.945260000228882,
      "epoch": 3.815483368994359,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003599809876675513,
      "loss": 5.015,
      "mean_token_accuracy": 0.20989690572023392,
      "num_tokens": 85076420.0,
      "step": 49040
    },
    {
      "entropy": 5.888666772842408,
      "epoch": 3.8158723983660767,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035995532430754205,
      "loss": 5.018,
      "mean_token_accuracy": 0.20737213790416717,
      "num_tokens": 85085664.0,
      "step": 49045
    },
    {
      "entropy": 5.940039873123169,
      "epoch": 3.8162614277377944,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003599296596584711,
      "loss": 5.077,
      "mean_token_accuracy": 0.20953987389802933,
      "num_tokens": 85094489.0,
      "step": 49050
    },
    {
      "entropy": 5.896390199661255,
      "epoch": 3.8166504571095117,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003599039937207278,
      "loss": 4.9669,
      "mean_token_accuracy": 0.2164403960108757,
      "num_tokens": 85103105.0,
      "step": 49055
    },
    {
      "entropy": 5.819214725494385,
      "epoch": 3.8170394864812294,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003598783264947015,
      "loss": 4.9044,
      "mean_token_accuracy": 0.21478349566459656,
      "num_tokens": 85110956.0,
      "step": 49060
    },
    {
      "entropy": 5.851478338241577,
      "epoch": 3.8174285158529466,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035985265798078165,
      "loss": 5.012,
      "mean_token_accuracy": 0.21419051438570022,
      "num_tokens": 85119775.0,
      "step": 49065
    },
    {
      "entropy": 5.912068700790405,
      "epoch": 3.8178175452246643,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035982698817935783,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20906738340854644,
      "num_tokens": 85128491.0,
      "step": 49070
    },
    {
      "entropy": 5.916338920593262,
      "epoch": 3.818206574596382,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003598013170908194,
      "loss": 4.9998,
      "mean_token_accuracy": 0.20591468513011932,
      "num_tokens": 85136866.0,
      "step": 49075
    },
    {
      "entropy": 5.8801473617553714,
      "epoch": 3.8185956039680997,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003597756447155559,
      "loss": 4.8954,
      "mean_token_accuracy": 0.21352397054433822,
      "num_tokens": 85145209.0,
      "step": 49080
    },
    {
      "entropy": 5.9180950164794925,
      "epoch": 3.8189846333398174,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00035974997105395705,
      "loss": 5.1569,
      "mean_token_accuracy": 0.19886057823896408,
      "num_tokens": 85154448.0,
      "step": 49085
    },
    {
      "entropy": 5.902649879455566,
      "epoch": 3.8193736627115347,
      "grad_norm": 1.59375,
      "learning_rate": 0.00035972429610641224,
      "loss": 4.9911,
      "mean_token_accuracy": 0.2078552082180977,
      "num_tokens": 85162345.0,
      "step": 49090
    },
    {
      "entropy": 5.931975507736206,
      "epoch": 3.8197626920832524,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035969861987331114,
      "loss": 4.9768,
      "mean_token_accuracy": 0.2181047171354294,
      "num_tokens": 85171297.0,
      "step": 49095
    },
    {
      "entropy": 5.97546420097351,
      "epoch": 3.8201517214549696,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003596729423550433,
      "loss": 4.979,
      "mean_token_accuracy": 0.21391634345054628,
      "num_tokens": 85179509.0,
      "step": 49100
    },
    {
      "entropy": 5.945387649536133,
      "epoch": 3.8205407508266873,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003596472635519983,
      "loss": 5.0447,
      "mean_token_accuracy": 0.20629948377609253,
      "num_tokens": 85188338.0,
      "step": 49105
    },
    {
      "entropy": 5.875040435791016,
      "epoch": 3.820929780198405,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003596215834645657,
      "loss": 4.9477,
      "mean_token_accuracy": 0.21994423121213913,
      "num_tokens": 85196229.0,
      "step": 49110
    },
    {
      "entropy": 5.875749206542968,
      "epoch": 3.8213188095701227,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035959590209313537,
      "loss": 4.9583,
      "mean_token_accuracy": 0.21297667175531387,
      "num_tokens": 85204897.0,
      "step": 49115
    },
    {
      "entropy": 5.885037660598755,
      "epoch": 3.82170783894184,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003595702194380968,
      "loss": 4.9255,
      "mean_token_accuracy": 0.21585139781236648,
      "num_tokens": 85213871.0,
      "step": 49120
    },
    {
      "entropy": 5.909312343597412,
      "epoch": 3.8220968683135577,
      "grad_norm": 1.453125,
      "learning_rate": 0.00035954453549983984,
      "loss": 5.0186,
      "mean_token_accuracy": 0.20311688929796218,
      "num_tokens": 85222573.0,
      "step": 49125
    },
    {
      "entropy": 5.860403203964234,
      "epoch": 3.8224858976852754,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035951885027875406,
      "loss": 4.9778,
      "mean_token_accuracy": 0.2101161390542984,
      "num_tokens": 85230535.0,
      "step": 49130
    },
    {
      "entropy": 5.867007303237915,
      "epoch": 3.8228749270569926,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035949316377522934,
      "loss": 4.9243,
      "mean_token_accuracy": 0.21654145419597626,
      "num_tokens": 85239785.0,
      "step": 49135
    },
    {
      "entropy": 5.82874174118042,
      "epoch": 3.8232639564287103,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003594674759896553,
      "loss": 4.8439,
      "mean_token_accuracy": 0.21349311918020247,
      "num_tokens": 85248262.0,
      "step": 49140
    },
    {
      "entropy": 5.916867017745972,
      "epoch": 3.823652985800428,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003594417869224217,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2099852591753006,
      "num_tokens": 85257647.0,
      "step": 49145
    },
    {
      "entropy": 5.8619428157806395,
      "epoch": 3.8240420151721457,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035941609657391845,
      "loss": 5.0016,
      "mean_token_accuracy": 0.20683867782354354,
      "num_tokens": 85266065.0,
      "step": 49150
    },
    {
      "entropy": 5.882521104812622,
      "epoch": 3.824431044543863,
      "grad_norm": 1.375,
      "learning_rate": 0.0003593904049445353,
      "loss": 5.0542,
      "mean_token_accuracy": 0.2088477209210396,
      "num_tokens": 85274100.0,
      "step": 49155
    },
    {
      "entropy": 5.89699239730835,
      "epoch": 3.8248200739155807,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035936471203466206,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21726315915584565,
      "num_tokens": 85282690.0,
      "step": 49160
    },
    {
      "entropy": 5.901795387268066,
      "epoch": 3.8252091032872984,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003593390178446886,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21812914162874222,
      "num_tokens": 85291136.0,
      "step": 49165
    },
    {
      "entropy": 5.8666411399841305,
      "epoch": 3.8255981326590156,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003593133223750047,
      "loss": 4.9627,
      "mean_token_accuracy": 0.2069456845521927,
      "num_tokens": 85299531.0,
      "step": 49170
    },
    {
      "entropy": 5.865469217300415,
      "epoch": 3.8259871620307333,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003592876256260004,
      "loss": 5.0008,
      "mean_token_accuracy": 0.21047211289405823,
      "num_tokens": 85308975.0,
      "step": 49175
    },
    {
      "entropy": 6.000840711593628,
      "epoch": 3.826376191402451,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003592619275980656,
      "loss": 5.1008,
      "mean_token_accuracy": 0.20032065808773042,
      "num_tokens": 85316878.0,
      "step": 49180
    },
    {
      "entropy": 5.876325464248657,
      "epoch": 3.8267652207741687,
      "grad_norm": 1.375,
      "learning_rate": 0.00035923622829159014,
      "loss": 4.9467,
      "mean_token_accuracy": 0.20984466820955278,
      "num_tokens": 85325330.0,
      "step": 49185
    },
    {
      "entropy": 5.897516870498658,
      "epoch": 3.827154250145886,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003592105277069639,
      "loss": 5.0618,
      "mean_token_accuracy": 0.20819537937641144,
      "num_tokens": 85334137.0,
      "step": 49190
    },
    {
      "entropy": 5.810792112350464,
      "epoch": 3.8275432795176036,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000359184825844577,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2127223029732704,
      "num_tokens": 85342505.0,
      "step": 49195
    },
    {
      "entropy": 5.9118194580078125,
      "epoch": 3.827932308889321,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035915912270481935,
      "loss": 5.0027,
      "mean_token_accuracy": 0.20893150717020034,
      "num_tokens": 85350831.0,
      "step": 49200
    },
    {
      "entropy": 5.941920757293701,
      "epoch": 3.8283213382610386,
      "grad_norm": 1.296875,
      "learning_rate": 0.000359133418288081,
      "loss": 5.0442,
      "mean_token_accuracy": 0.203595632314682,
      "num_tokens": 85360078.0,
      "step": 49205
    },
    {
      "entropy": 5.876425743103027,
      "epoch": 3.8287103676327563,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035910771259475187,
      "loss": 4.9029,
      "mean_token_accuracy": 0.21201393455266954,
      "num_tokens": 85368731.0,
      "step": 49210
    },
    {
      "entropy": 5.901688432693481,
      "epoch": 3.829099397004474,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003590820056252222,
      "loss": 5.0706,
      "mean_token_accuracy": 0.20542028099298476,
      "num_tokens": 85377193.0,
      "step": 49215
    },
    {
      "entropy": 5.8391945362091064,
      "epoch": 3.8294884263761912,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003590562973798818,
      "loss": 4.8881,
      "mean_token_accuracy": 0.2210157707333565,
      "num_tokens": 85386808.0,
      "step": 49220
    },
    {
      "entropy": 5.879534339904785,
      "epoch": 3.829877455747909,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003590305878591209,
      "loss": 4.9904,
      "mean_token_accuracy": 0.21246301382780075,
      "num_tokens": 85394831.0,
      "step": 49225
    },
    {
      "entropy": 5.852814722061157,
      "epoch": 3.8302664851196266,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035900487706332963,
      "loss": 4.9185,
      "mean_token_accuracy": 0.21568588465452193,
      "num_tokens": 85403604.0,
      "step": 49230
    },
    {
      "entropy": 5.943615961074829,
      "epoch": 3.830655514491344,
      "grad_norm": 1.359375,
      "learning_rate": 0.000358979164992898,
      "loss": 5.0148,
      "mean_token_accuracy": 0.2093362793326378,
      "num_tokens": 85412030.0,
      "step": 49235
    },
    {
      "entropy": 5.852046489715576,
      "epoch": 3.8310445438630616,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035895345164821624,
      "loss": 4.9036,
      "mean_token_accuracy": 0.22068717926740647,
      "num_tokens": 85421059.0,
      "step": 49240
    },
    {
      "entropy": 5.842216539382934,
      "epoch": 3.8314335732347793,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003589277370296745,
      "loss": 4.9855,
      "mean_token_accuracy": 0.21445718556642532,
      "num_tokens": 85430782.0,
      "step": 49245
    },
    {
      "entropy": 5.984556245803833,
      "epoch": 3.831822602606497,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035890202113766283,
      "loss": 5.1381,
      "mean_token_accuracy": 0.19815652966499328,
      "num_tokens": 85439749.0,
      "step": 49250
    },
    {
      "entropy": 5.921663427352906,
      "epoch": 3.8322116319782142,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035887630397257167,
      "loss": 5.0331,
      "mean_token_accuracy": 0.2135204181075096,
      "num_tokens": 85449156.0,
      "step": 49255
    },
    {
      "entropy": 5.924413013458252,
      "epoch": 3.832600661349932,
      "grad_norm": 1.390625,
      "learning_rate": 0.000358850585534791,
      "loss": 4.9391,
      "mean_token_accuracy": 0.20843184888362884,
      "num_tokens": 85457218.0,
      "step": 49260
    },
    {
      "entropy": 5.8369842052459715,
      "epoch": 3.8329896907216496,
      "grad_norm": 1.4375,
      "learning_rate": 0.00035882486582471124,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21167470663785934,
      "num_tokens": 85466164.0,
      "step": 49265
    },
    {
      "entropy": 5.8762514114379885,
      "epoch": 3.833378720093367,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003587991448427225,
      "loss": 4.946,
      "mean_token_accuracy": 0.2141008511185646,
      "num_tokens": 85474826.0,
      "step": 49270
    },
    {
      "entropy": 5.886213779449463,
      "epoch": 3.8337677494650846,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035877342258921514,
      "loss": 5.0119,
      "mean_token_accuracy": 0.21271411180496216,
      "num_tokens": 85484251.0,
      "step": 49275
    },
    {
      "entropy": 5.906606388092041,
      "epoch": 3.8341567788368023,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035874769906457936,
      "loss": 4.9933,
      "mean_token_accuracy": 0.20562374144792556,
      "num_tokens": 85492911.0,
      "step": 49280
    },
    {
      "entropy": 6.007534313201904,
      "epoch": 3.83454580820852,
      "grad_norm": 1.484375,
      "learning_rate": 0.00035872197426920567,
      "loss": 5.0437,
      "mean_token_accuracy": 0.2082899108529091,
      "num_tokens": 85501171.0,
      "step": 49285
    },
    {
      "entropy": 5.8936254501342775,
      "epoch": 3.8349348375802372,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003586962482034842,
      "loss": 4.9988,
      "mean_token_accuracy": 0.20873216390609742,
      "num_tokens": 85509277.0,
      "step": 49290
    },
    {
      "entropy": 5.910317754745483,
      "epoch": 3.835323866951955,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00035867052086780544,
      "loss": 5.0698,
      "mean_token_accuracy": 0.20565247982740403,
      "num_tokens": 85518012.0,
      "step": 49295
    },
    {
      "entropy": 5.92736291885376,
      "epoch": 3.835712896323672,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003586447922625596,
      "loss": 4.976,
      "mean_token_accuracy": 0.21612330973148347,
      "num_tokens": 85526750.0,
      "step": 49300
    },
    {
      "entropy": 5.925536346435547,
      "epoch": 3.83610192569539,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035861906238813726,
      "loss": 5.0372,
      "mean_token_accuracy": 0.20125157833099366,
      "num_tokens": 85535535.0,
      "step": 49305
    },
    {
      "entropy": 5.878973579406738,
      "epoch": 3.8364909550671076,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035859333124492866,
      "loss": 4.9407,
      "mean_token_accuracy": 0.21806163489818572,
      "num_tokens": 85543528.0,
      "step": 49310
    },
    {
      "entropy": 5.907936668395996,
      "epoch": 3.8368799844388253,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003585675988333244,
      "loss": 4.9689,
      "mean_token_accuracy": 0.2133386328816414,
      "num_tokens": 85552416.0,
      "step": 49315
    },
    {
      "entropy": 5.899492359161377,
      "epoch": 3.837269013810543,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003585418651537147,
      "loss": 5.011,
      "mean_token_accuracy": 0.21662275940179826,
      "num_tokens": 85561323.0,
      "step": 49320
    },
    {
      "entropy": 5.904488134384155,
      "epoch": 3.8376580431822602,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035851613020649036,
      "loss": 5.0052,
      "mean_token_accuracy": 0.205824775993824,
      "num_tokens": 85570452.0,
      "step": 49325
    },
    {
      "entropy": 5.88805661201477,
      "epoch": 3.838047072553978,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035849039399204157,
      "loss": 4.9615,
      "mean_token_accuracy": 0.2129218637943268,
      "num_tokens": 85579091.0,
      "step": 49330
    },
    {
      "entropy": 5.87508225440979,
      "epoch": 3.838436101925695,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035846465651075897,
      "loss": 5.0227,
      "mean_token_accuracy": 0.2129583939909935,
      "num_tokens": 85587892.0,
      "step": 49335
    },
    {
      "entropy": 5.824540901184082,
      "epoch": 3.838825131297413,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035843891776303304,
      "loss": 4.9594,
      "mean_token_accuracy": 0.21353771835565566,
      "num_tokens": 85596710.0,
      "step": 49340
    },
    {
      "entropy": 5.92581729888916,
      "epoch": 3.8392141606691306,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003584131777492543,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21097322851419448,
      "num_tokens": 85605591.0,
      "step": 49345
    },
    {
      "entropy": 5.910016870498657,
      "epoch": 3.8396031900408483,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003583874364698134,
      "loss": 5.0626,
      "mean_token_accuracy": 0.20946840047836304,
      "num_tokens": 85613988.0,
      "step": 49350
    },
    {
      "entropy": 5.831673192977905,
      "epoch": 3.8399922194125655,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003583616939251009,
      "loss": 4.9699,
      "mean_token_accuracy": 0.20580997169017792,
      "num_tokens": 85622880.0,
      "step": 49355
    },
    {
      "entropy": 5.9092248439788815,
      "epoch": 3.840381248784283,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003583359501155073,
      "loss": 5.0363,
      "mean_token_accuracy": 0.2075486183166504,
      "num_tokens": 85631611.0,
      "step": 49360
    },
    {
      "entropy": 5.922773170471191,
      "epoch": 3.840770278156001,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003583102050414234,
      "loss": 4.9504,
      "mean_token_accuracy": 0.20957036316394806,
      "num_tokens": 85639715.0,
      "step": 49365
    },
    {
      "entropy": 5.893848943710327,
      "epoch": 3.841159307527718,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035828445870323967,
      "loss": 5.0861,
      "mean_token_accuracy": 0.2073373943567276,
      "num_tokens": 85649609.0,
      "step": 49370
    },
    {
      "entropy": 5.879936361312867,
      "epoch": 3.841548336899436,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003582587111013469,
      "loss": 4.9778,
      "mean_token_accuracy": 0.22126031070947647,
      "num_tokens": 85658404.0,
      "step": 49375
    },
    {
      "entropy": 5.897621965408325,
      "epoch": 3.8419373662711536,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003582329622361356,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21554087847471237,
      "num_tokens": 85666682.0,
      "step": 49380
    },
    {
      "entropy": 5.931985139846802,
      "epoch": 3.8423263956428713,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003582072121079966,
      "loss": 5.0808,
      "mean_token_accuracy": 0.2006845772266388,
      "num_tokens": 85675833.0,
      "step": 49385
    },
    {
      "entropy": 5.9050031185150145,
      "epoch": 3.8427154250145885,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035818146071732064,
      "loss": 4.9348,
      "mean_token_accuracy": 0.2118572935461998,
      "num_tokens": 85684449.0,
      "step": 49390
    },
    {
      "entropy": 5.957858943939209,
      "epoch": 3.843104454386306,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035815570806449845,
      "loss": 5.1049,
      "mean_token_accuracy": 0.19879860132932664,
      "num_tokens": 85693164.0,
      "step": 49395
    },
    {
      "entropy": 5.892121362686157,
      "epoch": 3.8434934837580235,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035812995414992075,
      "loss": 5.0564,
      "mean_token_accuracy": 0.20386233925819397,
      "num_tokens": 85701722.0,
      "step": 49400
    },
    {
      "entropy": 6.008479738235474,
      "epoch": 3.843882513129741,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035810419897397823,
      "loss": 5.0731,
      "mean_token_accuracy": 0.19837931990623475,
      "num_tokens": 85710290.0,
      "step": 49405
    },
    {
      "entropy": 5.891609811782837,
      "epoch": 3.844271542501459,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003580784425370618,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2164851114153862,
      "num_tokens": 85719321.0,
      "step": 49410
    },
    {
      "entropy": 5.962700891494751,
      "epoch": 3.8446605718731766,
      "grad_norm": 1.375,
      "learning_rate": 0.0003580526848395622,
      "loss": 5.0789,
      "mean_token_accuracy": 0.20483056157827378,
      "num_tokens": 85728386.0,
      "step": 49415
    },
    {
      "entropy": 5.972732925415039,
      "epoch": 3.8450496012448943,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035802692588187036,
      "loss": 5.0597,
      "mean_token_accuracy": 0.2003506153821945,
      "num_tokens": 85737460.0,
      "step": 49420
    },
    {
      "entropy": 5.929497814178466,
      "epoch": 3.8454386306166115,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000358001165664377,
      "loss": 4.9643,
      "mean_token_accuracy": 0.20923516303300857,
      "num_tokens": 85745732.0,
      "step": 49425
    },
    {
      "entropy": 5.906579160690308,
      "epoch": 3.845827659988329,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003579754041874731,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21690774857997894,
      "num_tokens": 85754180.0,
      "step": 49430
    },
    {
      "entropy": 5.860488367080689,
      "epoch": 3.8462166893600465,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003579496414515495,
      "loss": 4.9804,
      "mean_token_accuracy": 0.21136310547590256,
      "num_tokens": 85764189.0,
      "step": 49435
    },
    {
      "entropy": 5.908385753631592,
      "epoch": 3.846605718731764,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035792387745699715,
      "loss": 4.9181,
      "mean_token_accuracy": 0.21308577060699463,
      "num_tokens": 85773104.0,
      "step": 49440
    },
    {
      "entropy": 5.944346189498901,
      "epoch": 3.846994748103482,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003578981122042069,
      "loss": 4.9794,
      "mean_token_accuracy": 0.2118445411324501,
      "num_tokens": 85781198.0,
      "step": 49445
    },
    {
      "entropy": 5.890374612808228,
      "epoch": 3.8473837774751996,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003578723456935698,
      "loss": 4.9497,
      "mean_token_accuracy": 0.20916337966918946,
      "num_tokens": 85789662.0,
      "step": 49450
    },
    {
      "entropy": 5.811501979827881,
      "epoch": 3.847772806846917,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003578465779254768,
      "loss": 4.8571,
      "mean_token_accuracy": 0.2253691166639328,
      "num_tokens": 85797758.0,
      "step": 49455
    },
    {
      "entropy": 5.87333345413208,
      "epoch": 3.8481618362186345,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035782080890031876,
      "loss": 4.9291,
      "mean_token_accuracy": 0.2186008647084236,
      "num_tokens": 85806310.0,
      "step": 49460
    },
    {
      "entropy": 5.875651264190674,
      "epoch": 3.848550865590352,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035779503861848677,
      "loss": 4.9518,
      "mean_token_accuracy": 0.21422866731882095,
      "num_tokens": 85815340.0,
      "step": 49465
    },
    {
      "entropy": 5.91192626953125,
      "epoch": 3.8489398949620695,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003577692670803719,
      "loss": 4.986,
      "mean_token_accuracy": 0.211429663002491,
      "num_tokens": 85824700.0,
      "step": 49470
    },
    {
      "entropy": 5.926465845108032,
      "epoch": 3.849328924333787,
      "grad_norm": 1.5,
      "learning_rate": 0.0003577434942863651,
      "loss": 4.9885,
      "mean_token_accuracy": 0.21352363526821136,
      "num_tokens": 85832710.0,
      "step": 49475
    },
    {
      "entropy": 5.850472021102905,
      "epoch": 3.849717953705505,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003577177202368575,
      "loss": 5.012,
      "mean_token_accuracy": 0.20687387585639955,
      "num_tokens": 85841359.0,
      "step": 49480
    },
    {
      "entropy": 5.851738739013672,
      "epoch": 3.8501069830772225,
      "grad_norm": 1.25,
      "learning_rate": 0.00035769194493224014,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20448113828897477,
      "num_tokens": 85850877.0,
      "step": 49485
    },
    {
      "entropy": 5.961606121063232,
      "epoch": 3.85049601244894,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035766616837290413,
      "loss": 5.0023,
      "mean_token_accuracy": 0.207453890144825,
      "num_tokens": 85859140.0,
      "step": 49490
    },
    {
      "entropy": 5.948023128509521,
      "epoch": 3.8508850418206575,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003576403905592406,
      "loss": 5.0326,
      "mean_token_accuracy": 0.21148218512535094,
      "num_tokens": 85867039.0,
      "step": 49495
    },
    {
      "entropy": 5.8297224044799805,
      "epoch": 3.8512740711923747,
      "grad_norm": 1.484375,
      "learning_rate": 0.00035761461149164066,
      "loss": 4.9088,
      "mean_token_accuracy": 0.21660172045230866,
      "num_tokens": 85875199.0,
      "step": 49500
    },
    {
      "entropy": 5.947691202163696,
      "epoch": 3.8516631005640924,
      "grad_norm": 1.453125,
      "learning_rate": 0.00035758883117049543,
      "loss": 5.0227,
      "mean_token_accuracy": 0.21306429505348207,
      "num_tokens": 85883937.0,
      "step": 49505
    },
    {
      "entropy": 5.851103019714356,
      "epoch": 3.85205212993581,
      "grad_norm": 1.25,
      "learning_rate": 0.0003575630495961962,
      "loss": 4.9085,
      "mean_token_accuracy": 0.21646764278411865,
      "num_tokens": 85892750.0,
      "step": 49510
    },
    {
      "entropy": 5.847301387786866,
      "epoch": 3.852441159307528,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035753726676913413,
      "loss": 4.9351,
      "mean_token_accuracy": 0.21430059373378754,
      "num_tokens": 85900967.0,
      "step": 49515
    },
    {
      "entropy": 5.871849822998047,
      "epoch": 3.8528301886792455,
      "grad_norm": 1.53125,
      "learning_rate": 0.00035751148268970046,
      "loss": 5.0326,
      "mean_token_accuracy": 0.2041352793574333,
      "num_tokens": 85908953.0,
      "step": 49520
    },
    {
      "entropy": 5.901280403137207,
      "epoch": 3.853219218050963,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003574856973582863,
      "loss": 4.9046,
      "mean_token_accuracy": 0.21914848685264587,
      "num_tokens": 85916754.0,
      "step": 49525
    },
    {
      "entropy": 5.860062217712402,
      "epoch": 3.8536082474226805,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035745991077528296,
      "loss": 4.907,
      "mean_token_accuracy": 0.21508288383483887,
      "num_tokens": 85925338.0,
      "step": 49530
    },
    {
      "entropy": 5.8786091804504395,
      "epoch": 3.8539972767943977,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00035743412294108175,
      "loss": 4.937,
      "mean_token_accuracy": 0.21505581587553024,
      "num_tokens": 85933053.0,
      "step": 49535
    },
    {
      "entropy": 5.961914110183716,
      "epoch": 3.8543863061661154,
      "grad_norm": 1.375,
      "learning_rate": 0.0003574083338560739,
      "loss": 5.0817,
      "mean_token_accuracy": 0.20651623904705046,
      "num_tokens": 85942552.0,
      "step": 49540
    },
    {
      "entropy": 5.871679353713989,
      "epoch": 3.854775335537833,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035738254352065074,
      "loss": 4.8736,
      "mean_token_accuracy": 0.21851248890161515,
      "num_tokens": 85950843.0,
      "step": 49545
    },
    {
      "entropy": 5.894000482559204,
      "epoch": 3.855164364909551,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035735675193520364,
      "loss": 4.9703,
      "mean_token_accuracy": 0.2138999357819557,
      "num_tokens": 85959383.0,
      "step": 49550
    },
    {
      "entropy": 5.846316909790039,
      "epoch": 3.855553394281268,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003573309591001239,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2154393747448921,
      "num_tokens": 85967134.0,
      "step": 49555
    },
    {
      "entropy": 5.915923118591309,
      "epoch": 3.855942423652986,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000357305165015803,
      "loss": 5.0077,
      "mean_token_accuracy": 0.21036537736654282,
      "num_tokens": 85975921.0,
      "step": 49560
    },
    {
      "entropy": 5.903443145751953,
      "epoch": 3.8563314530247035,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035727936968263206,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20756752789020538,
      "num_tokens": 85984447.0,
      "step": 49565
    },
    {
      "entropy": 5.928441381454467,
      "epoch": 3.8567204823964207,
      "grad_norm": 1.375,
      "learning_rate": 0.0003572535731010028,
      "loss": 5.0355,
      "mean_token_accuracy": 0.2065972164273262,
      "num_tokens": 85993548.0,
      "step": 49570
    },
    {
      "entropy": 5.960633373260498,
      "epoch": 3.8571095117681384,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003572277752713063,
      "loss": 5.0686,
      "mean_token_accuracy": 0.20573433339595795,
      "num_tokens": 86002471.0,
      "step": 49575
    },
    {
      "entropy": 5.9383544921875,
      "epoch": 3.857498541139856,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003572019761939343,
      "loss": 4.9775,
      "mean_token_accuracy": 0.20848557054996492,
      "num_tokens": 86010488.0,
      "step": 49580
    },
    {
      "entropy": 5.875814151763916,
      "epoch": 3.857887570511574,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035717617586927814,
      "loss": 4.9704,
      "mean_token_accuracy": 0.21251550912857056,
      "num_tokens": 86019686.0,
      "step": 49585
    },
    {
      "entropy": 5.93630838394165,
      "epoch": 3.858276599883291,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035715037429772935,
      "loss": 4.9843,
      "mean_token_accuracy": 0.21156545281410216,
      "num_tokens": 86028510.0,
      "step": 49590
    },
    {
      "entropy": 5.8833996772766115,
      "epoch": 3.8586656292550088,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003571245714796793,
      "loss": 4.9284,
      "mean_token_accuracy": 0.21667902916669846,
      "num_tokens": 86038029.0,
      "step": 49595
    },
    {
      "entropy": 5.860598230361939,
      "epoch": 3.8590546586267265,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003570987674155197,
      "loss": 4.9847,
      "mean_token_accuracy": 0.2169323444366455,
      "num_tokens": 86046530.0,
      "step": 49600
    },
    {
      "entropy": 5.819795417785644,
      "epoch": 3.8594436879984437,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003570729621056419,
      "loss": 4.8926,
      "mean_token_accuracy": 0.22292157858610154,
      "num_tokens": 86055502.0,
      "step": 49605
    },
    {
      "entropy": 5.9052674770355225,
      "epoch": 3.8598327173701614,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00035704715555043753,
      "loss": 4.9722,
      "mean_token_accuracy": 0.21512884348630906,
      "num_tokens": 86063357.0,
      "step": 49610
    },
    {
      "entropy": 5.862415218353272,
      "epoch": 3.860221746741879,
      "grad_norm": 1.5,
      "learning_rate": 0.00035702134775029813,
      "loss": 4.9404,
      "mean_token_accuracy": 0.21755334883928298,
      "num_tokens": 86071507.0,
      "step": 49615
    },
    {
      "entropy": 5.79955472946167,
      "epoch": 3.860610776113597,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003569955387056154,
      "loss": 4.8667,
      "mean_token_accuracy": 0.21954558342695235,
      "num_tokens": 86080456.0,
      "step": 49620
    },
    {
      "entropy": 5.848525524139404,
      "epoch": 3.860999805485314,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035696972841678094,
      "loss": 4.9027,
      "mean_token_accuracy": 0.21932505816221237,
      "num_tokens": 86088933.0,
      "step": 49625
    },
    {
      "entropy": 5.948590087890625,
      "epoch": 3.8613888348570318,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003569439168841862,
      "loss": 4.9727,
      "mean_token_accuracy": 0.2153840497136116,
      "num_tokens": 86097035.0,
      "step": 49630
    },
    {
      "entropy": 5.826184511184692,
      "epoch": 3.861777864228749,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035691810410822303,
      "loss": 4.8917,
      "mean_token_accuracy": 0.2149326503276825,
      "num_tokens": 86105898.0,
      "step": 49635
    },
    {
      "entropy": 5.929840469360352,
      "epoch": 3.8621668936004667,
      "grad_norm": 1.421875,
      "learning_rate": 0.00035689229008928304,
      "loss": 5.002,
      "mean_token_accuracy": 0.21724387705326081,
      "num_tokens": 86113757.0,
      "step": 49640
    },
    {
      "entropy": 5.886704635620117,
      "epoch": 3.8625559229721844,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035686647482775774,
      "loss": 5.0092,
      "mean_token_accuracy": 0.20430192053318025,
      "num_tokens": 86122487.0,
      "step": 49645
    },
    {
      "entropy": 6.024584054946899,
      "epoch": 3.862944952343902,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035684065832403903,
      "loss": 5.0343,
      "mean_token_accuracy": 0.21021942794322968,
      "num_tokens": 86132146.0,
      "step": 49650
    },
    {
      "entropy": 5.858136510848999,
      "epoch": 3.8633339817156194,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003568148405785187,
      "loss": 4.9057,
      "mean_token_accuracy": 0.21554238349199295,
      "num_tokens": 86140247.0,
      "step": 49655
    },
    {
      "entropy": 5.88458662033081,
      "epoch": 3.863723011087337,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003567890215915883,
      "loss": 5.0282,
      "mean_token_accuracy": 0.20507986545562745,
      "num_tokens": 86148410.0,
      "step": 49660
    },
    {
      "entropy": 5.840918445587159,
      "epoch": 3.8641120404590548,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035676320136363975,
      "loss": 4.9228,
      "mean_token_accuracy": 0.2198563814163208,
      "num_tokens": 86157089.0,
      "step": 49665
    },
    {
      "entropy": 5.967732954025268,
      "epoch": 3.864501069830772,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003567373798950647,
      "loss": 5.0042,
      "mean_token_accuracy": 0.20341504365205765,
      "num_tokens": 86165071.0,
      "step": 49670
    },
    {
      "entropy": 5.903130912780762,
      "epoch": 3.8648900992024897,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035671155718625504,
      "loss": 4.9538,
      "mean_token_accuracy": 0.22221267521381377,
      "num_tokens": 86173685.0,
      "step": 49675
    },
    {
      "entropy": 5.875367498397827,
      "epoch": 3.8652791285742074,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003566857332376025,
      "loss": 4.9736,
      "mean_token_accuracy": 0.2106228932738304,
      "num_tokens": 86181642.0,
      "step": 49680
    },
    {
      "entropy": 5.948966836929321,
      "epoch": 3.865668157945925,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035665990804949904,
      "loss": 5.1017,
      "mean_token_accuracy": 0.2040363222360611,
      "num_tokens": 86190261.0,
      "step": 49685
    },
    {
      "entropy": 6.020456266403198,
      "epoch": 3.8660571873176424,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003566340816223364,
      "loss": 5.1097,
      "mean_token_accuracy": 0.19796894937753678,
      "num_tokens": 86200086.0,
      "step": 49690
    },
    {
      "entropy": 5.994291639328003,
      "epoch": 3.86644621668936,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003566082539565066,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20669254064559936,
      "num_tokens": 86209312.0,
      "step": 49695
    },
    {
      "entropy": 5.9707887172698975,
      "epoch": 3.8668352460610778,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003565824250524013,
      "loss": 5.0436,
      "mean_token_accuracy": 0.2056431621313095,
      "num_tokens": 86218762.0,
      "step": 49700
    },
    {
      "entropy": 5.939906454086303,
      "epoch": 3.867224275432795,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035655659491041266,
      "loss": 5.1041,
      "mean_token_accuracy": 0.20547030866146088,
      "num_tokens": 86227478.0,
      "step": 49705
    },
    {
      "entropy": 5.900994682312012,
      "epoch": 3.8676133048045127,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003565307635309325,
      "loss": 5.039,
      "mean_token_accuracy": 0.2042345032095909,
      "num_tokens": 86236122.0,
      "step": 49710
    },
    {
      "entropy": 5.939699125289917,
      "epoch": 3.8680023341762304,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003565049309143527,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20409068316221238,
      "num_tokens": 86244838.0,
      "step": 49715
    },
    {
      "entropy": 5.944669008255005,
      "epoch": 3.868391363547948,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035647909706106533,
      "loss": 5.004,
      "mean_token_accuracy": 0.20713644921779634,
      "num_tokens": 86253362.0,
      "step": 49720
    },
    {
      "entropy": 5.899145078659058,
      "epoch": 3.8687803929196654,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003564532619714624,
      "loss": 4.966,
      "mean_token_accuracy": 0.21391140520572663,
      "num_tokens": 86262337.0,
      "step": 49725
    },
    {
      "entropy": 5.990558195114136,
      "epoch": 3.869169422291383,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003564274256459358,
      "loss": 5.0884,
      "mean_token_accuracy": 0.2038117155432701,
      "num_tokens": 86271021.0,
      "step": 49730
    },
    {
      "entropy": 5.8866363048553465,
      "epoch": 3.8695584516631003,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003564015880848777,
      "loss": 4.9432,
      "mean_token_accuracy": 0.21734942197799684,
      "num_tokens": 86279561.0,
      "step": 49735
    },
    {
      "entropy": 5.836661338806152,
      "epoch": 3.869947481034818,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035637574928867996,
      "loss": 4.8956,
      "mean_token_accuracy": 0.21963123828172684,
      "num_tokens": 86289358.0,
      "step": 49740
    },
    {
      "entropy": 5.905851221084594,
      "epoch": 3.8703365104065357,
      "grad_norm": 1.46875,
      "learning_rate": 0.00035634990925773475,
      "loss": 4.9846,
      "mean_token_accuracy": 0.21490618288517,
      "num_tokens": 86297758.0,
      "step": 49745
    },
    {
      "entropy": 5.845110511779785,
      "epoch": 3.8707255397782534,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003563240679924342,
      "loss": 4.9205,
      "mean_token_accuracy": 0.21522506028413774,
      "num_tokens": 86306214.0,
      "step": 49750
    },
    {
      "entropy": 5.9187065124511715,
      "epoch": 3.871114569149971,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003562982254931704,
      "loss": 5.0192,
      "mean_token_accuracy": 0.21234747171401977,
      "num_tokens": 86314249.0,
      "step": 49755
    },
    {
      "entropy": 5.915904760360718,
      "epoch": 3.8715035985216883,
      "grad_norm": 1.375,
      "learning_rate": 0.0003562723817603354,
      "loss": 5.0004,
      "mean_token_accuracy": 0.21233959943056108,
      "num_tokens": 86322516.0,
      "step": 49760
    },
    {
      "entropy": 6.021694707870483,
      "epoch": 3.871892627893406,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035624653679432135,
      "loss": 5.1096,
      "mean_token_accuracy": 0.20661191046237945,
      "num_tokens": 86331637.0,
      "step": 49765
    },
    {
      "entropy": 5.933902549743652,
      "epoch": 3.8722816572651233,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035622069059552036,
      "loss": 5.0023,
      "mean_token_accuracy": 0.20758727043867112,
      "num_tokens": 86340147.0,
      "step": 49770
    },
    {
      "entropy": 5.892926979064941,
      "epoch": 3.872670686636841,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003561948431643247,
      "loss": 4.9737,
      "mean_token_accuracy": 0.21016692817211152,
      "num_tokens": 86348599.0,
      "step": 49775
    },
    {
      "entropy": 5.897628927230835,
      "epoch": 3.8730597160085587,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00035616899450112655,
      "loss": 5.0188,
      "mean_token_accuracy": 0.21717392206192015,
      "num_tokens": 86357409.0,
      "step": 49780
    },
    {
      "entropy": 5.884993076324463,
      "epoch": 3.8734487453802764,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035614314460631814,
      "loss": 4.9217,
      "mean_token_accuracy": 0.21625805050134658,
      "num_tokens": 86365826.0,
      "step": 49785
    },
    {
      "entropy": 5.89710431098938,
      "epoch": 3.8738377747519936,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003561172934802917,
      "loss": 5.0253,
      "mean_token_accuracy": 0.21099531799554824,
      "num_tokens": 86373784.0,
      "step": 49790
    },
    {
      "entropy": 5.871261119842529,
      "epoch": 3.8742268041237113,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003560914411234393,
      "loss": 4.9663,
      "mean_token_accuracy": 0.21572889983654023,
      "num_tokens": 86382761.0,
      "step": 49795
    },
    {
      "entropy": 5.920084667205811,
      "epoch": 3.874615833495429,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003560655875361535,
      "loss": 4.9804,
      "mean_token_accuracy": 0.20598232746124268,
      "num_tokens": 86391025.0,
      "step": 49800
    },
    {
      "entropy": 5.87377552986145,
      "epoch": 3.8750048628671463,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003560397327188263,
      "loss": 5.0142,
      "mean_token_accuracy": 0.20639852583408355,
      "num_tokens": 86399802.0,
      "step": 49805
    },
    {
      "entropy": 5.827319955825805,
      "epoch": 3.875393892238864,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003560138766718502,
      "loss": 4.8484,
      "mean_token_accuracy": 0.22213265895843506,
      "num_tokens": 86407942.0,
      "step": 49810
    },
    {
      "entropy": 5.837137508392334,
      "epoch": 3.8757829216105817,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035598801939561755,
      "loss": 4.9391,
      "mean_token_accuracy": 0.21369633376598357,
      "num_tokens": 86416399.0,
      "step": 49815
    },
    {
      "entropy": 5.87867579460144,
      "epoch": 3.8761719509822994,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035596216089052045,
      "loss": 4.9506,
      "mean_token_accuracy": 0.2113206535577774,
      "num_tokens": 86424325.0,
      "step": 49820
    },
    {
      "entropy": 5.900385951995849,
      "epoch": 3.8765609803540166,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035593630115695154,
      "loss": 5.008,
      "mean_token_accuracy": 0.21024041175842284,
      "num_tokens": 86433023.0,
      "step": 49825
    },
    {
      "entropy": 5.853997087478637,
      "epoch": 3.8769500097257343,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003559104401953031,
      "loss": 4.9341,
      "mean_token_accuracy": 0.22106249183416365,
      "num_tokens": 86442156.0,
      "step": 49830
    },
    {
      "entropy": 5.920306062698364,
      "epoch": 3.8773390390974516,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035588457800596747,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21778531670570372,
      "num_tokens": 86450884.0,
      "step": 49835
    },
    {
      "entropy": 5.894975709915161,
      "epoch": 3.8777280684691693,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003558587145893371,
      "loss": 5.0816,
      "mean_token_accuracy": 0.20594456642866135,
      "num_tokens": 86460509.0,
      "step": 49840
    },
    {
      "entropy": 5.957233333587647,
      "epoch": 3.878117097840887,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003558328499458044,
      "loss": 5.1085,
      "mean_token_accuracy": 0.20111797451972963,
      "num_tokens": 86468547.0,
      "step": 49845
    },
    {
      "entropy": 5.93181004524231,
      "epoch": 3.8785061272126047,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003558069840757619,
      "loss": 5.0106,
      "mean_token_accuracy": 0.20987920463085175,
      "num_tokens": 86477208.0,
      "step": 49850
    },
    {
      "entropy": 5.913388586044311,
      "epoch": 3.8788951565843224,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000355781116979602,
      "loss": 5.024,
      "mean_token_accuracy": 0.2138962045311928,
      "num_tokens": 86485825.0,
      "step": 49855
    },
    {
      "entropy": 5.910169982910157,
      "epoch": 3.8792841859560396,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003557552486577173,
      "loss": 5.031,
      "mean_token_accuracy": 0.21134063899517058,
      "num_tokens": 86493364.0,
      "step": 49860
    },
    {
      "entropy": 5.915397310256958,
      "epoch": 3.8796732153277573,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003557293791105002,
      "loss": 4.982,
      "mean_token_accuracy": 0.21194352060556412,
      "num_tokens": 86502641.0,
      "step": 49865
    },
    {
      "entropy": 5.868005180358887,
      "epoch": 3.8800622446994746,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035570350833834334,
      "loss": 4.9342,
      "mean_token_accuracy": 0.2168838545680046,
      "num_tokens": 86511523.0,
      "step": 49870
    },
    {
      "entropy": 5.907673978805542,
      "epoch": 3.8804512740711923,
      "grad_norm": 1.453125,
      "learning_rate": 0.00035567763634163904,
      "loss": 4.9766,
      "mean_token_accuracy": 0.21375919580459596,
      "num_tokens": 86519782.0,
      "step": 49875
    },
    {
      "entropy": 5.930201768875122,
      "epoch": 3.88084030344291,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003556517631207802,
      "loss": 5.0145,
      "mean_token_accuracy": 0.21050166785717012,
      "num_tokens": 86529102.0,
      "step": 49880
    },
    {
      "entropy": 5.949878692626953,
      "epoch": 3.8812293328146277,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035562588867615907,
      "loss": 5.0029,
      "mean_token_accuracy": 0.20914242863655091,
      "num_tokens": 86538145.0,
      "step": 49885
    },
    {
      "entropy": 5.90383882522583,
      "epoch": 3.881618362186345,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00035560001300816846,
      "loss": 5.0438,
      "mean_token_accuracy": 0.2104927822947502,
      "num_tokens": 86547268.0,
      "step": 49890
    },
    {
      "entropy": 5.929309368133545,
      "epoch": 3.8820073915580626,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035557413611720095,
      "loss": 4.9855,
      "mean_token_accuracy": 0.21054494231939316,
      "num_tokens": 86555676.0,
      "step": 49895
    },
    {
      "entropy": 5.919291353225708,
      "epoch": 3.8823964209297803,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00035554825800364917,
      "loss": 4.9999,
      "mean_token_accuracy": 0.2058631807565689,
      "num_tokens": 86564056.0,
      "step": 49900
    },
    {
      "entropy": 5.942048501968384,
      "epoch": 3.8827854503014976,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003555223786679058,
      "loss": 5.0042,
      "mean_token_accuracy": 0.20885255187749863,
      "num_tokens": 86571909.0,
      "step": 49905
    },
    {
      "entropy": 5.930173730850219,
      "epoch": 3.8831744796732153,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035549649811036346,
      "loss": 5.098,
      "mean_token_accuracy": 0.20814307779073715,
      "num_tokens": 86580772.0,
      "step": 49910
    },
    {
      "entropy": 5.912869691848755,
      "epoch": 3.883563509044933,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035547061633141487,
      "loss": 5.0805,
      "mean_token_accuracy": 0.2074701428413391,
      "num_tokens": 86590018.0,
      "step": 49915
    },
    {
      "entropy": 6.0084434986114506,
      "epoch": 3.8839525384166507,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00035544473333145275,
      "loss": 5.0725,
      "mean_token_accuracy": 0.21657355278730392,
      "num_tokens": 86597702.0,
      "step": 49920
    },
    {
      "entropy": 5.910910606384277,
      "epoch": 3.884341567788368,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003554188491108699,
      "loss": 4.9432,
      "mean_token_accuracy": 0.21158004105091094,
      "num_tokens": 86605780.0,
      "step": 49925
    },
    {
      "entropy": 5.915832281112671,
      "epoch": 3.8847305971600856,
      "grad_norm": 1.375,
      "learning_rate": 0.000355392963670059,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21747543811798095,
      "num_tokens": 86614449.0,
      "step": 49930
    },
    {
      "entropy": 5.981594133377075,
      "epoch": 3.8851196265318033,
      "grad_norm": 1.375,
      "learning_rate": 0.0003553670770094129,
      "loss": 5.0483,
      "mean_token_accuracy": 0.20831324011087418,
      "num_tokens": 86622715.0,
      "step": 49935
    },
    {
      "entropy": 5.870710563659668,
      "epoch": 3.8855086559035206,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035534118912932425,
      "loss": 4.9487,
      "mean_token_accuracy": 0.21014446765184402,
      "num_tokens": 86631007.0,
      "step": 49940
    },
    {
      "entropy": 5.8792259216308596,
      "epoch": 3.8858976852752383,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035531530003018584,
      "loss": 4.9097,
      "mean_token_accuracy": 0.21683760732412338,
      "num_tokens": 86639079.0,
      "step": 49945
    },
    {
      "entropy": 5.941952896118164,
      "epoch": 3.886286714646956,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003552894097123907,
      "loss": 5.0238,
      "mean_token_accuracy": 0.2089035302400589,
      "num_tokens": 86647793.0,
      "step": 49950
    },
    {
      "entropy": 5.890746021270752,
      "epoch": 3.8866757440186737,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035526351817633157,
      "loss": 4.955,
      "mean_token_accuracy": 0.21389715373516083,
      "num_tokens": 86656321.0,
      "step": 49955
    },
    {
      "entropy": 5.843081855773926,
      "epoch": 3.887064773390391,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003552376254224013,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21543222814798355,
      "num_tokens": 86665295.0,
      "step": 49960
    },
    {
      "entropy": 5.859136486053467,
      "epoch": 3.8874538027621086,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003552117314509928,
      "loss": 4.8955,
      "mean_token_accuracy": 0.21749415546655654,
      "num_tokens": 86674229.0,
      "step": 49965
    },
    {
      "entropy": 5.936794567108154,
      "epoch": 3.887842832133826,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003551858362624989,
      "loss": 5.0569,
      "mean_token_accuracy": 0.20553871244192123,
      "num_tokens": 86682362.0,
      "step": 49970
    },
    {
      "entropy": 5.939830923080445,
      "epoch": 3.8882318615055436,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003551599398573125,
      "loss": 5.0401,
      "mean_token_accuracy": 0.20618510097265244,
      "num_tokens": 86690670.0,
      "step": 49975
    },
    {
      "entropy": 5.871878242492675,
      "epoch": 3.8886208908772613,
      "grad_norm": 1.375,
      "learning_rate": 0.00035513404223582666,
      "loss": 4.9364,
      "mean_token_accuracy": 0.2122965008020401,
      "num_tokens": 86698861.0,
      "step": 49980
    },
    {
      "entropy": 5.866643476486206,
      "epoch": 3.889009920248979,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035510814339843427,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21246618628501893,
      "num_tokens": 86706866.0,
      "step": 49985
    },
    {
      "entropy": 5.9182273864746096,
      "epoch": 3.889398949620696,
      "grad_norm": 1.46875,
      "learning_rate": 0.00035508224334552835,
      "loss": 5.0231,
      "mean_token_accuracy": 0.21047684997320176,
      "num_tokens": 86715313.0,
      "step": 49990
    },
    {
      "entropy": 5.80032868385315,
      "epoch": 3.889787978992414,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003550563420775018,
      "loss": 4.8842,
      "mean_token_accuracy": 0.21967146545648575,
      "num_tokens": 86724926.0,
      "step": 49995
    },
    {
      "entropy": 5.87828950881958,
      "epoch": 3.8901770083641316,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035503043959474775,
      "loss": 4.826,
      "mean_token_accuracy": 0.2227575346827507,
      "num_tokens": 86732645.0,
      "step": 50000
    },
    {
      "entropy": 5.862406301498413,
      "epoch": 3.890566037735849,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035500453589765906,
      "loss": 4.948,
      "mean_token_accuracy": 0.21635289043188094,
      "num_tokens": 86741351.0,
      "step": 50005
    },
    {
      "entropy": 5.903558826446533,
      "epoch": 3.8909550671075666,
      "grad_norm": 1.375,
      "learning_rate": 0.00035497863098662893,
      "loss": 4.9487,
      "mean_token_accuracy": 0.21577460318803787,
      "num_tokens": 86749220.0,
      "step": 50010
    },
    {
      "entropy": 5.893824863433838,
      "epoch": 3.8913440964792843,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035495272486205033,
      "loss": 4.9711,
      "mean_token_accuracy": 0.21402205228805543,
      "num_tokens": 86757006.0,
      "step": 50015
    },
    {
      "entropy": 5.923338842391968,
      "epoch": 3.891733125851002,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003549268175243165,
      "loss": 5.0237,
      "mean_token_accuracy": 0.2077975243330002,
      "num_tokens": 86765447.0,
      "step": 50020
    },
    {
      "entropy": 5.792822551727295,
      "epoch": 3.892122155222719,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003549009089738204,
      "loss": 4.8894,
      "mean_token_accuracy": 0.22218640744686127,
      "num_tokens": 86774546.0,
      "step": 50025
    },
    {
      "entropy": 5.855463600158691,
      "epoch": 3.892511184594437,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003548749992109551,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2131939098238945,
      "num_tokens": 86784155.0,
      "step": 50030
    },
    {
      "entropy": 5.912651252746582,
      "epoch": 3.8929002139661546,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003548490882361139,
      "loss": 4.9823,
      "mean_token_accuracy": 0.21194558143615722,
      "num_tokens": 86792463.0,
      "step": 50035
    },
    {
      "entropy": 5.977617883682251,
      "epoch": 3.893289243337872,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003548231760496898,
      "loss": 5.0805,
      "mean_token_accuracy": 0.20782548934221268,
      "num_tokens": 86801880.0,
      "step": 50040
    },
    {
      "entropy": 5.880118322372437,
      "epoch": 3.8936782727095895,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035479726265207604,
      "loss": 4.9117,
      "mean_token_accuracy": 0.21418636441230773,
      "num_tokens": 86810813.0,
      "step": 50045
    },
    {
      "entropy": 5.897434091567993,
      "epoch": 3.8940673020813072,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003547713480436659,
      "loss": 5.0263,
      "mean_token_accuracy": 0.20230769366025925,
      "num_tokens": 86819373.0,
      "step": 50050
    },
    {
      "entropy": 5.840103578567505,
      "epoch": 3.894456331453025,
      "grad_norm": 1.484375,
      "learning_rate": 0.00035474543222485254,
      "loss": 4.9286,
      "mean_token_accuracy": 0.2125045746564865,
      "num_tokens": 86827369.0,
      "step": 50055
    },
    {
      "entropy": 5.887427091598511,
      "epoch": 3.894845360824742,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035471951519602926,
      "loss": 5.0545,
      "mean_token_accuracy": 0.2062416896224022,
      "num_tokens": 86836119.0,
      "step": 50060
    },
    {
      "entropy": 5.957984352111817,
      "epoch": 3.89523439019646,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00035469359695758905,
      "loss": 5.1126,
      "mean_token_accuracy": 0.20347007364034653,
      "num_tokens": 86844976.0,
      "step": 50065
    },
    {
      "entropy": 5.980688619613647,
      "epoch": 3.895623419568177,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003546676775099254,
      "loss": 5.0215,
      "mean_token_accuracy": 0.21019212901592255,
      "num_tokens": 86853245.0,
      "step": 50070
    },
    {
      "entropy": 5.888351392745972,
      "epoch": 3.896012448939895,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003546417568534315,
      "loss": 4.9584,
      "mean_token_accuracy": 0.215139140188694,
      "num_tokens": 86862682.0,
      "step": 50075
    },
    {
      "entropy": 5.931468296051025,
      "epoch": 3.8964014783116125,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035461583498850083,
      "loss": 5.0673,
      "mean_token_accuracy": 0.2063592791557312,
      "num_tokens": 86871900.0,
      "step": 50080
    },
    {
      "entropy": 5.915826606750488,
      "epoch": 3.8967905076833302,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003545899119155265,
      "loss": 5.0396,
      "mean_token_accuracy": 0.21245780140161513,
      "num_tokens": 86880417.0,
      "step": 50085
    },
    {
      "entropy": 5.860255146026612,
      "epoch": 3.8971795370550475,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035456398763490193,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21974065601825715,
      "num_tokens": 86888489.0,
      "step": 50090
    },
    {
      "entropy": 5.910390424728393,
      "epoch": 3.897568566426765,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003545380621470205,
      "loss": 5.025,
      "mean_token_accuracy": 0.21794474720954896,
      "num_tokens": 86897189.0,
      "step": 50095
    },
    {
      "entropy": 5.9353594303131105,
      "epoch": 3.897957595798483,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003545121354522756,
      "loss": 5.0529,
      "mean_token_accuracy": 0.21174402087926864,
      "num_tokens": 86906174.0,
      "step": 50100
    },
    {
      "entropy": 5.845537614822388,
      "epoch": 3.8983466251702,
      "grad_norm": 1.421875,
      "learning_rate": 0.00035448620755106064,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21674865782260894,
      "num_tokens": 86915002.0,
      "step": 50105
    },
    {
      "entropy": 5.9149223327636715,
      "epoch": 3.898735654541918,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003544602784437689,
      "loss": 5.0054,
      "mean_token_accuracy": 0.2094147324562073,
      "num_tokens": 86923611.0,
      "step": 50110
    },
    {
      "entropy": 5.908443975448608,
      "epoch": 3.8991246839136355,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035443434813079393,
      "loss": 4.9721,
      "mean_token_accuracy": 0.20889769047498702,
      "num_tokens": 86932717.0,
      "step": 50115
    },
    {
      "entropy": 5.916933965682984,
      "epoch": 3.8995137132853532,
      "grad_norm": 1.375,
      "learning_rate": 0.0003544084166125292,
      "loss": 5.0468,
      "mean_token_accuracy": 0.20657638609409332,
      "num_tokens": 86941522.0,
      "step": 50120
    },
    {
      "entropy": 5.87265419960022,
      "epoch": 3.8999027426570705,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003543824838893682,
      "loss": 4.9821,
      "mean_token_accuracy": 0.20651367902755738,
      "num_tokens": 86950593.0,
      "step": 50125
    },
    {
      "entropy": 5.949397897720337,
      "epoch": 3.900291772028788,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035435654996170417,
      "loss": 5.0163,
      "mean_token_accuracy": 0.20709467083215713,
      "num_tokens": 86959110.0,
      "step": 50130
    },
    {
      "entropy": 5.9439630031585695,
      "epoch": 3.900680801400506,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000354330614829931,
      "loss": 5.0968,
      "mean_token_accuracy": 0.20207741260528564,
      "num_tokens": 86968631.0,
      "step": 50135
    },
    {
      "entropy": 5.92245569229126,
      "epoch": 3.901069830772223,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003543046784944419,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21018511354923247,
      "num_tokens": 86977377.0,
      "step": 50140
    },
    {
      "entropy": 6.05735330581665,
      "epoch": 3.901458860143941,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00035427874095563054,
      "loss": 5.1534,
      "mean_token_accuracy": 0.20767497420310974,
      "num_tokens": 86985931.0,
      "step": 50145
    },
    {
      "entropy": 5.882279348373413,
      "epoch": 3.9018478895156585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003542528022138905,
      "loss": 4.9264,
      "mean_token_accuracy": 0.21683316081762313,
      "num_tokens": 86994038.0,
      "step": 50150
    },
    {
      "entropy": 5.805412435531617,
      "epoch": 3.9022369188873762,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003542268622696153,
      "loss": 4.935,
      "mean_token_accuracy": 0.22011127918958664,
      "num_tokens": 87003740.0,
      "step": 50155
    },
    {
      "entropy": 5.886851501464844,
      "epoch": 3.9026259482590935,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003542009211231986,
      "loss": 4.8846,
      "mean_token_accuracy": 0.21532836705446243,
      "num_tokens": 87012120.0,
      "step": 50160
    },
    {
      "entropy": 5.910329675674438,
      "epoch": 3.903014977630811,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00035417497877503396,
      "loss": 4.9681,
      "mean_token_accuracy": 0.21857584714889527,
      "num_tokens": 87020292.0,
      "step": 50165
    },
    {
      "entropy": 5.914029788970947,
      "epoch": 3.9034040070025284,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035414903522551505,
      "loss": 4.9858,
      "mean_token_accuracy": 0.21186985820531845,
      "num_tokens": 87028524.0,
      "step": 50170
    },
    {
      "entropy": 5.866074943542481,
      "epoch": 3.903793036374246,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035412309047503553,
      "loss": 4.9668,
      "mean_token_accuracy": 0.2085736021399498,
      "num_tokens": 87037772.0,
      "step": 50175
    },
    {
      "entropy": 5.933314990997315,
      "epoch": 3.904182065745964,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000354097144523989,
      "loss": 4.922,
      "mean_token_accuracy": 0.21190500557422637,
      "num_tokens": 87046333.0,
      "step": 50180
    },
    {
      "entropy": 5.864216041564942,
      "epoch": 3.9045710951176815,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003540711973727693,
      "loss": 4.8621,
      "mean_token_accuracy": 0.21888171881437302,
      "num_tokens": 87055507.0,
      "step": 50185
    },
    {
      "entropy": 5.9024125099182125,
      "epoch": 3.904960124489399,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003540452490217699,
      "loss": 4.9547,
      "mean_token_accuracy": 0.21199716478586197,
      "num_tokens": 87063817.0,
      "step": 50190
    },
    {
      "entropy": 5.857805633544922,
      "epoch": 3.9053491538611165,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003540192994713847,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21837370991706848,
      "num_tokens": 87072404.0,
      "step": 50195
    },
    {
      "entropy": 5.876820659637451,
      "epoch": 3.905738183232834,
      "grad_norm": 1.421875,
      "learning_rate": 0.00035399334872200747,
      "loss": 4.8968,
      "mean_token_accuracy": 0.21984235495328902,
      "num_tokens": 87081074.0,
      "step": 50200
    },
    {
      "entropy": 5.870394945144653,
      "epoch": 3.9061272126045514,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003539673967740318,
      "loss": 4.9637,
      "mean_token_accuracy": 0.20919657200574876,
      "num_tokens": 87089651.0,
      "step": 50205
    },
    {
      "entropy": 5.845789861679077,
      "epoch": 3.906516241976269,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003539414436278516,
      "loss": 4.9141,
      "mean_token_accuracy": 0.21383048295974733,
      "num_tokens": 87098255.0,
      "step": 50210
    },
    {
      "entropy": 5.831580448150635,
      "epoch": 3.906905271347987,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035391548928386077,
      "loss": 4.8955,
      "mean_token_accuracy": 0.21181600987911225,
      "num_tokens": 87107880.0,
      "step": 50215
    },
    {
      "entropy": 5.921594285964966,
      "epoch": 3.9072943007197045,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003538895337424529,
      "loss": 4.9918,
      "mean_token_accuracy": 0.21468723118305205,
      "num_tokens": 87116276.0,
      "step": 50220
    },
    {
      "entropy": 5.90599513053894,
      "epoch": 3.9076833300914218,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00035386357700402187,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21896442472934724,
      "num_tokens": 87124682.0,
      "step": 50225
    },
    {
      "entropy": 5.851755809783936,
      "epoch": 3.9080723594631395,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003538376190689616,
      "loss": 4.8404,
      "mean_token_accuracy": 0.22835186272859573,
      "num_tokens": 87133428.0,
      "step": 50230
    },
    {
      "entropy": 5.8333264827728275,
      "epoch": 3.908461388834857,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003538116599376661,
      "loss": 4.9487,
      "mean_token_accuracy": 0.21107601821422578,
      "num_tokens": 87142471.0,
      "step": 50235
    },
    {
      "entropy": 5.806998157501221,
      "epoch": 3.9088504182065744,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035378569961052887,
      "loss": 4.8379,
      "mean_token_accuracy": 0.2224346086382866,
      "num_tokens": 87151813.0,
      "step": 50240
    },
    {
      "entropy": 5.877148580551148,
      "epoch": 3.909239447578292,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035375973808794416,
      "loss": 4.9892,
      "mean_token_accuracy": 0.21060765385627747,
      "num_tokens": 87160438.0,
      "step": 50245
    },
    {
      "entropy": 5.827613401412964,
      "epoch": 3.90962847695001,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003537337753703058,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21244496256113052,
      "num_tokens": 87168937.0,
      "step": 50250
    },
    {
      "entropy": 5.93652720451355,
      "epoch": 3.9100175063217275,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003537078114580077,
      "loss": 4.924,
      "mean_token_accuracy": 0.21513839513063432,
      "num_tokens": 87176511.0,
      "step": 50255
    },
    {
      "entropy": 5.901549863815307,
      "epoch": 3.9104065356934448,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003536818463514438,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21373755484819412,
      "num_tokens": 87185250.0,
      "step": 50260
    },
    {
      "entropy": 5.881203317642212,
      "epoch": 3.9107955650651625,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035365588005100817,
      "loss": 5.0533,
      "mean_token_accuracy": 0.20628661215305327,
      "num_tokens": 87193621.0,
      "step": 50265
    },
    {
      "entropy": 5.934397745132446,
      "epoch": 3.9111845944368797,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003536299125570947,
      "loss": 5.1154,
      "mean_token_accuracy": 0.1933930665254593,
      "num_tokens": 87202307.0,
      "step": 50270
    },
    {
      "entropy": 5.928903293609619,
      "epoch": 3.9115736238085974,
      "grad_norm": 1.375,
      "learning_rate": 0.0003536039438700975,
      "loss": 4.9765,
      "mean_token_accuracy": 0.2120120570063591,
      "num_tokens": 87210504.0,
      "step": 50275
    },
    {
      "entropy": 5.9600667476654055,
      "epoch": 3.911962653180315,
      "grad_norm": 1.4375,
      "learning_rate": 0.00035357797399041056,
      "loss": 5.0334,
      "mean_token_accuracy": 0.2073104739189148,
      "num_tokens": 87219541.0,
      "step": 50280
    },
    {
      "entropy": 5.917445373535156,
      "epoch": 3.912351682552033,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003535520029184279,
      "loss": 4.9946,
      "mean_token_accuracy": 0.21059515774250032,
      "num_tokens": 87227996.0,
      "step": 50285
    },
    {
      "entropy": 5.867440605163575,
      "epoch": 3.9127407119237505,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0003535260306545437,
      "loss": 4.9545,
      "mean_token_accuracy": 0.21784480214118956,
      "num_tokens": 87235840.0,
      "step": 50290
    },
    {
      "entropy": 5.904961252212525,
      "epoch": 3.9131297412954678,
      "grad_norm": 1.421875,
      "learning_rate": 0.00035350005719915196,
      "loss": 4.9991,
      "mean_token_accuracy": 0.209321291744709,
      "num_tokens": 87243942.0,
      "step": 50295
    },
    {
      "entropy": 5.884698963165283,
      "epoch": 3.9135187706671855,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035347408255264674,
      "loss": 4.9583,
      "mean_token_accuracy": 0.21425328254699708,
      "num_tokens": 87252435.0,
      "step": 50300
    },
    {
      "entropy": 5.90400242805481,
      "epoch": 3.9139078000389027,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003534481067154223,
      "loss": 5.0005,
      "mean_token_accuracy": 0.20740165561437607,
      "num_tokens": 87261251.0,
      "step": 50305
    },
    {
      "entropy": 5.958360767364502,
      "epoch": 3.9142968294106204,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003534221296878726,
      "loss": 5.0401,
      "mean_token_accuracy": 0.20490928292274474,
      "num_tokens": 87269320.0,
      "step": 50310
    },
    {
      "entropy": 5.897350883483886,
      "epoch": 3.914685858782338,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035339615147039196,
      "loss": 4.9616,
      "mean_token_accuracy": 0.21413628458976747,
      "num_tokens": 87277934.0,
      "step": 50315
    },
    {
      "entropy": 5.964792156219483,
      "epoch": 3.915074888154056,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035337017206337444,
      "loss": 5.0932,
      "mean_token_accuracy": 0.2037027359008789,
      "num_tokens": 87287042.0,
      "step": 50320
    },
    {
      "entropy": 5.8947258472442625,
      "epoch": 3.915463917525773,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003533441914672144,
      "loss": 5.0285,
      "mean_token_accuracy": 0.21239891946315764,
      "num_tokens": 87295495.0,
      "step": 50325
    },
    {
      "entropy": 5.909796190261841,
      "epoch": 3.9158529468974907,
      "grad_norm": 1.375,
      "learning_rate": 0.00035331820968230595,
      "loss": 4.913,
      "mean_token_accuracy": 0.22106371372938155,
      "num_tokens": 87304242.0,
      "step": 50330
    },
    {
      "entropy": 5.906122255325317,
      "epoch": 3.9162419762692084,
      "grad_norm": 1.25,
      "learning_rate": 0.00035329222670904325,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20458732694387435,
      "num_tokens": 87313896.0,
      "step": 50335
    },
    {
      "entropy": 5.859498310089111,
      "epoch": 3.9166310056409257,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003532662425478206,
      "loss": 4.9336,
      "mean_token_accuracy": 0.22614033669233322,
      "num_tokens": 87322187.0,
      "step": 50340
    },
    {
      "entropy": 5.863008880615235,
      "epoch": 3.9170200350126434,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00035324025719903236,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21537211239337922,
      "num_tokens": 87331170.0,
      "step": 50345
    },
    {
      "entropy": 5.873621988296509,
      "epoch": 3.917409064384361,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00035321427066307276,
      "loss": 4.9723,
      "mean_token_accuracy": 0.21462617665529252,
      "num_tokens": 87339673.0,
      "step": 50350
    },
    {
      "entropy": 5.920474815368652,
      "epoch": 3.917798093756079,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003531882829403361,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20745428949594497,
      "num_tokens": 87347443.0,
      "step": 50355
    },
    {
      "entropy": 5.872058153152466,
      "epoch": 3.918187123127796,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035316229403121666,
      "loss": 4.9989,
      "mean_token_accuracy": 0.2058949738740921,
      "num_tokens": 87355554.0,
      "step": 50360
    },
    {
      "entropy": 5.922288703918457,
      "epoch": 3.9185761524995137,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003531363039361088,
      "loss": 4.9936,
      "mean_token_accuracy": 0.21099065542221068,
      "num_tokens": 87364392.0,
      "step": 50365
    },
    {
      "entropy": 5.8887474060058596,
      "epoch": 3.9189651818712314,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035311031265540695,
      "loss": 4.968,
      "mean_token_accuracy": 0.21295930594205856,
      "num_tokens": 87373474.0,
      "step": 50370
    },
    {
      "entropy": 6.000442981719971,
      "epoch": 3.9193542112429487,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035308432018950537,
      "loss": 4.9911,
      "mean_token_accuracy": 0.21125533282756806,
      "num_tokens": 87382158.0,
      "step": 50375
    },
    {
      "entropy": 5.995359277725219,
      "epoch": 3.9197432406146664,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035305832653879856,
      "loss": 5.0029,
      "mean_token_accuracy": 0.2086159884929657,
      "num_tokens": 87390032.0,
      "step": 50380
    },
    {
      "entropy": 5.922431802749633,
      "epoch": 3.920132269986384,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035303233170368083,
      "loss": 5.0292,
      "mean_token_accuracy": 0.20407381951808928,
      "num_tokens": 87398409.0,
      "step": 50385
    },
    {
      "entropy": 5.893648767471314,
      "epoch": 3.920521299358102,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035300633568454673,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21105469316244124,
      "num_tokens": 87407218.0,
      "step": 50390
    },
    {
      "entropy": 5.877258014678955,
      "epoch": 3.920910328729819,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035298033848179064,
      "loss": 4.9457,
      "mean_token_accuracy": 0.2116571456193924,
      "num_tokens": 87415869.0,
      "step": 50395
    },
    {
      "entropy": 5.845294523239136,
      "epoch": 3.9212993581015367,
      "grad_norm": 1.3125,
      "learning_rate": 0.000352954340095807,
      "loss": 4.9678,
      "mean_token_accuracy": 0.21793511807918547,
      "num_tokens": 87424636.0,
      "step": 50400
    },
    {
      "entropy": 5.8885386943817135,
      "epoch": 3.921688387473254,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00035292834052699036,
      "loss": 4.9847,
      "mean_token_accuracy": 0.20811957120895386,
      "num_tokens": 87432833.0,
      "step": 50405
    },
    {
      "entropy": 5.94917778968811,
      "epoch": 3.9220774168449717,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003529023397757351,
      "loss": 5.0526,
      "mean_token_accuracy": 0.20956340730190276,
      "num_tokens": 87442002.0,
      "step": 50410
    },
    {
      "entropy": 5.907338953018188,
      "epoch": 3.9224664462166894,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035287633784243595,
      "loss": 4.995,
      "mean_token_accuracy": 0.2081120118498802,
      "num_tokens": 87451271.0,
      "step": 50415
    },
    {
      "entropy": 5.9394889831542965,
      "epoch": 3.922855475588407,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035285033472748725,
      "loss": 4.9893,
      "mean_token_accuracy": 0.2068886861205101,
      "num_tokens": 87460290.0,
      "step": 50420
    },
    {
      "entropy": 5.910650444030762,
      "epoch": 3.9232445049601243,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003528243304312837,
      "loss": 4.961,
      "mean_token_accuracy": 0.21754674315452577,
      "num_tokens": 87469131.0,
      "step": 50425
    },
    {
      "entropy": 5.891176176071167,
      "epoch": 3.923633534331842,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003527983249542197,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21013073176145552,
      "num_tokens": 87477445.0,
      "step": 50430
    },
    {
      "entropy": 5.9382692813873295,
      "epoch": 3.9240225637035597,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035277231829669006,
      "loss": 5.0659,
      "mean_token_accuracy": 0.20761577039957047,
      "num_tokens": 87486080.0,
      "step": 50435
    },
    {
      "entropy": 5.944386529922485,
      "epoch": 3.924411593075277,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003527463104590892,
      "loss": 5.0188,
      "mean_token_accuracy": 0.20501877516508102,
      "num_tokens": 87494669.0,
      "step": 50440
    },
    {
      "entropy": 5.931106424331665,
      "epoch": 3.9248006224469947,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003527203014418119,
      "loss": 5.0337,
      "mean_token_accuracy": 0.20832479596138,
      "num_tokens": 87503497.0,
      "step": 50445
    },
    {
      "entropy": 5.866706800460816,
      "epoch": 3.9251896518187124,
      "grad_norm": 1.5,
      "learning_rate": 0.0003526942912452526,
      "loss": 4.9905,
      "mean_token_accuracy": 0.2143278867006302,
      "num_tokens": 87512038.0,
      "step": 50450
    },
    {
      "entropy": 5.896438550949097,
      "epoch": 3.92557868119043,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003526682798698062,
      "loss": 5.0157,
      "mean_token_accuracy": 0.20606726855039598,
      "num_tokens": 87520176.0,
      "step": 50455
    },
    {
      "entropy": 5.955945777893066,
      "epoch": 3.9259677105621473,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003526422673158673,
      "loss": 5.0126,
      "mean_token_accuracy": 0.20951262414455413,
      "num_tokens": 87528961.0,
      "step": 50460
    },
    {
      "entropy": 5.911625576019287,
      "epoch": 3.926356739933865,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003526162535838305,
      "loss": 4.9816,
      "mean_token_accuracy": 0.20987471491098403,
      "num_tokens": 87538424.0,
      "step": 50465
    },
    {
      "entropy": 5.885615062713623,
      "epoch": 3.9267457693055827,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035259023867409065,
      "loss": 4.9844,
      "mean_token_accuracy": 0.21681899726390838,
      "num_tokens": 87547234.0,
      "step": 50470
    },
    {
      "entropy": 5.811842393875122,
      "epoch": 3.9271347986773,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003525642225870423,
      "loss": 4.8808,
      "mean_token_accuracy": 0.2244354635477066,
      "num_tokens": 87556276.0,
      "step": 50475
    },
    {
      "entropy": 5.953203630447388,
      "epoch": 3.9275238280490177,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035253820532308043,
      "loss": 5.0383,
      "mean_token_accuracy": 0.20998992025852203,
      "num_tokens": 87565808.0,
      "step": 50480
    },
    {
      "entropy": 5.883889532089233,
      "epoch": 3.9279128574207354,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0003525121868825997,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22596168220043183,
      "num_tokens": 87573124.0,
      "step": 50485
    },
    {
      "entropy": 5.874752426147461,
      "epoch": 3.928301886792453,
      "grad_norm": 1.375,
      "learning_rate": 0.0003524861672659949,
      "loss": 4.9901,
      "mean_token_accuracy": 0.2142250880599022,
      "num_tokens": 87581635.0,
      "step": 50490
    },
    {
      "entropy": 5.847004032135009,
      "epoch": 3.9286909161641703,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003524601464736609,
      "loss": 4.9628,
      "mean_token_accuracy": 0.2151826500892639,
      "num_tokens": 87589733.0,
      "step": 50495
    },
    {
      "entropy": 5.81771445274353,
      "epoch": 3.929079945535888,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003524341245059924,
      "loss": 4.948,
      "mean_token_accuracy": 0.21453103125095369,
      "num_tokens": 87598593.0,
      "step": 50500
    },
    {
      "entropy": 5.859150981903076,
      "epoch": 3.9294689749076053,
      "grad_norm": 1.375,
      "learning_rate": 0.0003524081013633843,
      "loss": 4.91,
      "mean_token_accuracy": 0.21639949083328247,
      "num_tokens": 87607418.0,
      "step": 50505
    },
    {
      "entropy": 5.889812994003296,
      "epoch": 3.929858004279323,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035238207704623147,
      "loss": 4.9686,
      "mean_token_accuracy": 0.21492732018232347,
      "num_tokens": 87615799.0,
      "step": 50510
    },
    {
      "entropy": 5.904260921478271,
      "epoch": 3.9302470336510407,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035235605155492885,
      "loss": 4.9711,
      "mean_token_accuracy": 0.21084872931241988,
      "num_tokens": 87624403.0,
      "step": 50515
    },
    {
      "entropy": 5.890996932983398,
      "epoch": 3.9306360630227584,
      "grad_norm": 1.375,
      "learning_rate": 0.00035233002488987123,
      "loss": 5.0038,
      "mean_token_accuracy": 0.20418586432933808,
      "num_tokens": 87632963.0,
      "step": 50520
    },
    {
      "entropy": 5.889699459075928,
      "epoch": 3.9310250923944756,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035230399705145365,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21526117771863937,
      "num_tokens": 87641441.0,
      "step": 50525
    },
    {
      "entropy": 5.86859359741211,
      "epoch": 3.9314141217661933,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003522779680400709,
      "loss": 5.0292,
      "mean_token_accuracy": 0.20645926743745804,
      "num_tokens": 87650383.0,
      "step": 50530
    },
    {
      "entropy": 5.837763023376465,
      "epoch": 3.931803151137911,
      "grad_norm": 1.25,
      "learning_rate": 0.00035225193785611795,
      "loss": 4.9419,
      "mean_token_accuracy": 0.2159390330314636,
      "num_tokens": 87658963.0,
      "step": 50535
    },
    {
      "entropy": 5.846101856231689,
      "epoch": 3.9321921805096283,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003522259064999898,
      "loss": 4.891,
      "mean_token_accuracy": 0.21844946444034577,
      "num_tokens": 87667459.0,
      "step": 50540
    },
    {
      "entropy": 5.950911617279052,
      "epoch": 3.932581209881346,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035219987397208137,
      "loss": 5.0479,
      "mean_token_accuracy": 0.2054503694176674,
      "num_tokens": 87677443.0,
      "step": 50545
    },
    {
      "entropy": 5.927232694625855,
      "epoch": 3.9329702392530637,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003521738402727878,
      "loss": 5.0054,
      "mean_token_accuracy": 0.2104513540863991,
      "num_tokens": 87686019.0,
      "step": 50550
    },
    {
      "entropy": 5.874931335449219,
      "epoch": 3.9333592686247814,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003521478054025041,
      "loss": 4.9742,
      "mean_token_accuracy": 0.21403136253356933,
      "num_tokens": 87695754.0,
      "step": 50555
    },
    {
      "entropy": 5.895137214660645,
      "epoch": 3.9337482979964986,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035212176936162524,
      "loss": 4.9815,
      "mean_token_accuracy": 0.2151082471013069,
      "num_tokens": 87704961.0,
      "step": 50560
    },
    {
      "entropy": 5.80647873878479,
      "epoch": 3.9341373273682163,
      "grad_norm": 1.578125,
      "learning_rate": 0.0003520957321505462,
      "loss": 4.849,
      "mean_token_accuracy": 0.22096519470214843,
      "num_tokens": 87713378.0,
      "step": 50565
    },
    {
      "entropy": 5.873860263824463,
      "epoch": 3.934526356739934,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035206969376966226,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21443846821784973,
      "num_tokens": 87722880.0,
      "step": 50570
    },
    {
      "entropy": 5.880841207504273,
      "epoch": 3.9349153861116513,
      "grad_norm": 1.5,
      "learning_rate": 0.00035204365421936826,
      "loss": 4.9411,
      "mean_token_accuracy": 0.2129504770040512,
      "num_tokens": 87730794.0,
      "step": 50575
    },
    {
      "entropy": 6.014259576797485,
      "epoch": 3.935304415483369,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003520176135000594,
      "loss": 5.045,
      "mean_token_accuracy": 0.20272360295057296,
      "num_tokens": 87739906.0,
      "step": 50580
    },
    {
      "entropy": 5.906285381317138,
      "epoch": 3.9356934448550867,
      "grad_norm": 1.375,
      "learning_rate": 0.0003519915716121309,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2147137239575386,
      "num_tokens": 87748656.0,
      "step": 50585
    },
    {
      "entropy": 5.860756206512451,
      "epoch": 3.9360824742268044,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003519655285559779,
      "loss": 4.9812,
      "mean_token_accuracy": 0.21556138396263122,
      "num_tokens": 87756672.0,
      "step": 50590
    },
    {
      "entropy": 5.882245397567749,
      "epoch": 3.9364715035985216,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035193948433199547,
      "loss": 4.9586,
      "mean_token_accuracy": 0.21297724694013595,
      "num_tokens": 87765506.0,
      "step": 50595
    },
    {
      "entropy": 5.922574281692505,
      "epoch": 3.9368605329702393,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035191343894057885,
      "loss": 5.0181,
      "mean_token_accuracy": 0.21283968389034272,
      "num_tokens": 87774401.0,
      "step": 50600
    },
    {
      "entropy": 5.950524616241455,
      "epoch": 3.9372495623419566,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035188739238212316,
      "loss": 5.0795,
      "mean_token_accuracy": 0.19888032972812653,
      "num_tokens": 87783277.0,
      "step": 50605
    },
    {
      "entropy": 5.895290279388428,
      "epoch": 3.9376385917136743,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003518613446570237,
      "loss": 5.0279,
      "mean_token_accuracy": 0.20867571234703064,
      "num_tokens": 87791532.0,
      "step": 50610
    },
    {
      "entropy": 5.8752950668334964,
      "epoch": 3.938027621085392,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035183529576567567,
      "loss": 4.9277,
      "mean_token_accuracy": 0.22382221966981888,
      "num_tokens": 87800179.0,
      "step": 50615
    },
    {
      "entropy": 5.882565116882324,
      "epoch": 3.9384166504571096,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003518092457084743,
      "loss": 5.0289,
      "mean_token_accuracy": 0.20625824332237244,
      "num_tokens": 87808940.0,
      "step": 50620
    },
    {
      "entropy": 5.894248819351196,
      "epoch": 3.9388056798288273,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003517831944858149,
      "loss": 5.0463,
      "mean_token_accuracy": 0.21112489849328994,
      "num_tokens": 87817730.0,
      "step": 50625
    },
    {
      "entropy": 5.930561494827271,
      "epoch": 3.9391947092005446,
      "grad_norm": 1.4375,
      "learning_rate": 0.00035175714209809275,
      "loss": 4.9347,
      "mean_token_accuracy": 0.21212247610092164,
      "num_tokens": 87825836.0,
      "step": 50630
    },
    {
      "entropy": 5.871069669723511,
      "epoch": 3.9395837385722623,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003517310885457032,
      "loss": 4.8361,
      "mean_token_accuracy": 0.22699456959962844,
      "num_tokens": 87834977.0,
      "step": 50635
    },
    {
      "entropy": 5.850802326202393,
      "epoch": 3.9399727679439795,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003517050338290414,
      "loss": 5.0249,
      "mean_token_accuracy": 0.20773414224386216,
      "num_tokens": 87843290.0,
      "step": 50640
    },
    {
      "entropy": 5.91695761680603,
      "epoch": 3.9403617973156972,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035167897794850285,
      "loss": 5.0088,
      "mean_token_accuracy": 0.20507010370492934,
      "num_tokens": 87851826.0,
      "step": 50645
    },
    {
      "entropy": 5.941468381881714,
      "epoch": 3.940750826687415,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00035165292090448287,
      "loss": 5.037,
      "mean_token_accuracy": 0.20769549012184144,
      "num_tokens": 87860537.0,
      "step": 50650
    },
    {
      "entropy": 5.937904214859008,
      "epoch": 3.9411398560591326,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035162686269737676,
      "loss": 5.031,
      "mean_token_accuracy": 0.20416270792484284,
      "num_tokens": 87869006.0,
      "step": 50655
    },
    {
      "entropy": 5.911216688156128,
      "epoch": 3.94152888543085,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035160080332758006,
      "loss": 4.9371,
      "mean_token_accuracy": 0.2107901766896248,
      "num_tokens": 87877491.0,
      "step": 50660
    },
    {
      "entropy": 5.899318027496338,
      "epoch": 3.9419179148025676,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000351574742795488,
      "loss": 5.0018,
      "mean_token_accuracy": 0.21113879680633546,
      "num_tokens": 87885346.0,
      "step": 50665
    },
    {
      "entropy": 5.92356424331665,
      "epoch": 3.9423069441742853,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003515486811014962,
      "loss": 5.0652,
      "mean_token_accuracy": 0.20590630024671555,
      "num_tokens": 87894469.0,
      "step": 50670
    },
    {
      "entropy": 5.92971248626709,
      "epoch": 3.9426959735460025,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003515226182459999,
      "loss": 5.0299,
      "mean_token_accuracy": 0.20347026437520982,
      "num_tokens": 87902735.0,
      "step": 50675
    },
    {
      "entropy": 5.956442260742188,
      "epoch": 3.9430850029177202,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035149655422939477,
      "loss": 4.966,
      "mean_token_accuracy": 0.21791040003299714,
      "num_tokens": 87911712.0,
      "step": 50680
    },
    {
      "entropy": 5.885373497009278,
      "epoch": 3.943474032289438,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003514704890520761,
      "loss": 4.946,
      "mean_token_accuracy": 0.21934774518013,
      "num_tokens": 87920655.0,
      "step": 50685
    },
    {
      "entropy": 5.8598816871643065,
      "epoch": 3.9438630616611556,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003514444227144395,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21397238224744797,
      "num_tokens": 87929221.0,
      "step": 50690
    },
    {
      "entropy": 5.940036964416504,
      "epoch": 3.944252091032873,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003514183552168804,
      "loss": 5.0353,
      "mean_token_accuracy": 0.2082590103149414,
      "num_tokens": 87938519.0,
      "step": 50695
    },
    {
      "entropy": 5.931116008758545,
      "epoch": 3.9446411204045906,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035139228655979457,
      "loss": 4.9839,
      "mean_token_accuracy": 0.2177615061402321,
      "num_tokens": 87947154.0,
      "step": 50700
    },
    {
      "entropy": 5.885914468765259,
      "epoch": 3.945030149776308,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035136621674357717,
      "loss": 4.9609,
      "mean_token_accuracy": 0.21244670152664186,
      "num_tokens": 87956352.0,
      "step": 50705
    },
    {
      "entropy": 5.869151544570923,
      "epoch": 3.9454191791480255,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003513401457686242,
      "loss": 4.916,
      "mean_token_accuracy": 0.21866313070058824,
      "num_tokens": 87965722.0,
      "step": 50710
    },
    {
      "entropy": 5.901678657531738,
      "epoch": 3.9458082085197432,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035131407363533083,
      "loss": 4.8931,
      "mean_token_accuracy": 0.225302854180336,
      "num_tokens": 87973869.0,
      "step": 50715
    },
    {
      "entropy": 5.793305921554565,
      "epoch": 3.946197237891461,
      "grad_norm": 1.375,
      "learning_rate": 0.0003512880003440929,
      "loss": 4.9349,
      "mean_token_accuracy": 0.2137158051133156,
      "num_tokens": 87982440.0,
      "step": 50720
    },
    {
      "entropy": 5.886638069152832,
      "epoch": 3.9465862672631786,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003512619258953061,
      "loss": 4.9727,
      "mean_token_accuracy": 0.2115711897611618,
      "num_tokens": 87990965.0,
      "step": 50725
    },
    {
      "entropy": 5.865990495681762,
      "epoch": 3.946975296634896,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003512358502893658,
      "loss": 4.9254,
      "mean_token_accuracy": 0.21386488229036332,
      "num_tokens": 87998884.0,
      "step": 50730
    },
    {
      "entropy": 5.856468915939331,
      "epoch": 3.9473643260066136,
      "grad_norm": 1.328125,
      "learning_rate": 0.000351209773526668,
      "loss": 4.9158,
      "mean_token_accuracy": 0.2082754999399185,
      "num_tokens": 88007404.0,
      "step": 50735
    },
    {
      "entropy": 5.799933910369873,
      "epoch": 3.947753355378331,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000351183695607608,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21204039752483367,
      "num_tokens": 88015559.0,
      "step": 50740
    },
    {
      "entropy": 5.919894170761109,
      "epoch": 3.9481423847500485,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035115761653258177,
      "loss": 4.9987,
      "mean_token_accuracy": 0.21437261253595352,
      "num_tokens": 88024374.0,
      "step": 50745
    },
    {
      "entropy": 5.886981344223022,
      "epoch": 3.9485314141217662,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035113153630198494,
      "loss": 4.9792,
      "mean_token_accuracy": 0.21586696803569794,
      "num_tokens": 88032372.0,
      "step": 50750
    },
    {
      "entropy": 5.862481117248535,
      "epoch": 3.948920443493484,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003511054549162132,
      "loss": 4.9599,
      "mean_token_accuracy": 0.2198229879140854,
      "num_tokens": 88040859.0,
      "step": 50755
    },
    {
      "entropy": 5.911664009094238,
      "epoch": 3.949309472865201,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035107937237566233,
      "loss": 4.9834,
      "mean_token_accuracy": 0.20788655430078506,
      "num_tokens": 88049894.0,
      "step": 50760
    },
    {
      "entropy": 5.871280860900879,
      "epoch": 3.949698502236919,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003510532886807281,
      "loss": 4.8839,
      "mean_token_accuracy": 0.21917642652988434,
      "num_tokens": 88057692.0,
      "step": 50765
    },
    {
      "entropy": 5.814662742614746,
      "epoch": 3.9500875316086366,
      "grad_norm": 1.25,
      "learning_rate": 0.0003510272038318062,
      "loss": 4.9349,
      "mean_token_accuracy": 0.21631101965904237,
      "num_tokens": 88066482.0,
      "step": 50770
    },
    {
      "entropy": 5.829773616790772,
      "epoch": 3.950476560980354,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003510011178292925,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21245686262845992,
      "num_tokens": 88075318.0,
      "step": 50775
    },
    {
      "entropy": 5.883805847167968,
      "epoch": 3.9508655903520715,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003509750306735828,
      "loss": 4.9535,
      "mean_token_accuracy": 0.21332098990678788,
      "num_tokens": 88083471.0,
      "step": 50780
    },
    {
      "entropy": 5.900699710845947,
      "epoch": 3.951254619723789,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000350948942365073,
      "loss": 4.9925,
      "mean_token_accuracy": 0.20952294319868087,
      "num_tokens": 88093473.0,
      "step": 50785
    },
    {
      "entropy": 5.972920656204224,
      "epoch": 3.951643649095507,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035092285290415885,
      "loss": 5.0526,
      "mean_token_accuracy": 0.20211092829704286,
      "num_tokens": 88102512.0,
      "step": 50790
    },
    {
      "entropy": 5.918581485748291,
      "epoch": 3.952032678467224,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003508967622912362,
      "loss": 5.0059,
      "mean_token_accuracy": 0.20745243579149247,
      "num_tokens": 88111171.0,
      "step": 50795
    },
    {
      "entropy": 5.991065835952758,
      "epoch": 3.952421707838942,
      "grad_norm": 1.546875,
      "learning_rate": 0.00035087067052670115,
      "loss": 5.0704,
      "mean_token_accuracy": 0.2020331710577011,
      "num_tokens": 88119565.0,
      "step": 50800
    },
    {
      "entropy": 5.961199235916138,
      "epoch": 3.9528107372106596,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035084457761094925,
      "loss": 5.0499,
      "mean_token_accuracy": 0.206028351187706,
      "num_tokens": 88128647.0,
      "step": 50805
    },
    {
      "entropy": 5.928804397583008,
      "epoch": 3.953199766582377,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003508184835443766,
      "loss": 4.9909,
      "mean_token_accuracy": 0.214894862473011,
      "num_tokens": 88138154.0,
      "step": 50810
    },
    {
      "entropy": 5.843280029296875,
      "epoch": 3.9535887959540945,
      "grad_norm": 1.453125,
      "learning_rate": 0.00035079238832737926,
      "loss": 4.991,
      "mean_token_accuracy": 0.2065334677696228,
      "num_tokens": 88147024.0,
      "step": 50815
    },
    {
      "entropy": 5.925642681121826,
      "epoch": 3.953977825325812,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000350766291960353,
      "loss": 5.0499,
      "mean_token_accuracy": 0.20830301344394683,
      "num_tokens": 88156132.0,
      "step": 50820
    },
    {
      "entropy": 5.973859119415283,
      "epoch": 3.95436685469753,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003507401944436939,
      "loss": 5.0841,
      "mean_token_accuracy": 0.20377764254808425,
      "num_tokens": 88165393.0,
      "step": 50825
    },
    {
      "entropy": 5.892049646377563,
      "epoch": 3.954755884069247,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003507140957777979,
      "loss": 4.9303,
      "mean_token_accuracy": 0.21226710677146912,
      "num_tokens": 88173452.0,
      "step": 50830
    },
    {
      "entropy": 5.894001817703247,
      "epoch": 3.955144913440965,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035068799596306096,
      "loss": 5.0033,
      "mean_token_accuracy": 0.2130879357457161,
      "num_tokens": 88182756.0,
      "step": 50835
    },
    {
      "entropy": 5.8498951435089115,
      "epoch": 3.955533942812682,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003506618949998791,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21155122816562652,
      "num_tokens": 88191040.0,
      "step": 50840
    },
    {
      "entropy": 5.92398509979248,
      "epoch": 3.9559229721844,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035063579288864853,
      "loss": 4.9627,
      "mean_token_accuracy": 0.2140589326620102,
      "num_tokens": 88199346.0,
      "step": 50845
    },
    {
      "entropy": 5.883610391616822,
      "epoch": 3.9563120015561175,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003506096896297651,
      "loss": 4.9968,
      "mean_token_accuracy": 0.21068155467510224,
      "num_tokens": 88208108.0,
      "step": 50850
    },
    {
      "entropy": 5.888367223739624,
      "epoch": 3.956701030927835,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035058358522362505,
      "loss": 4.9107,
      "mean_token_accuracy": 0.21909426897764206,
      "num_tokens": 88216747.0,
      "step": 50855
    },
    {
      "entropy": 5.962978744506836,
      "epoch": 3.9570900602995525,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035055747967062433,
      "loss": 5.1205,
      "mean_token_accuracy": 0.20052213221788406,
      "num_tokens": 88225943.0,
      "step": 50860
    },
    {
      "entropy": 5.896740961074829,
      "epoch": 3.95747908967127,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003505313729711592,
      "loss": 4.9976,
      "mean_token_accuracy": 0.22165123224258423,
      "num_tokens": 88235209.0,
      "step": 50865
    },
    {
      "entropy": 5.900451707839966,
      "epoch": 3.957868119042988,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035050526512562564,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21454713493585587,
      "num_tokens": 88244164.0,
      "step": 50870
    },
    {
      "entropy": 5.89416823387146,
      "epoch": 3.958257148414705,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003504791561344199,
      "loss": 4.9888,
      "mean_token_accuracy": 0.21069570183753966,
      "num_tokens": 88253099.0,
      "step": 50875
    },
    {
      "entropy": 5.933249568939209,
      "epoch": 3.958646177786423,
      "grad_norm": 1.484375,
      "learning_rate": 0.00035045304599793807,
      "loss": 5.024,
      "mean_token_accuracy": 0.21191158890724182,
      "num_tokens": 88260903.0,
      "step": 50880
    },
    {
      "entropy": 5.886406135559082,
      "epoch": 3.9590352071581405,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035042693471657637,
      "loss": 4.9572,
      "mean_token_accuracy": 0.22632163614034653,
      "num_tokens": 88270429.0,
      "step": 50885
    },
    {
      "entropy": 5.897616767883301,
      "epoch": 3.959424236529858,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035040082229073096,
      "loss": 5.0474,
      "mean_token_accuracy": 0.21058336943387984,
      "num_tokens": 88279777.0,
      "step": 50890
    },
    {
      "entropy": 5.8868523120880125,
      "epoch": 3.9598132659015755,
      "grad_norm": 1.375,
      "learning_rate": 0.0003503747087207981,
      "loss": 4.9416,
      "mean_token_accuracy": 0.21698378622531891,
      "num_tokens": 88288247.0,
      "step": 50895
    },
    {
      "entropy": 5.95027904510498,
      "epoch": 3.960202295273293,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00035034859400717407,
      "loss": 5.025,
      "mean_token_accuracy": 0.20573436766862868,
      "num_tokens": 88296552.0,
      "step": 50900
    },
    {
      "entropy": 5.8641704559326175,
      "epoch": 3.960591324645011,
      "grad_norm": 1.4375,
      "learning_rate": 0.00035032247815025504,
      "loss": 4.8549,
      "mean_token_accuracy": 0.22071026265621185,
      "num_tokens": 88304017.0,
      "step": 50905
    },
    {
      "entropy": 5.929469776153565,
      "epoch": 3.960980354016728,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035029636115043724,
      "loss": 5.0773,
      "mean_token_accuracy": 0.2078970640897751,
      "num_tokens": 88312650.0,
      "step": 50910
    },
    {
      "entropy": 5.874738597869873,
      "epoch": 3.961369383388446,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035027024300811707,
      "loss": 4.9416,
      "mean_token_accuracy": 0.21387553215026855,
      "num_tokens": 88321409.0,
      "step": 50915
    },
    {
      "entropy": 5.936597776412964,
      "epoch": 3.9617584127601635,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003502441237236907,
      "loss": 5.0044,
      "mean_token_accuracy": 0.20834557265043258,
      "num_tokens": 88329426.0,
      "step": 50920
    },
    {
      "entropy": 5.952096939086914,
      "epoch": 3.962147442131881,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003502180032975545,
      "loss": 5.0393,
      "mean_token_accuracy": 0.20776718705892563,
      "num_tokens": 88338151.0,
      "step": 50925
    },
    {
      "entropy": 5.9335136890411375,
      "epoch": 3.9625364715035984,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003501918817301049,
      "loss": 4.9653,
      "mean_token_accuracy": 0.21491142064332963,
      "num_tokens": 88346090.0,
      "step": 50930
    },
    {
      "entropy": 5.851593255996704,
      "epoch": 3.962925500875316,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035016575902173814,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21217916756868363,
      "num_tokens": 88354646.0,
      "step": 50935
    },
    {
      "entropy": 5.868579769134522,
      "epoch": 3.9633145302470334,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035013963517285064,
      "loss": 4.9505,
      "mean_token_accuracy": 0.2103911742568016,
      "num_tokens": 88363024.0,
      "step": 50940
    },
    {
      "entropy": 5.930336380004883,
      "epoch": 3.963703559618751,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00035011351018383874,
      "loss": 5.073,
      "mean_token_accuracy": 0.20166740119457244,
      "num_tokens": 88371399.0,
      "step": 50945
    },
    {
      "entropy": 5.909909582138061,
      "epoch": 3.964092588990469,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003500873840550989,
      "loss": 4.949,
      "mean_token_accuracy": 0.21228737980127335,
      "num_tokens": 88381178.0,
      "step": 50950
    },
    {
      "entropy": 5.891779518127441,
      "epoch": 3.9644816183621865,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035006125678702755,
      "loss": 4.932,
      "mean_token_accuracy": 0.21986540406942368,
      "num_tokens": 88389031.0,
      "step": 50955
    },
    {
      "entropy": 5.7916460037231445,
      "epoch": 3.964870647733904,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00035003512838002103,
      "loss": 4.8476,
      "mean_token_accuracy": 0.21537399142980576,
      "num_tokens": 88397242.0,
      "step": 50960
    },
    {
      "entropy": 5.8204034805297855,
      "epoch": 3.9652596771056214,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003500089988344759,
      "loss": 4.9598,
      "mean_token_accuracy": 0.21078855991363527,
      "num_tokens": 88406102.0,
      "step": 50965
    },
    {
      "entropy": 5.90370659828186,
      "epoch": 3.965648706477339,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034998286815078855,
      "loss": 4.9009,
      "mean_token_accuracy": 0.21408475041389466,
      "num_tokens": 88414833.0,
      "step": 50970
    },
    {
      "entropy": 5.971953916549682,
      "epoch": 3.9660377358490564,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00034995673632935557,
      "loss": 5.0277,
      "mean_token_accuracy": 0.21060966998338698,
      "num_tokens": 88423282.0,
      "step": 50975
    },
    {
      "entropy": 5.854204797744751,
      "epoch": 3.966426765220774,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003499306033705735,
      "loss": 4.9316,
      "mean_token_accuracy": 0.21204428523778915,
      "num_tokens": 88431745.0,
      "step": 50980
    },
    {
      "entropy": 5.825411367416382,
      "epoch": 3.966815794592492,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034990446927483877,
      "loss": 4.9653,
      "mean_token_accuracy": 0.21440336108207703,
      "num_tokens": 88439724.0,
      "step": 50985
    },
    {
      "entropy": 5.825381422042847,
      "epoch": 3.9672048239642095,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003498783340425479,
      "loss": 4.8811,
      "mean_token_accuracy": 0.2168792173266411,
      "num_tokens": 88448225.0,
      "step": 50990
    },
    {
      "entropy": 5.919149589538574,
      "epoch": 3.9675938533359267,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034985219767409747,
      "loss": 5.0099,
      "mean_token_accuracy": 0.21551765501499176,
      "num_tokens": 88456495.0,
      "step": 50995
    },
    {
      "entropy": 5.888489532470703,
      "epoch": 3.9679828827076444,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034982606016988407,
      "loss": 4.9546,
      "mean_token_accuracy": 0.21605252027511596,
      "num_tokens": 88465225.0,
      "step": 51000
    },
    {
      "epoch": 3.9679828827076444,
      "eval_entropy": 5.641780911005955,
      "eval_loss": 5.067383289337158,
      "eval_mean_token_accuracy": 0.21794036806088668,
      "eval_num_tokens": 88465225.0,
      "eval_runtime": 21.624,
      "eval_samples_per_second": 1921.849,
      "eval_steps_per_second": 240.243,
      "step": 51000
    },
    {
      "entropy": 5.946999359130859,
      "epoch": 3.968371912079362,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034979992153030424,
      "loss": 5.0312,
      "mean_token_accuracy": 0.21053029894828795,
      "num_tokens": 88474393.0,
      "step": 51005
    },
    {
      "entropy": 5.912715148925781,
      "epoch": 3.9687609414510794,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003497737817557548,
      "loss": 5.083,
      "mean_token_accuracy": 0.20431047379970552,
      "num_tokens": 88483257.0,
      "step": 51010
    },
    {
      "entropy": 5.926120662689209,
      "epoch": 3.969149970822797,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034974764084663213,
      "loss": 5.0379,
      "mean_token_accuracy": 0.21213313341140747,
      "num_tokens": 88492077.0,
      "step": 51015
    },
    {
      "entropy": 5.883660507202149,
      "epoch": 3.9695390001945148,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034972149880333316,
      "loss": 4.8903,
      "mean_token_accuracy": 0.21478844285011292,
      "num_tokens": 88501077.0,
      "step": 51020
    },
    {
      "entropy": 5.896730709075928,
      "epoch": 3.9699280295662325,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003496953556262542,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2159280851483345,
      "num_tokens": 88509281.0,
      "step": 51025
    },
    {
      "entropy": 5.851402521133423,
      "epoch": 3.9703170589379497,
      "grad_norm": 1.421875,
      "learning_rate": 0.00034966921131579216,
      "loss": 4.9681,
      "mean_token_accuracy": 0.21660338640213012,
      "num_tokens": 88518019.0,
      "step": 51030
    },
    {
      "entropy": 5.810026454925537,
      "epoch": 3.9707060883096674,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034964306587234367,
      "loss": 4.9107,
      "mean_token_accuracy": 0.21791840344667435,
      "num_tokens": 88525625.0,
      "step": 51035
    },
    {
      "entropy": 5.864147615432739,
      "epoch": 3.9710951176813847,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034961691929630544,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21436722874641417,
      "num_tokens": 88533733.0,
      "step": 51040
    },
    {
      "entropy": 5.853624725341797,
      "epoch": 3.9714841470531024,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003495907715880743,
      "loss": 4.9251,
      "mean_token_accuracy": 0.2131596803665161,
      "num_tokens": 88541565.0,
      "step": 51045
    },
    {
      "entropy": 5.930598449707031,
      "epoch": 3.97187317642482,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003495646227480469,
      "loss": 5.1032,
      "mean_token_accuracy": 0.2036469519138336,
      "num_tokens": 88549654.0,
      "step": 51050
    },
    {
      "entropy": 5.924565601348877,
      "epoch": 3.9722622057965378,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00034953847277662,
      "loss": 4.915,
      "mean_token_accuracy": 0.21339840739965438,
      "num_tokens": 88558572.0,
      "step": 51055
    },
    {
      "entropy": 5.924161911010742,
      "epoch": 3.9726512351682555,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034951232167419044,
      "loss": 4.9503,
      "mean_token_accuracy": 0.2136978477239609,
      "num_tokens": 88566810.0,
      "step": 51060
    },
    {
      "entropy": 5.997542428970337,
      "epoch": 3.9730402645399727,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034948616944115504,
      "loss": 5.1881,
      "mean_token_accuracy": 0.20055115967988968,
      "num_tokens": 88576439.0,
      "step": 51065
    },
    {
      "entropy": 5.91132173538208,
      "epoch": 3.9734292939116904,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003494600160779105,
      "loss": 4.8882,
      "mean_token_accuracy": 0.21542707532644273,
      "num_tokens": 88584801.0,
      "step": 51070
    },
    {
      "entropy": 5.880360460281372,
      "epoch": 3.9738183232834077,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003494338615848537,
      "loss": 5.0043,
      "mean_token_accuracy": 0.20871713906526565,
      "num_tokens": 88593894.0,
      "step": 51075
    },
    {
      "entropy": 5.824681615829467,
      "epoch": 3.9742073526551254,
      "grad_norm": 1.5,
      "learning_rate": 0.0003494077059623816,
      "loss": 4.9248,
      "mean_token_accuracy": 0.21743262708187103,
      "num_tokens": 88601572.0,
      "step": 51080
    },
    {
      "entropy": 5.856284046173096,
      "epoch": 3.974596382026843,
      "grad_norm": 1.421875,
      "learning_rate": 0.00034938154921089097,
      "loss": 4.9832,
      "mean_token_accuracy": 0.21898461431264876,
      "num_tokens": 88610447.0,
      "step": 51085
    },
    {
      "entropy": 5.857293748855591,
      "epoch": 3.9749854113985608,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003493553913307788,
      "loss": 4.9754,
      "mean_token_accuracy": 0.2016680732369423,
      "num_tokens": 88619363.0,
      "step": 51090
    },
    {
      "entropy": 5.915251636505127,
      "epoch": 3.975374440770278,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034932923232244183,
      "loss": 5.0279,
      "mean_token_accuracy": 0.2086326077580452,
      "num_tokens": 88627797.0,
      "step": 51095
    },
    {
      "entropy": 5.95707139968872,
      "epoch": 3.9757634701419957,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034930307218627706,
      "loss": 5.016,
      "mean_token_accuracy": 0.21039636582136154,
      "num_tokens": 88637080.0,
      "step": 51100
    },
    {
      "entropy": 5.906458711624145,
      "epoch": 3.9761524995137134,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003492769109226815,
      "loss": 4.9761,
      "mean_token_accuracy": 0.2155553102493286,
      "num_tokens": 88646252.0,
      "step": 51105
    },
    {
      "entropy": 5.9199940204620365,
      "epoch": 3.9765415288854307,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034925074853205194,
      "loss": 5.0274,
      "mean_token_accuracy": 0.2006924957036972,
      "num_tokens": 88655629.0,
      "step": 51110
    },
    {
      "entropy": 5.967312955856324,
      "epoch": 3.9769305582571484,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034922458501478556,
      "loss": 5.0027,
      "mean_token_accuracy": 0.22319788634777069,
      "num_tokens": 88664501.0,
      "step": 51115
    },
    {
      "entropy": 5.85486249923706,
      "epoch": 3.977319587628866,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003491984203712792,
      "loss": 4.9169,
      "mean_token_accuracy": 0.2135988250374794,
      "num_tokens": 88672511.0,
      "step": 51120
    },
    {
      "entropy": 5.935390758514404,
      "epoch": 3.9777086170005838,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003491722546019299,
      "loss": 5.0493,
      "mean_token_accuracy": 0.2100460097193718,
      "num_tokens": 88680987.0,
      "step": 51125
    },
    {
      "entropy": 5.931789827346802,
      "epoch": 3.978097646372301,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003491460877071347,
      "loss": 5.055,
      "mean_token_accuracy": 0.20573009997606279,
      "num_tokens": 88690234.0,
      "step": 51130
    },
    {
      "entropy": 5.971969985961914,
      "epoch": 3.9784866757440187,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003491199196872906,
      "loss": 5.0389,
      "mean_token_accuracy": 0.2136078119277954,
      "num_tokens": 88698840.0,
      "step": 51135
    },
    {
      "entropy": 5.898155307769775,
      "epoch": 3.978875705115736,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003490937505427947,
      "loss": 4.9429,
      "mean_token_accuracy": 0.21416521072387695,
      "num_tokens": 88706993.0,
      "step": 51140
    },
    {
      "entropy": 5.8424286365509035,
      "epoch": 3.9792647344874537,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003490675802740441,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21641771346330643,
      "num_tokens": 88715629.0,
      "step": 51145
    },
    {
      "entropy": 5.902287149429322,
      "epoch": 3.9796537638591714,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003490414088814359,
      "loss": 4.9247,
      "mean_token_accuracy": 0.22157925069332124,
      "num_tokens": 88724874.0,
      "step": 51150
    },
    {
      "entropy": 5.937044954299926,
      "epoch": 3.980042793230889,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003490152363653671,
      "loss": 5.0274,
      "mean_token_accuracy": 0.2071641817688942,
      "num_tokens": 88733173.0,
      "step": 51155
    },
    {
      "entropy": 5.847459983825684,
      "epoch": 3.9804318226026068,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003489890627262349,
      "loss": 5.0056,
      "mean_token_accuracy": 0.21110834628343583,
      "num_tokens": 88741557.0,
      "step": 51160
    },
    {
      "entropy": 5.838502264022827,
      "epoch": 3.980820851974324,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034896288796443654,
      "loss": 4.913,
      "mean_token_accuracy": 0.21508955657482148,
      "num_tokens": 88749958.0,
      "step": 51165
    },
    {
      "entropy": 5.89239821434021,
      "epoch": 3.9812098813460417,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034893671208036895,
      "loss": 4.913,
      "mean_token_accuracy": 0.2232305496931076,
      "num_tokens": 88758034.0,
      "step": 51170
    },
    {
      "entropy": 5.9087950706481935,
      "epoch": 3.981598910717759,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034891053507442944,
      "loss": 4.9627,
      "mean_token_accuracy": 0.2170702889561653,
      "num_tokens": 88767101.0,
      "step": 51175
    },
    {
      "entropy": 5.821817827224732,
      "epoch": 3.9819879400894767,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003488843569470153,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2146038755774498,
      "num_tokens": 88775590.0,
      "step": 51180
    },
    {
      "entropy": 5.880641222000122,
      "epoch": 3.9823769694611943,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034885817769852355,
      "loss": 4.9797,
      "mean_token_accuracy": 0.21303718388080597,
      "num_tokens": 88784575.0,
      "step": 51185
    },
    {
      "entropy": 5.9298052310943605,
      "epoch": 3.982765998832912,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034883199732935156,
      "loss": 5.007,
      "mean_token_accuracy": 0.20884982049465178,
      "num_tokens": 88793479.0,
      "step": 51190
    },
    {
      "entropy": 5.917762660980225,
      "epoch": 3.9831550282046293,
      "grad_norm": 1.53125,
      "learning_rate": 0.00034880581583989655,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21762374937534332,
      "num_tokens": 88801155.0,
      "step": 51195
    },
    {
      "entropy": 5.845101404190063,
      "epoch": 3.983544057576347,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003487796332305557,
      "loss": 4.8926,
      "mean_token_accuracy": 0.22117125242948532,
      "num_tokens": 88810228.0,
      "step": 51200
    },
    {
      "entropy": 5.814281415939331,
      "epoch": 3.9839330869480647,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034875344950172635,
      "loss": 5.0015,
      "mean_token_accuracy": 0.21652389764785768,
      "num_tokens": 88818090.0,
      "step": 51205
    },
    {
      "entropy": 5.828624057769775,
      "epoch": 3.984322116319782,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003487272646538058,
      "loss": 4.9514,
      "mean_token_accuracy": 0.21039635092020034,
      "num_tokens": 88826683.0,
      "step": 51210
    },
    {
      "entropy": 5.848966979980469,
      "epoch": 3.9847111456914996,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034870107868719135,
      "loss": 4.9039,
      "mean_token_accuracy": 0.22026965618133545,
      "num_tokens": 88836205.0,
      "step": 51215
    },
    {
      "entropy": 5.859218406677246,
      "epoch": 3.9851001750632173,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003486748916022804,
      "loss": 4.925,
      "mean_token_accuracy": 0.21897902637720107,
      "num_tokens": 88845492.0,
      "step": 51220
    },
    {
      "entropy": 5.900882005691528,
      "epoch": 3.985489204434935,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003486487033994702,
      "loss": 4.9642,
      "mean_token_accuracy": 0.20634374022483826,
      "num_tokens": 88853862.0,
      "step": 51225
    },
    {
      "entropy": 5.816099214553833,
      "epoch": 3.9858782338066523,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034862251407915813,
      "loss": 4.9397,
      "mean_token_accuracy": 0.21504469066858292,
      "num_tokens": 88863436.0,
      "step": 51230
    },
    {
      "entropy": 5.863626480102539,
      "epoch": 3.98626726317837,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034859632364174167,
      "loss": 4.935,
      "mean_token_accuracy": 0.21866805404424666,
      "num_tokens": 88871917.0,
      "step": 51235
    },
    {
      "entropy": 5.922841215133667,
      "epoch": 3.9866562925500877,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000348570132087618,
      "loss": 5.0655,
      "mean_token_accuracy": 0.21212382316589357,
      "num_tokens": 88880634.0,
      "step": 51240
    },
    {
      "entropy": 5.928359079360962,
      "epoch": 3.987045321921805,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003485439394171847,
      "loss": 5.057,
      "mean_token_accuracy": 0.20951623767614364,
      "num_tokens": 88889542.0,
      "step": 51245
    },
    {
      "entropy": 5.912408494949341,
      "epoch": 3.9874343512935226,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034851774563083925,
      "loss": 5.0397,
      "mean_token_accuracy": 0.2040437787771225,
      "num_tokens": 88898238.0,
      "step": 51250
    },
    {
      "entropy": 5.845378303527832,
      "epoch": 3.9878233806652403,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003484915507289791,
      "loss": 4.9624,
      "mean_token_accuracy": 0.2179309144616127,
      "num_tokens": 88906591.0,
      "step": 51255
    },
    {
      "entropy": 5.812925577163696,
      "epoch": 3.988212410036958,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034846535471200153,
      "loss": 4.8835,
      "mean_token_accuracy": 0.22041261494159697,
      "num_tokens": 88915008.0,
      "step": 51260
    },
    {
      "entropy": 5.950852584838867,
      "epoch": 3.9886014394086753,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003484391575803041,
      "loss": 5.007,
      "mean_token_accuracy": 0.20693634450435638,
      "num_tokens": 88923936.0,
      "step": 51265
    },
    {
      "entropy": 5.897033309936523,
      "epoch": 3.988990468780393,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003484129593342844,
      "loss": 4.9164,
      "mean_token_accuracy": 0.21858056038618087,
      "num_tokens": 88932770.0,
      "step": 51270
    },
    {
      "entropy": 5.876382493972779,
      "epoch": 3.9893794981521102,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034838675997433987,
      "loss": 4.9694,
      "mean_token_accuracy": 0.2075872913002968,
      "num_tokens": 88941233.0,
      "step": 51275
    },
    {
      "entropy": 5.87048373222351,
      "epoch": 3.989768527523828,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003483605595008681,
      "loss": 5.0478,
      "mean_token_accuracy": 0.2018609717488289,
      "num_tokens": 88949527.0,
      "step": 51280
    },
    {
      "entropy": 5.84958701133728,
      "epoch": 3.9901575568955456,
      "grad_norm": 1.375,
      "learning_rate": 0.00034833435791426653,
      "loss": 4.9647,
      "mean_token_accuracy": 0.2102692022919655,
      "num_tokens": 88957961.0,
      "step": 51285
    },
    {
      "entropy": 5.964362478256225,
      "epoch": 3.9905465862672633,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034830815521493295,
      "loss": 5.0343,
      "mean_token_accuracy": 0.20921943634748458,
      "num_tokens": 88966551.0,
      "step": 51290
    },
    {
      "entropy": 5.963025093078613,
      "epoch": 3.9909356156389806,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003482819514032647,
      "loss": 4.9822,
      "mean_token_accuracy": 0.21536827981472015,
      "num_tokens": 88974994.0,
      "step": 51295
    },
    {
      "entropy": 5.894117927551269,
      "epoch": 3.9913246450106983,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034825574647965945,
      "loss": 4.8939,
      "mean_token_accuracy": 0.21932432651519776,
      "num_tokens": 88982664.0,
      "step": 51300
    },
    {
      "entropy": 5.913053321838379,
      "epoch": 3.991713674382416,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034822954044451485,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2165893018245697,
      "num_tokens": 88990901.0,
      "step": 51305
    },
    {
      "entropy": 5.853804683685302,
      "epoch": 3.9921027037541332,
      "grad_norm": 1.375,
      "learning_rate": 0.0003482033332982286,
      "loss": 4.9471,
      "mean_token_accuracy": 0.218132184445858,
      "num_tokens": 88999439.0,
      "step": 51310
    },
    {
      "entropy": 5.897780084609986,
      "epoch": 3.992491733125851,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003481771250411982,
      "loss": 5.0462,
      "mean_token_accuracy": 0.20589831471443176,
      "num_tokens": 89007819.0,
      "step": 51315
    },
    {
      "entropy": 5.849972486495972,
      "epoch": 3.9928807624975686,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003481509156738214,
      "loss": 4.8686,
      "mean_token_accuracy": 0.22297106087207794,
      "num_tokens": 89016620.0,
      "step": 51320
    },
    {
      "entropy": 5.871960926055908,
      "epoch": 3.9932697918692863,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034812470519649593,
      "loss": 4.9255,
      "mean_token_accuracy": 0.2118513196706772,
      "num_tokens": 89025327.0,
      "step": 51325
    },
    {
      "entropy": 5.923964405059815,
      "epoch": 3.9936588212410036,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003480984936096194,
      "loss": 5.0178,
      "mean_token_accuracy": 0.213408000767231,
      "num_tokens": 89034527.0,
      "step": 51330
    },
    {
      "entropy": 5.882249736785889,
      "epoch": 3.9940478506127213,
      "grad_norm": 1.53125,
      "learning_rate": 0.00034807228091358964,
      "loss": 4.9828,
      "mean_token_accuracy": 0.20976798981428146,
      "num_tokens": 89043923.0,
      "step": 51335
    },
    {
      "entropy": 5.870542573928833,
      "epoch": 3.994436879984439,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003480460671088043,
      "loss": 4.9169,
      "mean_token_accuracy": 0.22477118223905562,
      "num_tokens": 89052619.0,
      "step": 51340
    },
    {
      "entropy": 5.891075086593628,
      "epoch": 3.9948259093561562,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034801985219566124,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20686447918415068,
      "num_tokens": 89062228.0,
      "step": 51345
    },
    {
      "entropy": 5.97468523979187,
      "epoch": 3.995214938727874,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000347993636174558,
      "loss": 5.0741,
      "mean_token_accuracy": 0.2113485410809517,
      "num_tokens": 89070784.0,
      "step": 51350
    },
    {
      "entropy": 5.881267833709717,
      "epoch": 3.9956039680995916,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034796741904589263,
      "loss": 5.033,
      "mean_token_accuracy": 0.21018338054418564,
      "num_tokens": 89079758.0,
      "step": 51355
    },
    {
      "entropy": 5.859053611755371,
      "epoch": 3.9959929974713093,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034794120081006275,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21520709693431855,
      "num_tokens": 89088176.0,
      "step": 51360
    },
    {
      "entropy": 5.911710119247436,
      "epoch": 3.9963820268430266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003479149814674663,
      "loss": 4.9648,
      "mean_token_accuracy": 0.2098015785217285,
      "num_tokens": 89097329.0,
      "step": 51365
    },
    {
      "entropy": 5.897600364685059,
      "epoch": 3.9967710562147443,
      "grad_norm": 1.3125,
      "learning_rate": 0.000347888761018501,
      "loss": 4.9953,
      "mean_token_accuracy": 0.21509635895490647,
      "num_tokens": 89105973.0,
      "step": 51370
    },
    {
      "entropy": 5.8907938480377195,
      "epoch": 3.9971600855864615,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003478625394635648,
      "loss": 4.973,
      "mean_token_accuracy": 0.20917249023914336,
      "num_tokens": 89115618.0,
      "step": 51375
    },
    {
      "entropy": 5.929049444198609,
      "epoch": 3.997549114958179,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034783631680305554,
      "loss": 4.9432,
      "mean_token_accuracy": 0.21390130668878554,
      "num_tokens": 89124072.0,
      "step": 51380
    },
    {
      "entropy": 5.8912557601928714,
      "epoch": 3.997938144329897,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003478100930373711,
      "loss": 4.9918,
      "mean_token_accuracy": 0.21044548153877257,
      "num_tokens": 89132693.0,
      "step": 51385
    },
    {
      "entropy": 5.807271480560303,
      "epoch": 3.9983271737016146,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003477838681669094,
      "loss": 4.9261,
      "mean_token_accuracy": 0.21389484852552415,
      "num_tokens": 89141315.0,
      "step": 51390
    },
    {
      "entropy": 5.9639739990234375,
      "epoch": 3.9987162030733323,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003477576421920682,
      "loss": 5.125,
      "mean_token_accuracy": 0.2041578084230423,
      "num_tokens": 89148983.0,
      "step": 51395
    },
    {
      "entropy": 5.941881608963013,
      "epoch": 3.9991052324450496,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003477314151132458,
      "loss": 5.0684,
      "mean_token_accuracy": 0.2040597453713417,
      "num_tokens": 89158209.0,
      "step": 51400
    },
    {
      "entropy": 5.985758304595947,
      "epoch": 3.9994942618167673,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034770518693083986,
      "loss": 5.0997,
      "mean_token_accuracy": 0.20041133016347884,
      "num_tokens": 89166805.0,
      "step": 51405
    },
    {
      "entropy": 5.910163831710816,
      "epoch": 3.9998832911884845,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034767895764524845,
      "loss": 4.9794,
      "mean_token_accuracy": 0.21112971603870392,
      "num_tokens": 89175678.0,
      "step": 51410
    },
    {
      "entropy": 5.97582467397054,
      "epoch": 4.000233417623031,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00034765272725686947,
      "loss": 4.9776,
      "mean_token_accuracy": 0.21074849863847098,
      "num_tokens": 89183078.0,
      "step": 51415
    },
    {
      "entropy": 5.8420196056365965,
      "epoch": 4.000622446994748,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034762649576610105,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21914774626493455,
      "num_tokens": 89191249.0,
      "step": 51420
    },
    {
      "entropy": 5.8402751922607425,
      "epoch": 4.0010114763664655,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034760026317334113,
      "loss": 4.8632,
      "mean_token_accuracy": 0.2195607140660286,
      "num_tokens": 89199848.0,
      "step": 51425
    },
    {
      "entropy": 5.8676862716674805,
      "epoch": 4.001400505738183,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003475740294789878,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20982439815998077,
      "num_tokens": 89209258.0,
      "step": 51430
    },
    {
      "entropy": 5.90981969833374,
      "epoch": 4.001789535109901,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000347547794683439,
      "loss": 5.0042,
      "mean_token_accuracy": 0.21824139952659607,
      "num_tokens": 89218260.0,
      "step": 51435
    },
    {
      "entropy": 5.909048080444336,
      "epoch": 4.002178564481619,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003475215587870929,
      "loss": 4.9413,
      "mean_token_accuracy": 0.21335484087467194,
      "num_tokens": 89226214.0,
      "step": 51440
    },
    {
      "entropy": 5.877802944183349,
      "epoch": 4.002567593853336,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003474953217903477,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21453987509012223,
      "num_tokens": 89234936.0,
      "step": 51445
    },
    {
      "entropy": 5.877139234542847,
      "epoch": 4.002956623225053,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003474690836936013,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21407377272844313,
      "num_tokens": 89243516.0,
      "step": 51450
    },
    {
      "entropy": 5.898499965667725,
      "epoch": 4.003345652596771,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003474428444972519,
      "loss": 4.954,
      "mean_token_accuracy": 0.2105195105075836,
      "num_tokens": 89252680.0,
      "step": 51455
    },
    {
      "entropy": 5.880385971069336,
      "epoch": 4.0037346819684885,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003474166042016977,
      "loss": 4.8566,
      "mean_token_accuracy": 0.22143347710371017,
      "num_tokens": 89260965.0,
      "step": 51460
    },
    {
      "entropy": 5.984302663803101,
      "epoch": 4.004123711340206,
      "grad_norm": 1.46875,
      "learning_rate": 0.00034739036280733674,
      "loss": 5.0432,
      "mean_token_accuracy": 0.20459716767072678,
      "num_tokens": 89270269.0,
      "step": 51465
    },
    {
      "entropy": 5.871835517883301,
      "epoch": 4.004512740711924,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034736412031456714,
      "loss": 4.8775,
      "mean_token_accuracy": 0.2225031092762947,
      "num_tokens": 89278825.0,
      "step": 51470
    },
    {
      "entropy": 5.830157518386841,
      "epoch": 4.004901770083642,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034733787672378734,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21398176103830338,
      "num_tokens": 89287413.0,
      "step": 51475
    },
    {
      "entropy": 5.851902198791504,
      "epoch": 4.005290799455359,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003473116320353953,
      "loss": 4.9815,
      "mean_token_accuracy": 0.2110906034708023,
      "num_tokens": 89295379.0,
      "step": 51480
    },
    {
      "entropy": 5.862320184707642,
      "epoch": 4.005679828827076,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003472853862497895,
      "loss": 4.9448,
      "mean_token_accuracy": 0.20952885895967482,
      "num_tokens": 89303671.0,
      "step": 51485
    },
    {
      "entropy": 5.949629163742065,
      "epoch": 4.006068858198794,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003472591393673679,
      "loss": 4.9616,
      "mean_token_accuracy": 0.21784767359495164,
      "num_tokens": 89312177.0,
      "step": 51490
    },
    {
      "entropy": 5.839616012573242,
      "epoch": 4.0064578875705115,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034723289138852885,
      "loss": 4.9139,
      "mean_token_accuracy": 0.21895986348390578,
      "num_tokens": 89321002.0,
      "step": 51495
    },
    {
      "entropy": 5.84339075088501,
      "epoch": 4.006846916942229,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003472066423136707,
      "loss": 4.8985,
      "mean_token_accuracy": 0.21512024253606796,
      "num_tokens": 89329761.0,
      "step": 51500
    },
    {
      "entropy": 5.896760654449463,
      "epoch": 4.007235946313947,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003471803921431917,
      "loss": 4.9273,
      "mean_token_accuracy": 0.21654462516307832,
      "num_tokens": 89338345.0,
      "step": 51505
    },
    {
      "entropy": 5.902386665344238,
      "epoch": 4.007624975685665,
      "grad_norm": 1.4375,
      "learning_rate": 0.00034715414087749003,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21777443438768387,
      "num_tokens": 89347229.0,
      "step": 51510
    },
    {
      "entropy": 5.912940263748169,
      "epoch": 4.008014005057382,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003471278885169642,
      "loss": 4.9338,
      "mean_token_accuracy": 0.21891834437847138,
      "num_tokens": 89355590.0,
      "step": 51515
    },
    {
      "entropy": 5.937684059143066,
      "epoch": 4.008403034429099,
      "grad_norm": 1.421875,
      "learning_rate": 0.00034710163506201247,
      "loss": 4.9648,
      "mean_token_accuracy": 0.210394087433815,
      "num_tokens": 89363442.0,
      "step": 51520
    },
    {
      "entropy": 5.897476720809936,
      "epoch": 4.008792063800817,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034707538051303316,
      "loss": 4.9551,
      "mean_token_accuracy": 0.21237749755382537,
      "num_tokens": 89372363.0,
      "step": 51525
    },
    {
      "entropy": 5.933529996871949,
      "epoch": 4.0091810931725345,
      "grad_norm": 1.375,
      "learning_rate": 0.0003470491248704247,
      "loss": 5.038,
      "mean_token_accuracy": 0.20637766122817994,
      "num_tokens": 89380870.0,
      "step": 51530
    },
    {
      "entropy": 5.868474245071411,
      "epoch": 4.009570122544252,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034702286813458543,
      "loss": 4.8943,
      "mean_token_accuracy": 0.21358744651079178,
      "num_tokens": 89389047.0,
      "step": 51535
    },
    {
      "entropy": 5.843909788131714,
      "epoch": 4.00995915191597,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034699661030591374,
      "loss": 4.9147,
      "mean_token_accuracy": 0.22017130702733995,
      "num_tokens": 89397413.0,
      "step": 51540
    },
    {
      "entropy": 5.877579927444458,
      "epoch": 4.0103481812876876,
      "grad_norm": 1.375,
      "learning_rate": 0.0003469703513848081,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21067177802324294,
      "num_tokens": 89405507.0,
      "step": 51545
    },
    {
      "entropy": 5.861230993270874,
      "epoch": 4.010737210659404,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000346944091371667,
      "loss": 4.9446,
      "mean_token_accuracy": 0.2288581058382988,
      "num_tokens": 89413544.0,
      "step": 51550
    },
    {
      "entropy": 5.863749408721924,
      "epoch": 4.011126240031122,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003469178302668888,
      "loss": 4.9169,
      "mean_token_accuracy": 0.2223828077316284,
      "num_tokens": 89421567.0,
      "step": 51555
    },
    {
      "entropy": 5.794481229782105,
      "epoch": 4.01151526940284,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003468915680708719,
      "loss": 4.8132,
      "mean_token_accuracy": 0.23565987944602967,
      "num_tokens": 89430425.0,
      "step": 51560
    },
    {
      "entropy": 5.942691946029663,
      "epoch": 4.0119042987745575,
      "grad_norm": 1.359375,
      "learning_rate": 0.000346865304784015,
      "loss": 4.9976,
      "mean_token_accuracy": 0.20705384761095047,
      "num_tokens": 89438219.0,
      "step": 51565
    },
    {
      "entropy": 5.867607688903808,
      "epoch": 4.012293328146275,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003468390404067164,
      "loss": 4.9904,
      "mean_token_accuracy": 0.2116684392094612,
      "num_tokens": 89447145.0,
      "step": 51570
    },
    {
      "entropy": 5.968862152099609,
      "epoch": 4.012682357517993,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003468127749393747,
      "loss": 5.0456,
      "mean_token_accuracy": 0.2131548821926117,
      "num_tokens": 89457060.0,
      "step": 51575
    },
    {
      "entropy": 5.978016185760498,
      "epoch": 4.0130713868897105,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003467865083823885,
      "loss": 5.0401,
      "mean_token_accuracy": 0.21333996057510377,
      "num_tokens": 89465925.0,
      "step": 51580
    },
    {
      "entropy": 5.927683925628662,
      "epoch": 4.013460416261427,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003467602407361563,
      "loss": 4.9546,
      "mean_token_accuracy": 0.21414511799812316,
      "num_tokens": 89474863.0,
      "step": 51585
    },
    {
      "entropy": 5.866687679290772,
      "epoch": 4.013849445633145,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034673397200107663,
      "loss": 4.8874,
      "mean_token_accuracy": 0.2190448597073555,
      "num_tokens": 89483623.0,
      "step": 51590
    },
    {
      "entropy": 5.899594116210937,
      "epoch": 4.014238475004863,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003467077021775481,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21481382101774216,
      "num_tokens": 89492677.0,
      "step": 51595
    },
    {
      "entropy": 5.869575262069702,
      "epoch": 4.0146275043765804,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034668143126596933,
      "loss": 4.9346,
      "mean_token_accuracy": 0.2163792982697487,
      "num_tokens": 89501611.0,
      "step": 51600
    },
    {
      "entropy": 5.9594621658325195,
      "epoch": 4.015016533748298,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034665515926673903,
      "loss": 4.9929,
      "mean_token_accuracy": 0.21615516990423203,
      "num_tokens": 89510842.0,
      "step": 51605
    },
    {
      "entropy": 5.926382780075073,
      "epoch": 4.015405563120016,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003466288861802556,
      "loss": 4.937,
      "mean_token_accuracy": 0.21702005416154863,
      "num_tokens": 89520015.0,
      "step": 51610
    },
    {
      "entropy": 5.867415714263916,
      "epoch": 4.0157945924917335,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003466026120069179,
      "loss": 4.9556,
      "mean_token_accuracy": 0.2193633422255516,
      "num_tokens": 89529272.0,
      "step": 51615
    },
    {
      "entropy": 5.8528331279754635,
      "epoch": 4.01618362186345,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003465763367471245,
      "loss": 4.9165,
      "mean_token_accuracy": 0.21379163414239882,
      "num_tokens": 89537641.0,
      "step": 51620
    },
    {
      "entropy": 5.96091365814209,
      "epoch": 4.016572651235168,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034655006040127415,
      "loss": 5.0457,
      "mean_token_accuracy": 0.2057829350233078,
      "num_tokens": 89546454.0,
      "step": 51625
    },
    {
      "entropy": 5.885729694366455,
      "epoch": 4.016961680606886,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003465237829697655,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22202978879213334,
      "num_tokens": 89555292.0,
      "step": 51630
    },
    {
      "entropy": 5.942846202850342,
      "epoch": 4.017350709978603,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034649750445299725,
      "loss": 5.0466,
      "mean_token_accuracy": 0.21129234731197358,
      "num_tokens": 89563464.0,
      "step": 51635
    },
    {
      "entropy": 5.878468751907349,
      "epoch": 4.017739739350321,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003464712248513682,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2125246226787567,
      "num_tokens": 89572636.0,
      "step": 51640
    },
    {
      "entropy": 5.923302507400512,
      "epoch": 4.018128768722039,
      "grad_norm": 1.375,
      "learning_rate": 0.0003464449441652771,
      "loss": 5.0598,
      "mean_token_accuracy": 0.20053177177906037,
      "num_tokens": 89581080.0,
      "step": 51645
    },
    {
      "entropy": 5.90190372467041,
      "epoch": 4.018517798093756,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003464186623951227,
      "loss": 4.9255,
      "mean_token_accuracy": 0.2173517107963562,
      "num_tokens": 89589903.0,
      "step": 51650
    },
    {
      "entropy": 5.952067613601685,
      "epoch": 4.018906827465473,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003463923795413037,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22032854557037354,
      "num_tokens": 89597799.0,
      "step": 51655
    },
    {
      "entropy": 5.881106662750244,
      "epoch": 4.019295856837191,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034636609560421906,
      "loss": 5.0106,
      "mean_token_accuracy": 0.20971026569604873,
      "num_tokens": 89606301.0,
      "step": 51660
    },
    {
      "entropy": 5.895504188537598,
      "epoch": 4.019684886208909,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034633981058426744,
      "loss": 4.9696,
      "mean_token_accuracy": 0.21039099991321564,
      "num_tokens": 89616081.0,
      "step": 51665
    },
    {
      "entropy": 5.875878858566284,
      "epoch": 4.020073915580626,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003463135244818477,
      "loss": 4.8726,
      "mean_token_accuracy": 0.21888205856084825,
      "num_tokens": 89624014.0,
      "step": 51670
    },
    {
      "entropy": 5.8793755054473875,
      "epoch": 4.020462944952344,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003462872372973588,
      "loss": 4.9652,
      "mean_token_accuracy": 0.21547592878341676,
      "num_tokens": 89631636.0,
      "step": 51675
    },
    {
      "entropy": 5.825847959518432,
      "epoch": 4.020851974324062,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003462609490311996,
      "loss": 4.8655,
      "mean_token_accuracy": 0.22567352801561355,
      "num_tokens": 89639771.0,
      "step": 51680
    },
    {
      "entropy": 5.9022430896759035,
      "epoch": 4.021241003695779,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003462346596837689,
      "loss": 5.0264,
      "mean_token_accuracy": 0.2118979051709175,
      "num_tokens": 89648186.0,
      "step": 51685
    },
    {
      "entropy": 5.8689357280731205,
      "epoch": 4.021630033067496,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003462083692554655,
      "loss": 4.8785,
      "mean_token_accuracy": 0.22017496973276138,
      "num_tokens": 89656812.0,
      "step": 51690
    },
    {
      "entropy": 5.939738035202026,
      "epoch": 4.022019062439214,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034618207774668854,
      "loss": 4.908,
      "mean_token_accuracy": 0.22098247557878495,
      "num_tokens": 89664824.0,
      "step": 51695
    },
    {
      "entropy": 5.9381858825683596,
      "epoch": 4.022408091810932,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034615578515783687,
      "loss": 4.9981,
      "mean_token_accuracy": 0.21227751970291137,
      "num_tokens": 89674323.0,
      "step": 51700
    },
    {
      "entropy": 5.915309286117553,
      "epoch": 4.022797121182649,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034612949148930935,
      "loss": 4.9978,
      "mean_token_accuracy": 0.20927779674530028,
      "num_tokens": 89683233.0,
      "step": 51705
    },
    {
      "entropy": 5.890073919296265,
      "epoch": 4.023186150554367,
      "grad_norm": 1.421875,
      "learning_rate": 0.000346103196741505,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22136933654546737,
      "num_tokens": 89691704.0,
      "step": 51710
    },
    {
      "entropy": 5.953982305526734,
      "epoch": 4.023575179926085,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034607690091482284,
      "loss": 5.0003,
      "mean_token_accuracy": 0.21588967442512513,
      "num_tokens": 89700829.0,
      "step": 51715
    },
    {
      "entropy": 5.932560729980469,
      "epoch": 4.023964209297802,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003460506040096619,
      "loss": 4.9082,
      "mean_token_accuracy": 0.21344215720891951,
      "num_tokens": 89709026.0,
      "step": 51720
    },
    {
      "entropy": 5.850546026229859,
      "epoch": 4.024353238669519,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003460243060264211,
      "loss": 4.9471,
      "mean_token_accuracy": 0.21272530555725097,
      "num_tokens": 89717934.0,
      "step": 51725
    },
    {
      "entropy": 5.7960621356964115,
      "epoch": 4.024742268041237,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034599800696549944,
      "loss": 4.8345,
      "mean_token_accuracy": 0.21892877221107482,
      "num_tokens": 89726922.0,
      "step": 51730
    },
    {
      "entropy": 5.839752292633056,
      "epoch": 4.025131297412955,
      "grad_norm": 1.296875,
      "learning_rate": 0.000345971706827296,
      "loss": 4.9209,
      "mean_token_accuracy": 0.21678055077791214,
      "num_tokens": 89735656.0,
      "step": 51735
    },
    {
      "entropy": 5.8838310718536375,
      "epoch": 4.025520326784672,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034594540561221,
      "loss": 4.9918,
      "mean_token_accuracy": 0.2187206581234932,
      "num_tokens": 89744741.0,
      "step": 51740
    },
    {
      "entropy": 5.905480194091797,
      "epoch": 4.02590935615639,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034591910332064027,
      "loss": 4.9212,
      "mean_token_accuracy": 0.22085680067539215,
      "num_tokens": 89753588.0,
      "step": 51745
    },
    {
      "entropy": 6.008329296112061,
      "epoch": 4.026298385528108,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003458927999529861,
      "loss": 5.0225,
      "mean_token_accuracy": 0.21085150092840194,
      "num_tokens": 89761527.0,
      "step": 51750
    },
    {
      "entropy": 5.950730848312378,
      "epoch": 4.026687414899825,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034586649550964656,
      "loss": 4.9801,
      "mean_token_accuracy": 0.20928021073341369,
      "num_tokens": 89771524.0,
      "step": 51755
    },
    {
      "entropy": 5.884275245666504,
      "epoch": 4.027076444271542,
      "grad_norm": 1.484375,
      "learning_rate": 0.00034584018999102073,
      "loss": 4.9323,
      "mean_token_accuracy": 0.21991979479789733,
      "num_tokens": 89779664.0,
      "step": 51760
    },
    {
      "entropy": 5.990915679931641,
      "epoch": 4.02746547364326,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034581388339750775,
      "loss": 5.1526,
      "mean_token_accuracy": 0.2040310189127922,
      "num_tokens": 89787984.0,
      "step": 51765
    },
    {
      "entropy": 5.892055606842041,
      "epoch": 4.027854503014978,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003457875757295067,
      "loss": 4.971,
      "mean_token_accuracy": 0.2088671639561653,
      "num_tokens": 89796175.0,
      "step": 51770
    },
    {
      "entropy": 6.022421360015869,
      "epoch": 4.028243532386695,
      "grad_norm": 1.4375,
      "learning_rate": 0.00034576126698741697,
      "loss": 5.1434,
      "mean_token_accuracy": 0.20306421518325807,
      "num_tokens": 89804947.0,
      "step": 51775
    },
    {
      "entropy": 5.892018985748291,
      "epoch": 4.028632561758413,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003457349571716376,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2099792927503586,
      "num_tokens": 89812947.0,
      "step": 51780
    },
    {
      "entropy": 5.858810138702393,
      "epoch": 4.02902159113013,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034570864628256795,
      "loss": 4.9224,
      "mean_token_accuracy": 0.21623849123716354,
      "num_tokens": 89821897.0,
      "step": 51785
    },
    {
      "entropy": 5.921432781219482,
      "epoch": 4.029410620501848,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034568233432060705,
      "loss": 4.9998,
      "mean_token_accuracy": 0.21561850160360335,
      "num_tokens": 89830690.0,
      "step": 51790
    },
    {
      "entropy": 5.925419330596924,
      "epoch": 4.029799649873565,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003456560212861543,
      "loss": 4.9733,
      "mean_token_accuracy": 0.21075913459062576,
      "num_tokens": 89839494.0,
      "step": 51795
    },
    {
      "entropy": 5.865956878662109,
      "epoch": 4.030188679245283,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003456297071796087,
      "loss": 4.8429,
      "mean_token_accuracy": 0.22570459991693498,
      "num_tokens": 89848450.0,
      "step": 51800
    },
    {
      "entropy": 5.814689588546753,
      "epoch": 4.030577708617001,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034560339200136983,
      "loss": 4.9046,
      "mean_token_accuracy": 0.21962074637413026,
      "num_tokens": 89856950.0,
      "step": 51805
    },
    {
      "entropy": 5.813710451126099,
      "epoch": 4.030966737988718,
      "grad_norm": 1.421875,
      "learning_rate": 0.00034557707575183684,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22321503162384032,
      "num_tokens": 89865889.0,
      "step": 51810
    },
    {
      "entropy": 5.92737979888916,
      "epoch": 4.031355767360436,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034555075843140917,
      "loss": 4.9162,
      "mean_token_accuracy": 0.22035107910633087,
      "num_tokens": 89874275.0,
      "step": 51815
    },
    {
      "entropy": 5.93666090965271,
      "epoch": 4.031744796732153,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034552444004048587,
      "loss": 5.0054,
      "mean_token_accuracy": 0.21510865539312363,
      "num_tokens": 89882975.0,
      "step": 51820
    },
    {
      "entropy": 5.860597038269043,
      "epoch": 4.032133826103871,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034549812057946657,
      "loss": 4.9557,
      "mean_token_accuracy": 0.21119217574596405,
      "num_tokens": 89892111.0,
      "step": 51825
    },
    {
      "entropy": 5.8945800304412845,
      "epoch": 4.032522855475588,
      "grad_norm": 1.375,
      "learning_rate": 0.0003454718000487505,
      "loss": 4.972,
      "mean_token_accuracy": 0.21828320473432541,
      "num_tokens": 89900731.0,
      "step": 51830
    },
    {
      "entropy": 5.91560959815979,
      "epoch": 4.032911884847306,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003454454784487369,
      "loss": 4.9584,
      "mean_token_accuracy": 0.21158554702997207,
      "num_tokens": 89909360.0,
      "step": 51835
    },
    {
      "entropy": 5.959977006912231,
      "epoch": 4.033300914219024,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003454191557798254,
      "loss": 5.0078,
      "mean_token_accuracy": 0.20687488466501236,
      "num_tokens": 89917425.0,
      "step": 51840
    },
    {
      "entropy": 5.957312440872192,
      "epoch": 4.033689943590741,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034539283204241525,
      "loss": 5.0671,
      "mean_token_accuracy": 0.20787511467933656,
      "num_tokens": 89926385.0,
      "step": 51845
    },
    {
      "entropy": 5.951608943939209,
      "epoch": 4.034078972962459,
      "grad_norm": 1.421875,
      "learning_rate": 0.00034536650723690596,
      "loss": 4.9151,
      "mean_token_accuracy": 0.21929338425397873,
      "num_tokens": 89935485.0,
      "step": 51850
    },
    {
      "entropy": 5.948643779754638,
      "epoch": 4.034468002334176,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034534018136369687,
      "loss": 4.9384,
      "mean_token_accuracy": 0.21792497634887695,
      "num_tokens": 89943736.0,
      "step": 51855
    },
    {
      "entropy": 5.787169504165649,
      "epoch": 4.034857031705894,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003453138544231875,
      "loss": 4.864,
      "mean_token_accuracy": 0.22355013489723205,
      "num_tokens": 89952352.0,
      "step": 51860
    },
    {
      "entropy": 5.881773519515991,
      "epoch": 4.035246061077611,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003452875264157774,
      "loss": 4.9452,
      "mean_token_accuracy": 0.2107370138168335,
      "num_tokens": 89960485.0,
      "step": 51865
    },
    {
      "entropy": 5.883715534210205,
      "epoch": 4.035635090449329,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034526119734186586,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21434458941221238,
      "num_tokens": 89969395.0,
      "step": 51870
    },
    {
      "entropy": 5.927192735671997,
      "epoch": 4.036024119821047,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034523486720185246,
      "loss": 4.9478,
      "mean_token_accuracy": 0.2138140991330147,
      "num_tokens": 89978078.0,
      "step": 51875
    },
    {
      "entropy": 5.908332061767578,
      "epoch": 4.036413149192764,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034520853599613675,
      "loss": 4.9564,
      "mean_token_accuracy": 0.2150764063000679,
      "num_tokens": 89986294.0,
      "step": 51880
    },
    {
      "entropy": 5.9305031299591064,
      "epoch": 4.036802178564481,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034518220372511835,
      "loss": 4.9657,
      "mean_token_accuracy": 0.208026622235775,
      "num_tokens": 89994124.0,
      "step": 51885
    },
    {
      "entropy": 5.872569942474366,
      "epoch": 4.037191207936199,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003451558703891967,
      "loss": 4.8988,
      "mean_token_accuracy": 0.2195541650056839,
      "num_tokens": 90002624.0,
      "step": 51890
    },
    {
      "entropy": 5.9115403175354,
      "epoch": 4.037580237307917,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003451295359887713,
      "loss": 4.9142,
      "mean_token_accuracy": 0.2196933165192604,
      "num_tokens": 90012073.0,
      "step": 51895
    },
    {
      "entropy": 5.958462858200074,
      "epoch": 4.037969266679634,
      "grad_norm": 1.375,
      "learning_rate": 0.0003451032005242418,
      "loss": 4.9446,
      "mean_token_accuracy": 0.2085406392812729,
      "num_tokens": 90020790.0,
      "step": 51900
    },
    {
      "entropy": 5.89507303237915,
      "epoch": 4.038358296051352,
      "grad_norm": 1.484375,
      "learning_rate": 0.00034507686399600786,
      "loss": 4.911,
      "mean_token_accuracy": 0.21860469430685042,
      "num_tokens": 90029318.0,
      "step": 51905
    },
    {
      "entropy": 5.892300939559936,
      "epoch": 4.03874732542307,
      "grad_norm": 1.375,
      "learning_rate": 0.00034505052640446907,
      "loss": 4.9064,
      "mean_token_accuracy": 0.22318902909755706,
      "num_tokens": 90037419.0,
      "step": 51910
    },
    {
      "entropy": 5.877291536331176,
      "epoch": 4.039136354794787,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000345024187750025,
      "loss": 4.9424,
      "mean_token_accuracy": 0.21932954341173172,
      "num_tokens": 90045806.0,
      "step": 51915
    },
    {
      "entropy": 5.898646831512451,
      "epoch": 4.039525384166504,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034499784803307534,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2210891529917717,
      "num_tokens": 90053969.0,
      "step": 51920
    },
    {
      "entropy": 5.8642332553863525,
      "epoch": 4.039914413538222,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003449715072540198,
      "loss": 4.8808,
      "mean_token_accuracy": 0.22469929605722427,
      "num_tokens": 90063310.0,
      "step": 51925
    },
    {
      "entropy": 5.834958457946778,
      "epoch": 4.04030344290994,
      "grad_norm": 1.765625,
      "learning_rate": 0.00034494516541325795,
      "loss": 5.0029,
      "mean_token_accuracy": 0.21384450048208237,
      "num_tokens": 90072315.0,
      "step": 51930
    },
    {
      "entropy": 5.791952466964721,
      "epoch": 4.040692472281657,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003449188225111895,
      "loss": 4.7957,
      "mean_token_accuracy": 0.22513895779848098,
      "num_tokens": 90079817.0,
      "step": 51935
    },
    {
      "entropy": 5.893533563613891,
      "epoch": 4.041081501653375,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034489247854821426,
      "loss": 4.9791,
      "mean_token_accuracy": 0.20553414821624755,
      "num_tokens": 90089188.0,
      "step": 51940
    },
    {
      "entropy": 5.92146863937378,
      "epoch": 4.041470531025093,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00034486613352473195,
      "loss": 4.9529,
      "mean_token_accuracy": 0.2129001185297966,
      "num_tokens": 90098186.0,
      "step": 51945
    },
    {
      "entropy": 5.811070394515991,
      "epoch": 4.04185956039681,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034483978744114224,
      "loss": 4.8376,
      "mean_token_accuracy": 0.2223162680864334,
      "num_tokens": 90106136.0,
      "step": 51950
    },
    {
      "entropy": 5.860670137405395,
      "epoch": 4.042248589768527,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003448134402978449,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21533261388540267,
      "num_tokens": 90114754.0,
      "step": 51955
    },
    {
      "entropy": 5.875341939926147,
      "epoch": 4.042637619140245,
      "grad_norm": 1.421875,
      "learning_rate": 0.00034478709209523974,
      "loss": 4.9078,
      "mean_token_accuracy": 0.21165003180503844,
      "num_tokens": 90122994.0,
      "step": 51960
    },
    {
      "entropy": 5.996651124954224,
      "epoch": 4.043026648511963,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003447607428337265,
      "loss": 5.0645,
      "mean_token_accuracy": 0.20892146825790406,
      "num_tokens": 90132083.0,
      "step": 51965
    },
    {
      "entropy": 5.884886360168457,
      "epoch": 4.04341567788368,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00034473439251370505,
      "loss": 4.857,
      "mean_token_accuracy": 0.21886216402053832,
      "num_tokens": 90140790.0,
      "step": 51970
    },
    {
      "entropy": 5.806748867034912,
      "epoch": 4.043804707255398,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003447080411355752,
      "loss": 4.8206,
      "mean_token_accuracy": 0.22117093354463577,
      "num_tokens": 90149023.0,
      "step": 51975
    },
    {
      "entropy": 5.934732437133789,
      "epoch": 4.044193736627116,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034468168869973673,
      "loss": 5.0283,
      "mean_token_accuracy": 0.20419780015945435,
      "num_tokens": 90157274.0,
      "step": 51980
    },
    {
      "entropy": 5.870685195922851,
      "epoch": 4.0445827659988325,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034465533520658967,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21150976419448853,
      "num_tokens": 90166374.0,
      "step": 51985
    },
    {
      "entropy": 5.897975206375122,
      "epoch": 4.04497179537055,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034462898065653366,
      "loss": 4.9395,
      "mean_token_accuracy": 0.2129041776061058,
      "num_tokens": 90175191.0,
      "step": 51990
    },
    {
      "entropy": 5.907422780990601,
      "epoch": 4.045360824742268,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003446026250499687,
      "loss": 4.9986,
      "mean_token_accuracy": 0.20995161384344102,
      "num_tokens": 90183637.0,
      "step": 51995
    },
    {
      "entropy": 5.960235595703125,
      "epoch": 4.045749854113986,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003445762683872946,
      "loss": 4.9627,
      "mean_token_accuracy": 0.21673441529273987,
      "num_tokens": 90192258.0,
      "step": 52000
    },
    {
      "entropy": 6.014357423782348,
      "epoch": 4.046138883485703,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003445499106689115,
      "loss": 5.0449,
      "mean_token_accuracy": 0.21231255382299424,
      "num_tokens": 90201154.0,
      "step": 52005
    },
    {
      "entropy": 5.8809507369995115,
      "epoch": 4.046527912857421,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034452355189521915,
      "loss": 4.86,
      "mean_token_accuracy": 0.2281960055232048,
      "num_tokens": 90209343.0,
      "step": 52010
    },
    {
      "entropy": 5.9388152122497555,
      "epoch": 4.046916942229139,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003444971920666176,
      "loss": 4.9741,
      "mean_token_accuracy": 0.2047293797135353,
      "num_tokens": 90217689.0,
      "step": 52015
    },
    {
      "entropy": 5.919192218780518,
      "epoch": 4.0473059716008555,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003444708311835068,
      "loss": 4.9818,
      "mean_token_accuracy": 0.21262185275554657,
      "num_tokens": 90226855.0,
      "step": 52020
    },
    {
      "entropy": 5.852221345901489,
      "epoch": 4.047695000972573,
      "grad_norm": 1.375,
      "learning_rate": 0.00034444446924628664,
      "loss": 4.9001,
      "mean_token_accuracy": 0.22046876400709153,
      "num_tokens": 90236555.0,
      "step": 52025
    },
    {
      "entropy": 5.995780181884766,
      "epoch": 4.048084030344291,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034441810625535725,
      "loss": 5.0652,
      "mean_token_accuracy": 0.198277547955513,
      "num_tokens": 90245801.0,
      "step": 52030
    },
    {
      "entropy": 5.953625965118408,
      "epoch": 4.048473059716009,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003443917422111185,
      "loss": 4.9388,
      "mean_token_accuracy": 0.20780290216207503,
      "num_tokens": 90253892.0,
      "step": 52035
    },
    {
      "entropy": 5.919342517852783,
      "epoch": 4.048862089087726,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034436537711397054,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2163647249341011,
      "num_tokens": 90263245.0,
      "step": 52040
    },
    {
      "entropy": 5.875862884521484,
      "epoch": 4.049251118459444,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003443390109643134,
      "loss": 4.9426,
      "mean_token_accuracy": 0.21816371977329255,
      "num_tokens": 90271813.0,
      "step": 52045
    },
    {
      "entropy": 5.884547376632691,
      "epoch": 4.049640147831162,
      "grad_norm": 1.5,
      "learning_rate": 0.0003443126437625472,
      "loss": 4.986,
      "mean_token_accuracy": 0.21461054384708406,
      "num_tokens": 90280682.0,
      "step": 52050
    },
    {
      "entropy": 5.922876644134521,
      "epoch": 4.0500291772028785,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003442862755090719,
      "loss": 4.9417,
      "mean_token_accuracy": 0.21894809305667878,
      "num_tokens": 90289054.0,
      "step": 52055
    },
    {
      "entropy": 5.8500096797943115,
      "epoch": 4.050418206574596,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034425990620428763,
      "loss": 4.9043,
      "mean_token_accuracy": 0.21728812009096146,
      "num_tokens": 90297636.0,
      "step": 52060
    },
    {
      "entropy": 5.870144462585449,
      "epoch": 4.050807235946314,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003442335358485946,
      "loss": 4.9708,
      "mean_token_accuracy": 0.21283353865146637,
      "num_tokens": 90306653.0,
      "step": 52065
    },
    {
      "entropy": 6.007606124877929,
      "epoch": 4.051196265318032,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003442071644423928,
      "loss": 5.1365,
      "mean_token_accuracy": 0.20033505707979202,
      "num_tokens": 90316325.0,
      "step": 52070
    },
    {
      "entropy": 5.913882637023926,
      "epoch": 4.051585294689749,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003441807919860824,
      "loss": 4.9927,
      "mean_token_accuracy": 0.2099497362971306,
      "num_tokens": 90325559.0,
      "step": 52075
    },
    {
      "entropy": 5.934242296218872,
      "epoch": 4.051974324061467,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00034415441848006364,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21674539744853974,
      "num_tokens": 90333908.0,
      "step": 52080
    },
    {
      "entropy": 5.961170768737793,
      "epoch": 4.052363353433184,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034412804392473665,
      "loss": 4.967,
      "mean_token_accuracy": 0.21664415150880814,
      "num_tokens": 90342442.0,
      "step": 52085
    },
    {
      "entropy": 5.815949535369873,
      "epoch": 4.0527523828049015,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003441016683205016,
      "loss": 4.8514,
      "mean_token_accuracy": 0.2209737092256546,
      "num_tokens": 90351320.0,
      "step": 52090
    },
    {
      "entropy": 5.817745685577393,
      "epoch": 4.053141412176619,
      "grad_norm": 1.4375,
      "learning_rate": 0.00034407529166775874,
      "loss": 4.8367,
      "mean_token_accuracy": 0.2221623659133911,
      "num_tokens": 90359867.0,
      "step": 52095
    },
    {
      "entropy": 5.86921501159668,
      "epoch": 4.053530441548337,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003440489139669083,
      "loss": 4.922,
      "mean_token_accuracy": 0.21708546578884125,
      "num_tokens": 90369433.0,
      "step": 52100
    },
    {
      "entropy": 5.887403249740601,
      "epoch": 4.053919470920055,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034402253521835036,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2166532412171364,
      "num_tokens": 90379055.0,
      "step": 52105
    },
    {
      "entropy": 5.904199075698853,
      "epoch": 4.054308500291772,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003439961554224855,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20952796339988708,
      "num_tokens": 90387478.0,
      "step": 52110
    },
    {
      "entropy": 5.920164632797241,
      "epoch": 4.05469752966349,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003439697745797137,
      "loss": 4.9239,
      "mean_token_accuracy": 0.2136171743273735,
      "num_tokens": 90396267.0,
      "step": 52115
    },
    {
      "entropy": 5.839287519454956,
      "epoch": 4.055086559035207,
      "grad_norm": 1.921875,
      "learning_rate": 0.00034394339269043533,
      "loss": 4.885,
      "mean_token_accuracy": 0.2294818639755249,
      "num_tokens": 90405351.0,
      "step": 52120
    },
    {
      "entropy": 5.923763561248779,
      "epoch": 4.0554755884069245,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003439170097550507,
      "loss": 5.0134,
      "mean_token_accuracy": 0.21471408158540725,
      "num_tokens": 90413446.0,
      "step": 52125
    },
    {
      "entropy": 5.928883171081543,
      "epoch": 4.055864617778642,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003438906257739602,
      "loss": 4.9897,
      "mean_token_accuracy": 0.21087175011634826,
      "num_tokens": 90421870.0,
      "step": 52130
    },
    {
      "entropy": 5.970314025878906,
      "epoch": 4.05625364715036,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003438642407475641,
      "loss": 5.0405,
      "mean_token_accuracy": 0.21086794883012772,
      "num_tokens": 90430229.0,
      "step": 52135
    },
    {
      "entropy": 5.965768432617187,
      "epoch": 4.0566426765220776,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003438378546762627,
      "loss": 5.0137,
      "mean_token_accuracy": 0.210268434882164,
      "num_tokens": 90439205.0,
      "step": 52140
    },
    {
      "entropy": 5.948880577087403,
      "epoch": 4.057031705893795,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003438114675604565,
      "loss": 4.9867,
      "mean_token_accuracy": 0.20920566320419312,
      "num_tokens": 90447038.0,
      "step": 52145
    },
    {
      "entropy": 5.853871488571167,
      "epoch": 4.057420735265513,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003437850794005457,
      "loss": 4.8942,
      "mean_token_accuracy": 0.21822089403867723,
      "num_tokens": 90455231.0,
      "step": 52150
    },
    {
      "entropy": 5.872321128845215,
      "epoch": 4.05780976463723,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003437586901969309,
      "loss": 4.9409,
      "mean_token_accuracy": 0.2098012775182724,
      "num_tokens": 90464900.0,
      "step": 52155
    },
    {
      "entropy": 5.9274921894073485,
      "epoch": 4.0581987940089475,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003437322999500124,
      "loss": 4.9914,
      "mean_token_accuracy": 0.2078676387667656,
      "num_tokens": 90474354.0,
      "step": 52160
    },
    {
      "entropy": 5.846142911911011,
      "epoch": 4.058587823380665,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034370590866019063,
      "loss": 4.9292,
      "mean_token_accuracy": 0.21790043711662294,
      "num_tokens": 90482884.0,
      "step": 52165
    },
    {
      "entropy": 5.859018421173095,
      "epoch": 4.058976852752383,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000343679516327866,
      "loss": 5.0037,
      "mean_token_accuracy": 0.20999144911766052,
      "num_tokens": 90491626.0,
      "step": 52170
    },
    {
      "entropy": 5.883349084854126,
      "epoch": 4.0593658821241005,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003436531229534391,
      "loss": 4.915,
      "mean_token_accuracy": 0.21653378158807754,
      "num_tokens": 90499686.0,
      "step": 52175
    },
    {
      "entropy": 5.868328475952149,
      "epoch": 4.059754911495818,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003436267285373103,
      "loss": 4.9302,
      "mean_token_accuracy": 0.2137988477945328,
      "num_tokens": 90508523.0,
      "step": 52180
    },
    {
      "entropy": 5.929788732528687,
      "epoch": 4.060143940867536,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034360033307988014,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22054191082715988,
      "num_tokens": 90517383.0,
      "step": 52185
    },
    {
      "entropy": 5.9340404033660885,
      "epoch": 4.060532970239253,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034357393658154914,
      "loss": 5.0021,
      "mean_token_accuracy": 0.2118535116314888,
      "num_tokens": 90526052.0,
      "step": 52190
    },
    {
      "entropy": 5.943279170989991,
      "epoch": 4.0609219996109704,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003435475390427178,
      "loss": 5.0152,
      "mean_token_accuracy": 0.20467731207609177,
      "num_tokens": 90535521.0,
      "step": 52195
    },
    {
      "entropy": 5.925900602340699,
      "epoch": 4.061311028982688,
      "grad_norm": 1.5,
      "learning_rate": 0.00034352114046378655,
      "loss": 4.9396,
      "mean_token_accuracy": 0.2136240139603615,
      "num_tokens": 90544660.0,
      "step": 52200
    },
    {
      "entropy": 5.87631573677063,
      "epoch": 4.061700058354406,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034349474084515607,
      "loss": 4.9491,
      "mean_token_accuracy": 0.22159697711467743,
      "num_tokens": 90553648.0,
      "step": 52205
    },
    {
      "entropy": 5.886921977996826,
      "epoch": 4.0620890877261235,
      "grad_norm": 1.421875,
      "learning_rate": 0.00034346834018722706,
      "loss": 4.9893,
      "mean_token_accuracy": 0.205280601978302,
      "num_tokens": 90561819.0,
      "step": 52210
    },
    {
      "entropy": 5.888171720504761,
      "epoch": 4.062478117097841,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034344193849039997,
      "loss": 4.9881,
      "mean_token_accuracy": 0.2101266413927078,
      "num_tokens": 90570599.0,
      "step": 52215
    },
    {
      "entropy": 5.874212026596069,
      "epoch": 4.062867146469558,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003434155357550753,
      "loss": 5.0028,
      "mean_token_accuracy": 0.2181297391653061,
      "num_tokens": 90579620.0,
      "step": 52220
    },
    {
      "entropy": 5.926593542098999,
      "epoch": 4.063256175841276,
      "grad_norm": 1.25,
      "learning_rate": 0.00034338913198165373,
      "loss": 4.9812,
      "mean_token_accuracy": 0.2083246111869812,
      "num_tokens": 90589355.0,
      "step": 52225
    },
    {
      "entropy": 5.855221319198608,
      "epoch": 4.063645205212993,
      "grad_norm": 1.46875,
      "learning_rate": 0.000343362727170536,
      "loss": 4.9277,
      "mean_token_accuracy": 0.214627043902874,
      "num_tokens": 90597371.0,
      "step": 52230
    },
    {
      "entropy": 5.886782693862915,
      "epoch": 4.064034234584711,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003433363213221227,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21431716084480285,
      "num_tokens": 90606252.0,
      "step": 52235
    },
    {
      "entropy": 5.930133533477783,
      "epoch": 4.064423263956429,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034330991443681444,
      "loss": 4.917,
      "mean_token_accuracy": 0.21605076044797897,
      "num_tokens": 90614691.0,
      "step": 52240
    },
    {
      "entropy": 5.932524824142456,
      "epoch": 4.0648122933281465,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000343283506515012,
      "loss": 4.9533,
      "mean_token_accuracy": 0.2125212848186493,
      "num_tokens": 90622711.0,
      "step": 52245
    },
    {
      "entropy": 5.934361267089844,
      "epoch": 4.065201322699864,
      "grad_norm": 1.375,
      "learning_rate": 0.00034325709755711606,
      "loss": 5.0172,
      "mean_token_accuracy": 0.20878956019878386,
      "num_tokens": 90631490.0,
      "step": 52250
    },
    {
      "entropy": 5.87332239151001,
      "epoch": 4.065590352071581,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034323068756352725,
      "loss": 4.8829,
      "mean_token_accuracy": 0.2188686177134514,
      "num_tokens": 90640022.0,
      "step": 52255
    },
    {
      "entropy": 5.821243524551392,
      "epoch": 4.065979381443299,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003432042765346464,
      "loss": 4.8268,
      "mean_token_accuracy": 0.2282026842236519,
      "num_tokens": 90648589.0,
      "step": 52260
    },
    {
      "entropy": 5.899890518188476,
      "epoch": 4.066368410815016,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003431778644708742,
      "loss": 4.9647,
      "mean_token_accuracy": 0.2132944092154503,
      "num_tokens": 90657426.0,
      "step": 52265
    },
    {
      "entropy": 5.879000043869018,
      "epoch": 4.066757440186734,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003431514513726114,
      "loss": 4.9568,
      "mean_token_accuracy": 0.20911829471588134,
      "num_tokens": 90665445.0,
      "step": 52270
    },
    {
      "entropy": 5.914884042739868,
      "epoch": 4.067146469558452,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003431250372402588,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21387426257133485,
      "num_tokens": 90674442.0,
      "step": 52275
    },
    {
      "entropy": 5.957050561904907,
      "epoch": 4.0675354989301695,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034309862207421724,
      "loss": 5.0187,
      "mean_token_accuracy": 0.20510548800230027,
      "num_tokens": 90682907.0,
      "step": 52280
    },
    {
      "entropy": 5.948835563659668,
      "epoch": 4.067924528301887,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00034307220587488743,
      "loss": 4.988,
      "mean_token_accuracy": 0.20683171302080156,
      "num_tokens": 90691842.0,
      "step": 52285
    },
    {
      "entropy": 5.8799725532531735,
      "epoch": 4.068313557673604,
      "grad_norm": 1.46875,
      "learning_rate": 0.00034304578864267026,
      "loss": 4.9009,
      "mean_token_accuracy": 0.21437583416700362,
      "num_tokens": 90700480.0,
      "step": 52290
    },
    {
      "entropy": 5.94196720123291,
      "epoch": 4.068702587045322,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00034301937037796654,
      "loss": 5.0397,
      "mean_token_accuracy": 0.21065893024206161,
      "num_tokens": 90709822.0,
      "step": 52295
    },
    {
      "entropy": 5.9230893611907955,
      "epoch": 4.069091616417039,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034299295108117716,
      "loss": 5.0463,
      "mean_token_accuracy": 0.20771856158971785,
      "num_tokens": 90718432.0,
      "step": 52300
    },
    {
      "entropy": 5.86773362159729,
      "epoch": 4.069480645788757,
      "grad_norm": 1.25,
      "learning_rate": 0.00034296653075270296,
      "loss": 5.0007,
      "mean_token_accuracy": 0.210592183470726,
      "num_tokens": 90727318.0,
      "step": 52305
    },
    {
      "entropy": 5.925537109375,
      "epoch": 4.069869675160475,
      "grad_norm": 1.46875,
      "learning_rate": 0.00034294010939294486,
      "loss": 4.9284,
      "mean_token_accuracy": 0.21462321728467942,
      "num_tokens": 90735864.0,
      "step": 52310
    },
    {
      "entropy": 5.867591476440429,
      "epoch": 4.0702587045321925,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034291368700230376,
      "loss": 4.8772,
      "mean_token_accuracy": 0.21418919861316682,
      "num_tokens": 90744600.0,
      "step": 52315
    },
    {
      "entropy": 5.960395097732544,
      "epoch": 4.070647733903909,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003428872635811804,
      "loss": 5.0189,
      "mean_token_accuracy": 0.2099539503455162,
      "num_tokens": 90753409.0,
      "step": 52320
    },
    {
      "entropy": 5.915169191360474,
      "epoch": 4.071036763275627,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000342860839129976,
      "loss": 4.913,
      "mean_token_accuracy": 0.21352265924215316,
      "num_tokens": 90762583.0,
      "step": 52325
    },
    {
      "entropy": 5.995546483993531,
      "epoch": 4.071425792647345,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003428344136490914,
      "loss": 5.0652,
      "mean_token_accuracy": 0.2130970135331154,
      "num_tokens": 90771270.0,
      "step": 52330
    },
    {
      "entropy": 5.891939401626587,
      "epoch": 4.071814822019062,
      "grad_norm": 1.375,
      "learning_rate": 0.00034280798713892744,
      "loss": 4.9923,
      "mean_token_accuracy": 0.21517182141542435,
      "num_tokens": 90780414.0,
      "step": 52335
    },
    {
      "entropy": 5.895632600784301,
      "epoch": 4.07220385139078,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003427815595998853,
      "loss": 5.0382,
      "mean_token_accuracy": 0.2129811942577362,
      "num_tokens": 90789688.0,
      "step": 52340
    },
    {
      "entropy": 5.932392883300781,
      "epoch": 4.072592880762498,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003427551310323658,
      "loss": 5.0233,
      "mean_token_accuracy": 0.20880593955516816,
      "num_tokens": 90798690.0,
      "step": 52345
    },
    {
      "entropy": 5.905458784103393,
      "epoch": 4.0729819101342155,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034272870143677015,
      "loss": 4.885,
      "mean_token_accuracy": 0.2222653791308403,
      "num_tokens": 90806642.0,
      "step": 52350
    },
    {
      "entropy": 5.8652630805969235,
      "epoch": 4.073370939505932,
      "grad_norm": 1.484375,
      "learning_rate": 0.00034270227081349913,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20596519261598586,
      "num_tokens": 90814836.0,
      "step": 52355
    },
    {
      "entropy": 5.890783452987671,
      "epoch": 4.07375996887765,
      "grad_norm": 1.359375,
      "learning_rate": 0.000342675839162954,
      "loss": 5.0195,
      "mean_token_accuracy": 0.21460380256175995,
      "num_tokens": 90823760.0,
      "step": 52360
    },
    {
      "entropy": 5.906784439086914,
      "epoch": 4.074148998249368,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034264940648553565,
      "loss": 4.9734,
      "mean_token_accuracy": 0.2226565584540367,
      "num_tokens": 90832588.0,
      "step": 52365
    },
    {
      "entropy": 5.908009243011475,
      "epoch": 4.074538027621085,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003426229727816453,
      "loss": 4.9544,
      "mean_token_accuracy": 0.2203405722975731,
      "num_tokens": 90841289.0,
      "step": 52370
    },
    {
      "entropy": 5.860874605178833,
      "epoch": 4.074927056992803,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003425965380516839,
      "loss": 4.9367,
      "mean_token_accuracy": 0.21664959341287612,
      "num_tokens": 90850180.0,
      "step": 52375
    },
    {
      "entropy": 5.897396087646484,
      "epoch": 4.075316086364521,
      "grad_norm": 1.5,
      "learning_rate": 0.0003425701022960527,
      "loss": 5.032,
      "mean_token_accuracy": 0.20738847255706788,
      "num_tokens": 90858503.0,
      "step": 52380
    },
    {
      "entropy": 5.892825794219971,
      "epoch": 4.0757051157362385,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00034254366551515276,
      "loss": 4.9518,
      "mean_token_accuracy": 0.21447215974330902,
      "num_tokens": 90866834.0,
      "step": 52385
    },
    {
      "entropy": 5.991115140914917,
      "epoch": 4.076094145107955,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003425172277093852,
      "loss": 5.0702,
      "mean_token_accuracy": 0.2106425344944,
      "num_tokens": 90875166.0,
      "step": 52390
    },
    {
      "entropy": 5.942475700378418,
      "epoch": 4.076483174479673,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000342490788879151,
      "loss": 4.8872,
      "mean_token_accuracy": 0.22466082870960236,
      "num_tokens": 90883447.0,
      "step": 52395
    },
    {
      "entropy": 5.889201879501343,
      "epoch": 4.076872203851391,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034246434902485156,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21718860864639283,
      "num_tokens": 90892218.0,
      "step": 52400
    },
    {
      "entropy": 5.90369143486023,
      "epoch": 4.077261233223108,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00034243790814688816,
      "loss": 5.0037,
      "mean_token_accuracy": 0.20884954184293747,
      "num_tokens": 90900834.0,
      "step": 52405
    },
    {
      "entropy": 5.9149556159973145,
      "epoch": 4.077650262594826,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034241146624566175,
      "loss": 5.0261,
      "mean_token_accuracy": 0.2133224129676819,
      "num_tokens": 90909902.0,
      "step": 52410
    },
    {
      "entropy": 5.903930139541626,
      "epoch": 4.078039291966544,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003423850233215737,
      "loss": 4.902,
      "mean_token_accuracy": 0.21740110963582993,
      "num_tokens": 90918339.0,
      "step": 52415
    },
    {
      "entropy": 5.861764574050904,
      "epoch": 4.078428321338261,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003423585793750251,
      "loss": 4.7951,
      "mean_token_accuracy": 0.23406800478696824,
      "num_tokens": 90927226.0,
      "step": 52420
    },
    {
      "entropy": 5.8558708190917965,
      "epoch": 4.078817350709978,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034233213440641724,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22018247246742248,
      "num_tokens": 90935222.0,
      "step": 52425
    },
    {
      "entropy": 5.876427173614502,
      "epoch": 4.079206380081696,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00034230568841615145,
      "loss": 4.9234,
      "mean_token_accuracy": 0.21340503990650178,
      "num_tokens": 90943492.0,
      "step": 52430
    },
    {
      "entropy": 5.92194447517395,
      "epoch": 4.079595409453414,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000342279241404629,
      "loss": 5.0139,
      "mean_token_accuracy": 0.2157876193523407,
      "num_tokens": 90952715.0,
      "step": 52435
    },
    {
      "entropy": 6.006902027130127,
      "epoch": 4.079984438825131,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034225279337225106,
      "loss": 5.0706,
      "mean_token_accuracy": 0.2130955696105957,
      "num_tokens": 90962488.0,
      "step": 52440
    },
    {
      "entropy": 5.9346466064453125,
      "epoch": 4.080373468196849,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003422263443194191,
      "loss": 4.9833,
      "mean_token_accuracy": 0.2131215363740921,
      "num_tokens": 90971580.0,
      "step": 52445
    },
    {
      "entropy": 5.877329111099243,
      "epoch": 4.080762497568567,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003421998942465344,
      "loss": 4.9456,
      "mean_token_accuracy": 0.20889061987400054,
      "num_tokens": 90980743.0,
      "step": 52450
    },
    {
      "entropy": 5.866905641555786,
      "epoch": 4.081151526940284,
      "grad_norm": 1.4375,
      "learning_rate": 0.00034217344315399823,
      "loss": 4.8808,
      "mean_token_accuracy": 0.2173201024532318,
      "num_tokens": 90989857.0,
      "step": 52455
    },
    {
      "entropy": 5.9375217914581295,
      "epoch": 4.081540556312001,
      "grad_norm": 1.515625,
      "learning_rate": 0.000342146991042212,
      "loss": 4.941,
      "mean_token_accuracy": 0.21729382872581482,
      "num_tokens": 90998261.0,
      "step": 52460
    },
    {
      "entropy": 5.947642707824707,
      "epoch": 4.081929585683719,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00034212053791157695,
      "loss": 5.0159,
      "mean_token_accuracy": 0.20193351209163665,
      "num_tokens": 91006369.0,
      "step": 52465
    },
    {
      "entropy": 5.889685106277466,
      "epoch": 4.082318615055437,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034209408376249464,
      "loss": 4.924,
      "mean_token_accuracy": 0.21703642457723618,
      "num_tokens": 91015268.0,
      "step": 52470
    },
    {
      "entropy": 5.867584657669068,
      "epoch": 4.082707644427154,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003420676285953664,
      "loss": 4.8891,
      "mean_token_accuracy": 0.21270812600851058,
      "num_tokens": 91023573.0,
      "step": 52475
    },
    {
      "entropy": 5.807168102264404,
      "epoch": 4.083096673798872,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003420411724105937,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21511021703481675,
      "num_tokens": 91032383.0,
      "step": 52480
    },
    {
      "entropy": 5.926326179504395,
      "epoch": 4.08348570317059,
      "grad_norm": 1.375,
      "learning_rate": 0.00034201471520857793,
      "loss": 4.9866,
      "mean_token_accuracy": 0.20646649301052095,
      "num_tokens": 91041129.0,
      "step": 52485
    },
    {
      "entropy": 5.879046535491943,
      "epoch": 4.083874732542307,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003419882569897204,
      "loss": 4.8923,
      "mean_token_accuracy": 0.21938259154558182,
      "num_tokens": 91049422.0,
      "step": 52490
    },
    {
      "entropy": 5.888399267196656,
      "epoch": 4.084263761914024,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034196179775442274,
      "loss": 4.9868,
      "mean_token_accuracy": 0.21392204314470292,
      "num_tokens": 91059539.0,
      "step": 52495
    },
    {
      "entropy": 5.880028963088989,
      "epoch": 4.084652791285742,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003419353375030864,
      "loss": 4.9526,
      "mean_token_accuracy": 0.21438999027013778,
      "num_tokens": 91068068.0,
      "step": 52500
    },
    {
      "entropy": 5.9420241355896,
      "epoch": 4.08504182065746,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003419088762361128,
      "loss": 4.9768,
      "mean_token_accuracy": 0.2147290050983429,
      "num_tokens": 91077552.0,
      "step": 52505
    },
    {
      "entropy": 5.984156084060669,
      "epoch": 4.085430850029177,
      "grad_norm": 1.5,
      "learning_rate": 0.0003418824139539035,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20760720372200012,
      "num_tokens": 91086705.0,
      "step": 52510
    },
    {
      "entropy": 5.993846988677978,
      "epoch": 4.085819879400895,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003418559506568601,
      "loss": 5.0004,
      "mean_token_accuracy": 0.20637951791286469,
      "num_tokens": 91095307.0,
      "step": 52515
    },
    {
      "entropy": 5.88824691772461,
      "epoch": 4.086208908772612,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034182948634538403,
      "loss": 4.8477,
      "mean_token_accuracy": 0.2186955600976944,
      "num_tokens": 91104002.0,
      "step": 52520
    },
    {
      "entropy": 5.963755178451538,
      "epoch": 4.08659793814433,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003418030210198769,
      "loss": 5.0092,
      "mean_token_accuracy": 0.2078005403280258,
      "num_tokens": 91112007.0,
      "step": 52525
    },
    {
      "entropy": 5.872098445892334,
      "epoch": 4.086986967516047,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034177655468074027,
      "loss": 4.9265,
      "mean_token_accuracy": 0.21992426812648774,
      "num_tokens": 91120773.0,
      "step": 52530
    },
    {
      "entropy": 5.859691476821899,
      "epoch": 4.087375996887765,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003417500873283756,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21801672279834747,
      "num_tokens": 91129410.0,
      "step": 52535
    },
    {
      "entropy": 5.878656625747681,
      "epoch": 4.087765026259483,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003417236189631846,
      "loss": 4.9155,
      "mean_token_accuracy": 0.22114154696464539,
      "num_tokens": 91137960.0,
      "step": 52540
    },
    {
      "entropy": 5.916850852966308,
      "epoch": 4.0881540556312,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000341697149585569,
      "loss": 5.0556,
      "mean_token_accuracy": 0.2056829333305359,
      "num_tokens": 91146246.0,
      "step": 52545
    },
    {
      "entropy": 5.904967784881592,
      "epoch": 4.088543085002918,
      "grad_norm": 1.375,
      "learning_rate": 0.0003416706791959302,
      "loss": 4.9325,
      "mean_token_accuracy": 0.22398316711187363,
      "num_tokens": 91154706.0,
      "step": 52550
    },
    {
      "entropy": 5.844547891616822,
      "epoch": 4.088932114374635,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034164420779467003,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22033488005399704,
      "num_tokens": 91164088.0,
      "step": 52555
    },
    {
      "entropy": 5.928648614883423,
      "epoch": 4.089321143746353,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003416177353821901,
      "loss": 4.9568,
      "mean_token_accuracy": 0.21298209875822066,
      "num_tokens": 91172712.0,
      "step": 52560
    },
    {
      "entropy": 5.901903533935547,
      "epoch": 4.08971017311807,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000341591261958892,
      "loss": 4.9148,
      "mean_token_accuracy": 0.22317806482315064,
      "num_tokens": 91180931.0,
      "step": 52565
    },
    {
      "entropy": 5.831737327575683,
      "epoch": 4.090099202489788,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034156478752517754,
      "loss": 4.8726,
      "mean_token_accuracy": 0.21744970679283143,
      "num_tokens": 91190135.0,
      "step": 52570
    },
    {
      "entropy": 5.887293195724487,
      "epoch": 4.090488231861506,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00034153831208144837,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22460052818059922,
      "num_tokens": 91197544.0,
      "step": 52575
    },
    {
      "entropy": 5.93710503578186,
      "epoch": 4.090877261233223,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003415118356281062,
      "loss": 4.9897,
      "mean_token_accuracy": 0.2090143695473671,
      "num_tokens": 91206186.0,
      "step": 52580
    },
    {
      "entropy": 5.963474893569947,
      "epoch": 4.091266290604941,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003414853581655528,
      "loss": 5.0026,
      "mean_token_accuracy": 0.2136582151055336,
      "num_tokens": 91215083.0,
      "step": 52585
    },
    {
      "entropy": 5.964396619796753,
      "epoch": 4.091655319976658,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034145887969419,
      "loss": 4.9618,
      "mean_token_accuracy": 0.21241963654756546,
      "num_tokens": 91223934.0,
      "step": 52590
    },
    {
      "entropy": 5.932173728942871,
      "epoch": 4.092044349348376,
      "grad_norm": 1.375,
      "learning_rate": 0.00034143240021441935,
      "loss": 5.0021,
      "mean_token_accuracy": 0.20295039862394332,
      "num_tokens": 91232466.0,
      "step": 52595
    },
    {
      "entropy": 5.934010362625122,
      "epoch": 4.092433378720093,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003414059197266428,
      "loss": 4.935,
      "mean_token_accuracy": 0.2155100405216217,
      "num_tokens": 91241299.0,
      "step": 52600
    },
    {
      "entropy": 5.9018495082855225,
      "epoch": 4.092822408091811,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034137943823126215,
      "loss": 4.9809,
      "mean_token_accuracy": 0.21469002813100815,
      "num_tokens": 91250483.0,
      "step": 52605
    },
    {
      "entropy": 5.940790128707886,
      "epoch": 4.093211437463529,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003413529557286792,
      "loss": 4.991,
      "mean_token_accuracy": 0.21367815583944322,
      "num_tokens": 91259099.0,
      "step": 52610
    },
    {
      "entropy": 5.871824645996094,
      "epoch": 4.093600466835246,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003413264722192957,
      "loss": 4.8856,
      "mean_token_accuracy": 0.23089529126882552,
      "num_tokens": 91267816.0,
      "step": 52615
    },
    {
      "entropy": 5.882388687133789,
      "epoch": 4.093989496206964,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003412999877035136,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21121062934398652,
      "num_tokens": 91277058.0,
      "step": 52620
    },
    {
      "entropy": 5.915574359893799,
      "epoch": 4.094378525578681,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034127350218173463,
      "loss": 4.9718,
      "mean_token_accuracy": 0.2180978015065193,
      "num_tokens": 91285403.0,
      "step": 52625
    },
    {
      "entropy": 5.917043685913086,
      "epoch": 4.094767554950399,
      "grad_norm": 1.46875,
      "learning_rate": 0.00034124701565436076,
      "loss": 5.0063,
      "mean_token_accuracy": 0.19991597682237625,
      "num_tokens": 91293999.0,
      "step": 52630
    },
    {
      "entropy": 5.9734724998474125,
      "epoch": 4.095156584322116,
      "grad_norm": 1.375,
      "learning_rate": 0.000341220528121794,
      "loss": 5.0143,
      "mean_token_accuracy": 0.20683002918958665,
      "num_tokens": 91303415.0,
      "step": 52635
    },
    {
      "entropy": 5.909023380279541,
      "epoch": 4.095545613693834,
      "grad_norm": 1.46875,
      "learning_rate": 0.000341194039584436,
      "loss": 4.9452,
      "mean_token_accuracy": 0.21730211973190308,
      "num_tokens": 91311791.0,
      "step": 52640
    },
    {
      "entropy": 5.895079803466797,
      "epoch": 4.095934643065552,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003411675500426888,
      "loss": 4.934,
      "mean_token_accuracy": 0.2072712317109108,
      "num_tokens": 91320175.0,
      "step": 52645
    },
    {
      "entropy": 5.906395006179809,
      "epoch": 4.096323672437269,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00034114105949695445,
      "loss": 4.9863,
      "mean_token_accuracy": 0.2061407297849655,
      "num_tokens": 91330210.0,
      "step": 52650
    },
    {
      "entropy": 5.992322540283203,
      "epoch": 4.096712701808986,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003411145679476347,
      "loss": 5.0419,
      "mean_token_accuracy": 0.2041787639260292,
      "num_tokens": 91338331.0,
      "step": 52655
    },
    {
      "entropy": 5.893662643432617,
      "epoch": 4.097101731180704,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003410880753951317,
      "loss": 4.8739,
      "mean_token_accuracy": 0.22068906128406524,
      "num_tokens": 91346492.0,
      "step": 52660
    },
    {
      "entropy": 5.841854906082153,
      "epoch": 4.097490760552422,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003410615818398473,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21056584417819976,
      "num_tokens": 91355734.0,
      "step": 52665
    },
    {
      "entropy": 5.937055826187134,
      "epoch": 4.097879789924139,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003410350872821835,
      "loss": 4.9597,
      "mean_token_accuracy": 0.21435479670763016,
      "num_tokens": 91363982.0,
      "step": 52670
    },
    {
      "entropy": 5.926706171035766,
      "epoch": 4.098268819295857,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034100859172254245,
      "loss": 4.987,
      "mean_token_accuracy": 0.2204262375831604,
      "num_tokens": 91373031.0,
      "step": 52675
    },
    {
      "entropy": 5.9008454322814945,
      "epoch": 4.098657848667575,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00034098209516132607,
      "loss": 5.017,
      "mean_token_accuracy": 0.21275673508644105,
      "num_tokens": 91381676.0,
      "step": 52680
    },
    {
      "entropy": 5.953544092178345,
      "epoch": 4.099046878039292,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003409555975989363,
      "loss": 5.0945,
      "mean_token_accuracy": 0.2017646238207817,
      "num_tokens": 91390488.0,
      "step": 52685
    },
    {
      "entropy": 5.916071081161499,
      "epoch": 4.099435907411009,
      "grad_norm": 1.421875,
      "learning_rate": 0.00034092909903577547,
      "loss": 4.8264,
      "mean_token_accuracy": 0.22398722916841507,
      "num_tokens": 91399069.0,
      "step": 52690
    },
    {
      "entropy": 5.892436122894287,
      "epoch": 4.099824936782727,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003409025994722454,
      "loss": 4.9351,
      "mean_token_accuracy": 0.21681943535804749,
      "num_tokens": 91407921.0,
      "step": 52695
    },
    {
      "entropy": 5.8305457592010494,
      "epoch": 4.100213966154445,
      "grad_norm": 1.375,
      "learning_rate": 0.0003408760989087482,
      "loss": 4.899,
      "mean_token_accuracy": 0.2193138048052788,
      "num_tokens": 91416077.0,
      "step": 52700
    },
    {
      "entropy": 5.944351577758789,
      "epoch": 4.100602995526162,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034084959734568616,
      "loss": 5.0786,
      "mean_token_accuracy": 0.20689990520477294,
      "num_tokens": 91424941.0,
      "step": 52705
    },
    {
      "entropy": 6.088378095626831,
      "epoch": 4.10099202489788,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034082309478346127,
      "loss": 5.1043,
      "mean_token_accuracy": 0.20074537694454192,
      "num_tokens": 91433052.0,
      "step": 52710
    },
    {
      "entropy": 5.886245441436768,
      "epoch": 4.101381054269598,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034079659122247574,
      "loss": 4.8886,
      "mean_token_accuracy": 0.21408981680870057,
      "num_tokens": 91441109.0,
      "step": 52715
    },
    {
      "entropy": 5.846343755722046,
      "epoch": 4.101770083641315,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003407700866631317,
      "loss": 4.9307,
      "mean_token_accuracy": 0.2107575699687004,
      "num_tokens": 91449256.0,
      "step": 52720
    },
    {
      "entropy": 5.848797464370728,
      "epoch": 4.102159113013032,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003407435811058312,
      "loss": 4.9022,
      "mean_token_accuracy": 0.21143162548542022,
      "num_tokens": 91458398.0,
      "step": 52725
    },
    {
      "entropy": 5.869709730148315,
      "epoch": 4.10254814238475,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003407170745509764,
      "loss": 4.8783,
      "mean_token_accuracy": 0.21406591832637786,
      "num_tokens": 91467291.0,
      "step": 52730
    },
    {
      "entropy": 5.96812105178833,
      "epoch": 4.1029371717564675,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00034069056699896974,
      "loss": 5.0348,
      "mean_token_accuracy": 0.2082229033112526,
      "num_tokens": 91475616.0,
      "step": 52735
    },
    {
      "entropy": 5.967672681808471,
      "epoch": 4.103326201128185,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003406640584502132,
      "loss": 5.0348,
      "mean_token_accuracy": 0.2154568389058113,
      "num_tokens": 91484289.0,
      "step": 52740
    },
    {
      "entropy": 5.850242614746094,
      "epoch": 4.103715230499903,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003406375489051091,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22030099034309386,
      "num_tokens": 91492483.0,
      "step": 52745
    },
    {
      "entropy": 5.848580265045166,
      "epoch": 4.104104259871621,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003406110383640597,
      "loss": 4.9434,
      "mean_token_accuracy": 0.21315933912992477,
      "num_tokens": 91502002.0,
      "step": 52750
    },
    {
      "entropy": 5.968935966491699,
      "epoch": 4.1044932892433375,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034058452682746734,
      "loss": 5.0422,
      "mean_token_accuracy": 0.21156175881624223,
      "num_tokens": 91510392.0,
      "step": 52755
    },
    {
      "entropy": 5.937663316726685,
      "epoch": 4.104882318615055,
      "grad_norm": 1.421875,
      "learning_rate": 0.000340558014295734,
      "loss": 4.9296,
      "mean_token_accuracy": 0.2151510939002037,
      "num_tokens": 91518650.0,
      "step": 52760
    },
    {
      "entropy": 5.938623332977295,
      "epoch": 4.105271347986773,
      "grad_norm": 1.421875,
      "learning_rate": 0.00034053150076926214,
      "loss": 5.0786,
      "mean_token_accuracy": 0.20556239038705826,
      "num_tokens": 91528062.0,
      "step": 52765
    },
    {
      "entropy": 5.85027174949646,
      "epoch": 4.1056603773584905,
      "grad_norm": 1.5,
      "learning_rate": 0.0003405049862484541,
      "loss": 4.9506,
      "mean_token_accuracy": 0.2107527047395706,
      "num_tokens": 91536615.0,
      "step": 52770
    },
    {
      "entropy": 5.960754632949829,
      "epoch": 4.106049406730208,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003404784707337122,
      "loss": 5.0667,
      "mean_token_accuracy": 0.20279465913772582,
      "num_tokens": 91544912.0,
      "step": 52775
    },
    {
      "entropy": 5.902083206176758,
      "epoch": 4.106438436101926,
      "grad_norm": 1.4375,
      "learning_rate": 0.00034045195422543865,
      "loss": 4.9725,
      "mean_token_accuracy": 0.211878901720047,
      "num_tokens": 91553568.0,
      "step": 52780
    },
    {
      "entropy": 5.962917137145996,
      "epoch": 4.106827465473644,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034042543672403594,
      "loss": 5.0813,
      "mean_token_accuracy": 0.20828411132097244,
      "num_tokens": 91562473.0,
      "step": 52785
    },
    {
      "entropy": 5.859739160537719,
      "epoch": 4.10721649484536,
      "grad_norm": 1.375,
      "learning_rate": 0.00034039891822990634,
      "loss": 4.7885,
      "mean_token_accuracy": 0.22724731713533403,
      "num_tokens": 91570750.0,
      "step": 52790
    },
    {
      "entropy": 5.920616674423218,
      "epoch": 4.107605524217078,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003403723987434523,
      "loss": 4.9607,
      "mean_token_accuracy": 0.21304586082696914,
      "num_tokens": 91580463.0,
      "step": 52795
    },
    {
      "entropy": 5.864268255233765,
      "epoch": 4.107994553588796,
      "grad_norm": 1.4375,
      "learning_rate": 0.00034034587826507604,
      "loss": 4.8684,
      "mean_token_accuracy": 0.22026564478874205,
      "num_tokens": 91588690.0,
      "step": 52800
    },
    {
      "entropy": 5.807320785522461,
      "epoch": 4.1083835829605135,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034031935679518016,
      "loss": 4.8816,
      "mean_token_accuracy": 0.2250775709748268,
      "num_tokens": 91597439.0,
      "step": 52805
    },
    {
      "entropy": 5.859689998626709,
      "epoch": 4.108772612332231,
      "grad_norm": 1.3125,
      "learning_rate": 0.000340292834334167,
      "loss": 4.897,
      "mean_token_accuracy": 0.2160307750105858,
      "num_tokens": 91606711.0,
      "step": 52810
    },
    {
      "entropy": 5.932690000534057,
      "epoch": 4.109161641703949,
      "grad_norm": 1.375,
      "learning_rate": 0.000340266310882439,
      "loss": 4.896,
      "mean_token_accuracy": 0.218765552341938,
      "num_tokens": 91615683.0,
      "step": 52815
    },
    {
      "entropy": 5.871548652648926,
      "epoch": 4.109550671075667,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034023978644039864,
      "loss": 4.949,
      "mean_token_accuracy": 0.21354755908250808,
      "num_tokens": 91623816.0,
      "step": 52820
    },
    {
      "entropy": 5.8265052318573,
      "epoch": 4.109939700447383,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003402132610084483,
      "loss": 4.8498,
      "mean_token_accuracy": 0.21989983320236206,
      "num_tokens": 91633600.0,
      "step": 52825
    },
    {
      "entropy": 5.8325258731842045,
      "epoch": 4.110328729819101,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034018673458699055,
      "loss": 4.9145,
      "mean_token_accuracy": 0.2131930634379387,
      "num_tokens": 91642072.0,
      "step": 52830
    },
    {
      "entropy": 5.88376932144165,
      "epoch": 4.110717759190819,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003401602071764279,
      "loss": 4.9671,
      "mean_token_accuracy": 0.2141076758503914,
      "num_tokens": 91651515.0,
      "step": 52835
    },
    {
      "entropy": 5.944074440002441,
      "epoch": 4.1111067885625365,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034013367877716284,
      "loss": 5.0204,
      "mean_token_accuracy": 0.209410397708416,
      "num_tokens": 91660492.0,
      "step": 52840
    },
    {
      "entropy": 5.972733640670777,
      "epoch": 4.111495817934254,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003401071493895977,
      "loss": 5.0555,
      "mean_token_accuracy": 0.2058595448732376,
      "num_tokens": 91669091.0,
      "step": 52845
    },
    {
      "entropy": 5.894702196121216,
      "epoch": 4.111884847305972,
      "grad_norm": 1.375,
      "learning_rate": 0.0003400806190141354,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21729811280965805,
      "num_tokens": 91676957.0,
      "step": 52850
    },
    {
      "entropy": 5.848164463043213,
      "epoch": 4.112273876677689,
      "grad_norm": 1.578125,
      "learning_rate": 0.00034005408765117815,
      "loss": 4.9675,
      "mean_token_accuracy": 0.21747759282588958,
      "num_tokens": 91686230.0,
      "step": 52855
    },
    {
      "entropy": 5.868437242507935,
      "epoch": 4.112662906049406,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034002755530112877,
      "loss": 4.9185,
      "mean_token_accuracy": 0.22282164692878723,
      "num_tokens": 91695917.0,
      "step": 52860
    },
    {
      "entropy": 5.931685304641723,
      "epoch": 4.113051935421124,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003400010219643897,
      "loss": 4.9225,
      "mean_token_accuracy": 0.21332666873931885,
      "num_tokens": 91704619.0,
      "step": 52865
    },
    {
      "entropy": 5.925198221206665,
      "epoch": 4.113440964792842,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0003399744876413636,
      "loss": 5.0034,
      "mean_token_accuracy": 0.20861937701702118,
      "num_tokens": 91712524.0,
      "step": 52870
    },
    {
      "entropy": 5.883453750610352,
      "epoch": 4.1138299941645595,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000339947952332453,
      "loss": 5.0196,
      "mean_token_accuracy": 0.21128413677215577,
      "num_tokens": 91721455.0,
      "step": 52875
    },
    {
      "entropy": 5.863312101364135,
      "epoch": 4.114219023536277,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033992141603806064,
      "loss": 4.8223,
      "mean_token_accuracy": 0.2245756134390831,
      "num_tokens": 91729567.0,
      "step": 52880
    },
    {
      "entropy": 5.877848529815674,
      "epoch": 4.114608052907995,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033989487875858915,
      "loss": 4.9083,
      "mean_token_accuracy": 0.2183839499950409,
      "num_tokens": 91738373.0,
      "step": 52885
    },
    {
      "entropy": 5.850502872467041,
      "epoch": 4.114997082279712,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033986834049444113,
      "loss": 4.8933,
      "mean_token_accuracy": 0.22003017663955687,
      "num_tokens": 91747249.0,
      "step": 52890
    },
    {
      "entropy": 5.861805438995361,
      "epoch": 4.115386111651429,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033984180124601936,
      "loss": 4.879,
      "mean_token_accuracy": 0.21951905339956285,
      "num_tokens": 91754986.0,
      "step": 52895
    },
    {
      "entropy": 5.878744792938233,
      "epoch": 4.115775141023147,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033981526101372635,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2142985239624977,
      "num_tokens": 91764127.0,
      "step": 52900
    },
    {
      "entropy": 5.890066146850586,
      "epoch": 4.116164170394865,
      "grad_norm": 1.40625,
      "learning_rate": 0.000339788719797965,
      "loss": 4.9044,
      "mean_token_accuracy": 0.21786743998527527,
      "num_tokens": 91772564.0,
      "step": 52905
    },
    {
      "entropy": 5.871653366088867,
      "epoch": 4.1165531997665825,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003397621775991379,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21094811111688613,
      "num_tokens": 91780810.0,
      "step": 52910
    },
    {
      "entropy": 5.902502250671387,
      "epoch": 4.1169422291383,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003397356344176479,
      "loss": 4.9243,
      "mean_token_accuracy": 0.2120744213461876,
      "num_tokens": 91789065.0,
      "step": 52915
    },
    {
      "entropy": 5.872078275680542,
      "epoch": 4.117331258510018,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003397090902538976,
      "loss": 4.849,
      "mean_token_accuracy": 0.22367219775915145,
      "num_tokens": 91796891.0,
      "step": 52920
    },
    {
      "entropy": 5.896863698959351,
      "epoch": 4.117720287881735,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033968254510828995,
      "loss": 4.9546,
      "mean_token_accuracy": 0.215956012904644,
      "num_tokens": 91805447.0,
      "step": 52925
    },
    {
      "entropy": 5.82548713684082,
      "epoch": 4.118109317253452,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003396559989812275,
      "loss": 4.8796,
      "mean_token_accuracy": 0.21698842644691468,
      "num_tokens": 91814232.0,
      "step": 52930
    },
    {
      "entropy": 5.878021860122681,
      "epoch": 4.11849834662517,
      "grad_norm": 1.375,
      "learning_rate": 0.0003396294518731133,
      "loss": 4.9899,
      "mean_token_accuracy": 0.21511362344026566,
      "num_tokens": 91823557.0,
      "step": 52935
    },
    {
      "entropy": 5.959102582931519,
      "epoch": 4.118887375996888,
      "grad_norm": 1.5,
      "learning_rate": 0.00033960290378435,
      "loss": 4.9341,
      "mean_token_accuracy": 0.22572496086359023,
      "num_tokens": 91831579.0,
      "step": 52940
    },
    {
      "entropy": 5.945687341690063,
      "epoch": 4.1192764053686055,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00033957635471534045,
      "loss": 5.0048,
      "mean_token_accuracy": 0.21300116330385208,
      "num_tokens": 91839456.0,
      "step": 52945
    },
    {
      "entropy": 5.905714321136474,
      "epoch": 4.119665434740323,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003395498046664875,
      "loss": 5.0104,
      "mean_token_accuracy": 0.21242806911468506,
      "num_tokens": 91848051.0,
      "step": 52950
    },
    {
      "entropy": 5.875723028182984,
      "epoch": 4.120054464112041,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000339523253638194,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21487711668014525,
      "num_tokens": 91856153.0,
      "step": 52955
    },
    {
      "entropy": 5.8718345165252686,
      "epoch": 4.120443493483758,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003394967016308629,
      "loss": 4.9644,
      "mean_token_accuracy": 0.2136249616742134,
      "num_tokens": 91865083.0,
      "step": 52960
    },
    {
      "entropy": 5.93446569442749,
      "epoch": 4.120832522855475,
      "grad_norm": 1.5,
      "learning_rate": 0.0003394701486448968,
      "loss": 5.0493,
      "mean_token_accuracy": 0.20271414518356323,
      "num_tokens": 91872655.0,
      "step": 52965
    },
    {
      "entropy": 5.960166358947754,
      "epoch": 4.121221552227193,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033944359468069903,
      "loss": 5.0199,
      "mean_token_accuracy": 0.20944157093763352,
      "num_tokens": 91881735.0,
      "step": 52970
    },
    {
      "entropy": 5.849606847763061,
      "epoch": 4.121610581598911,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033941703973867216,
      "loss": 4.8875,
      "mean_token_accuracy": 0.22056015878915786,
      "num_tokens": 91890730.0,
      "step": 52975
    },
    {
      "entropy": 5.903974914550782,
      "epoch": 4.1219996109706285,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033939048381921935,
      "loss": 4.954,
      "mean_token_accuracy": 0.21133678704500197,
      "num_tokens": 91899393.0,
      "step": 52980
    },
    {
      "entropy": 5.938225984573364,
      "epoch": 4.122388640342346,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003393639269227434,
      "loss": 5.0215,
      "mean_token_accuracy": 0.21261327862739562,
      "num_tokens": 91908946.0,
      "step": 52985
    },
    {
      "entropy": 5.929977226257324,
      "epoch": 4.122777669714063,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003393373690496473,
      "loss": 4.8788,
      "mean_token_accuracy": 0.217441688477993,
      "num_tokens": 91917592.0,
      "step": 52990
    },
    {
      "entropy": 5.919007968902588,
      "epoch": 4.123166699085781,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000339310810200334,
      "loss": 4.8964,
      "mean_token_accuracy": 0.21340139359235763,
      "num_tokens": 91925576.0,
      "step": 52995
    },
    {
      "entropy": 5.849965286254883,
      "epoch": 4.123555728457498,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00033928425037520643,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22110699564218522,
      "num_tokens": 91933889.0,
      "step": 53000
    },
    {
      "entropy": 5.87031831741333,
      "epoch": 4.123944757829216,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003392576895746678,
      "loss": 4.9158,
      "mean_token_accuracy": 0.21415314972400665,
      "num_tokens": 91942733.0,
      "step": 53005
    },
    {
      "entropy": 5.945229816436767,
      "epoch": 4.124333787200934,
      "grad_norm": 1.59375,
      "learning_rate": 0.0003392311277991209,
      "loss": 4.9836,
      "mean_token_accuracy": 0.2145493820309639,
      "num_tokens": 91950780.0,
      "step": 53010
    },
    {
      "entropy": 5.929593753814697,
      "epoch": 4.1247228165726515,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033920456504896895,
      "loss": 4.9884,
      "mean_token_accuracy": 0.21254315376281738,
      "num_tokens": 91958908.0,
      "step": 53015
    },
    {
      "entropy": 5.833272933959961,
      "epoch": 4.125111845944369,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033917800132461493,
      "loss": 4.8658,
      "mean_token_accuracy": 0.2193354532122612,
      "num_tokens": 91967288.0,
      "step": 53020
    },
    {
      "entropy": 5.8546998500823975,
      "epoch": 4.125500875316086,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003391514366264619,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2152662456035614,
      "num_tokens": 91975806.0,
      "step": 53025
    },
    {
      "entropy": 5.947770500183106,
      "epoch": 4.125889904687804,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003391248709549129,
      "loss": 4.9223,
      "mean_token_accuracy": 0.21123569905757905,
      "num_tokens": 91984992.0,
      "step": 53030
    },
    {
      "entropy": 5.953722429275513,
      "epoch": 4.126278934059521,
      "grad_norm": 1.375,
      "learning_rate": 0.0003390983043103711,
      "loss": 4.9604,
      "mean_token_accuracy": 0.21307571679353715,
      "num_tokens": 91993899.0,
      "step": 53035
    },
    {
      "entropy": 5.865326404571533,
      "epoch": 4.126667963431239,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003390717366932395,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21270429342985153,
      "num_tokens": 92002628.0,
      "step": 53040
    },
    {
      "entropy": 5.826969194412231,
      "epoch": 4.127056992802957,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003390451681039212,
      "loss": 4.8806,
      "mean_token_accuracy": 0.2256618097424507,
      "num_tokens": 92011406.0,
      "step": 53045
    },
    {
      "entropy": 5.924251317977905,
      "epoch": 4.1274460221746745,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033901859854281954,
      "loss": 4.9808,
      "mean_token_accuracy": 0.21165835857391357,
      "num_tokens": 92019572.0,
      "step": 53050
    },
    {
      "entropy": 5.876049518585205,
      "epoch": 4.127835051546391,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003389920280103375,
      "loss": 4.9693,
      "mean_token_accuracy": 0.22090078443288802,
      "num_tokens": 92028273.0,
      "step": 53055
    },
    {
      "entropy": 5.96041579246521,
      "epoch": 4.128224080918109,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033896545650687824,
      "loss": 4.9875,
      "mean_token_accuracy": 0.21151614487171172,
      "num_tokens": 92036460.0,
      "step": 53060
    },
    {
      "entropy": 5.973018455505371,
      "epoch": 4.128613110289827,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003389388840328451,
      "loss": 5.0431,
      "mean_token_accuracy": 0.2069725662469864,
      "num_tokens": 92045109.0,
      "step": 53065
    },
    {
      "entropy": 5.915903806686401,
      "epoch": 4.129002139661544,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003389123105886411,
      "loss": 5.0339,
      "mean_token_accuracy": 0.20924107283353804,
      "num_tokens": 92053369.0,
      "step": 53070
    },
    {
      "entropy": 5.894375038146973,
      "epoch": 4.129391169033262,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003388857361746695,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22037282586097717,
      "num_tokens": 92062040.0,
      "step": 53075
    },
    {
      "entropy": 5.8793778896331785,
      "epoch": 4.12978019840498,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033885916079133354,
      "loss": 4.892,
      "mean_token_accuracy": 0.22042303383350373,
      "num_tokens": 92070832.0,
      "step": 53080
    },
    {
      "entropy": 5.861036920547486,
      "epoch": 4.1301692277766975,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033883258443903646,
      "loss": 4.9496,
      "mean_token_accuracy": 0.20970155745744706,
      "num_tokens": 92079862.0,
      "step": 53085
    },
    {
      "entropy": 6.02113618850708,
      "epoch": 4.130558257148414,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033880600711818157,
      "loss": 5.0514,
      "mean_token_accuracy": 0.20777950882911683,
      "num_tokens": 92089403.0,
      "step": 53090
    },
    {
      "entropy": 5.981896018981933,
      "epoch": 4.130947286520132,
      "grad_norm": 1.53125,
      "learning_rate": 0.00033877942882917203,
      "loss": 4.9727,
      "mean_token_accuracy": 0.2144062638282776,
      "num_tokens": 92097902.0,
      "step": 53095
    },
    {
      "entropy": 5.884845924377442,
      "epoch": 4.13133631589185,
      "grad_norm": 1.375,
      "learning_rate": 0.00033875284957241117,
      "loss": 4.9807,
      "mean_token_accuracy": 0.2114662081003189,
      "num_tokens": 92107149.0,
      "step": 53100
    },
    {
      "entropy": 5.8629707336425785,
      "epoch": 4.131725345263567,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003387262693483023,
      "loss": 4.9827,
      "mean_token_accuracy": 0.2068902686238289,
      "num_tokens": 92115650.0,
      "step": 53105
    },
    {
      "entropy": 5.951466989517212,
      "epoch": 4.132114374635285,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033869968815724866,
      "loss": 4.9822,
      "mean_token_accuracy": 0.2173279970884323,
      "num_tokens": 92124494.0,
      "step": 53110
    },
    {
      "entropy": 5.8978346824646,
      "epoch": 4.132503404007003,
      "grad_norm": 1.375,
      "learning_rate": 0.0003386731059996537,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21720950901508332,
      "num_tokens": 92133223.0,
      "step": 53115
    },
    {
      "entropy": 5.854339122772217,
      "epoch": 4.1328924333787205,
      "grad_norm": 1.453125,
      "learning_rate": 0.00033864652287592074,
      "loss": 4.9176,
      "mean_token_accuracy": 0.22681528180837632,
      "num_tokens": 92141546.0,
      "step": 53120
    },
    {
      "entropy": 5.893673658370972,
      "epoch": 4.133281462750437,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033861993878645305,
      "loss": 5.0173,
      "mean_token_accuracy": 0.21151699125766754,
      "num_tokens": 92150630.0,
      "step": 53125
    },
    {
      "entropy": 5.9865946769714355,
      "epoch": 4.133670492122155,
      "grad_norm": 1.46875,
      "learning_rate": 0.000338593353731654,
      "loss": 5.0233,
      "mean_token_accuracy": 0.21498106569051742,
      "num_tokens": 92160278.0,
      "step": 53130
    },
    {
      "entropy": 5.955991554260254,
      "epoch": 4.134059521493873,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003385667677119271,
      "loss": 4.9275,
      "mean_token_accuracy": 0.21284471899271012,
      "num_tokens": 92168078.0,
      "step": 53135
    },
    {
      "entropy": 5.8389918327331545,
      "epoch": 4.13444855086559,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033854018072767574,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2197596475481987,
      "num_tokens": 92176076.0,
      "step": 53140
    },
    {
      "entropy": 5.758515214920044,
      "epoch": 4.134837580237308,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033851359277930316,
      "loss": 4.8703,
      "mean_token_accuracy": 0.21958047747612,
      "num_tokens": 92183978.0,
      "step": 53145
    },
    {
      "entropy": 5.836000633239746,
      "epoch": 4.135226609609026,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003384870038672129,
      "loss": 4.9027,
      "mean_token_accuracy": 0.21999291628599166,
      "num_tokens": 92192788.0,
      "step": 53150
    },
    {
      "entropy": 5.832181978225708,
      "epoch": 4.1356156389807435,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003384604139918085,
      "loss": 4.8006,
      "mean_token_accuracy": 0.22784579992294313,
      "num_tokens": 92201210.0,
      "step": 53155
    },
    {
      "entropy": 5.915728330612183,
      "epoch": 4.13600466835246,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033843382315349325,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21042668372392653,
      "num_tokens": 92210107.0,
      "step": 53160
    },
    {
      "entropy": 5.948424530029297,
      "epoch": 4.136393697724178,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033840723135267073,
      "loss": 5.0167,
      "mean_token_accuracy": 0.2129012778401375,
      "num_tokens": 92219206.0,
      "step": 53165
    },
    {
      "entropy": 5.87229323387146,
      "epoch": 4.136782727095896,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033838063858974434,
      "loss": 4.9266,
      "mean_token_accuracy": 0.2198166862130165,
      "num_tokens": 92227974.0,
      "step": 53170
    },
    {
      "entropy": 5.8090348720550535,
      "epoch": 4.137171756467613,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003383540448651177,
      "loss": 4.8373,
      "mean_token_accuracy": 0.2178857818245888,
      "num_tokens": 92235743.0,
      "step": 53175
    },
    {
      "entropy": 5.858920001983643,
      "epoch": 4.137560785839331,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033832745017919424,
      "loss": 4.9549,
      "mean_token_accuracy": 0.2116394892334938,
      "num_tokens": 92243756.0,
      "step": 53180
    },
    {
      "entropy": 5.868471002578735,
      "epoch": 4.137949815211049,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033830085453237757,
      "loss": 4.9802,
      "mean_token_accuracy": 0.20792966932058335,
      "num_tokens": 92252770.0,
      "step": 53185
    },
    {
      "entropy": 5.916534519195556,
      "epoch": 4.138338844582766,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003382742579250712,
      "loss": 5.005,
      "mean_token_accuracy": 0.20948924124240875,
      "num_tokens": 92261431.0,
      "step": 53190
    },
    {
      "entropy": 5.985233736038208,
      "epoch": 4.138727873954483,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003382476603576785,
      "loss": 4.9872,
      "mean_token_accuracy": 0.2044818952679634,
      "num_tokens": 92269669.0,
      "step": 53195
    },
    {
      "entropy": 5.893612194061279,
      "epoch": 4.139116903326201,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003382210618306034,
      "loss": 4.9565,
      "mean_token_accuracy": 0.2104913666844368,
      "num_tokens": 92278723.0,
      "step": 53200
    },
    {
      "entropy": 5.9354432106018065,
      "epoch": 4.139505932697919,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003381944623442492,
      "loss": 5.0639,
      "mean_token_accuracy": 0.2117059901356697,
      "num_tokens": 92287905.0,
      "step": 53205
    },
    {
      "entropy": 5.8875645160675045,
      "epoch": 4.139894962069636,
      "grad_norm": 1.5,
      "learning_rate": 0.0003381678618990197,
      "loss": 4.9358,
      "mean_token_accuracy": 0.21197791695594786,
      "num_tokens": 92296882.0,
      "step": 53210
    },
    {
      "entropy": 5.9310266971588135,
      "epoch": 4.140283991441354,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003381412604953185,
      "loss": 4.9826,
      "mean_token_accuracy": 0.21040562242269517,
      "num_tokens": 92306425.0,
      "step": 53215
    },
    {
      "entropy": 5.934935855865478,
      "epoch": 4.140673020813072,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003381146581335491,
      "loss": 4.9951,
      "mean_token_accuracy": 0.21959634870290756,
      "num_tokens": 92315607.0,
      "step": 53220
    },
    {
      "entropy": 5.782837390899658,
      "epoch": 4.141062050184789,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003380880548141152,
      "loss": 4.8022,
      "mean_token_accuracy": 0.22800897657871247,
      "num_tokens": 92324673.0,
      "step": 53225
    },
    {
      "entropy": 5.863712882995605,
      "epoch": 4.141451079556506,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033806145053742043,
      "loss": 4.9919,
      "mean_token_accuracy": 0.21263565868139267,
      "num_tokens": 92333073.0,
      "step": 53230
    },
    {
      "entropy": 5.964238452911377,
      "epoch": 4.141840108928224,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003380348453038686,
      "loss": 4.9697,
      "mean_token_accuracy": 0.2206350177526474,
      "num_tokens": 92341559.0,
      "step": 53235
    },
    {
      "entropy": 5.972965478897095,
      "epoch": 4.142229138299942,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003380082391138633,
      "loss": 5.0301,
      "mean_token_accuracy": 0.21591034084558486,
      "num_tokens": 92350735.0,
      "step": 53240
    },
    {
      "entropy": 5.923988008499146,
      "epoch": 4.142618167671659,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003379816319678083,
      "loss": 5.0071,
      "mean_token_accuracy": 0.2115897372364998,
      "num_tokens": 92359554.0,
      "step": 53245
    },
    {
      "entropy": 5.916676235198975,
      "epoch": 4.143007197043377,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003379550238661072,
      "loss": 5.052,
      "mean_token_accuracy": 0.2146249607205391,
      "num_tokens": 92368098.0,
      "step": 53250
    },
    {
      "entropy": 5.843868160247803,
      "epoch": 4.143396226415095,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033792841480916394,
      "loss": 4.9128,
      "mean_token_accuracy": 0.2171401023864746,
      "num_tokens": 92376373.0,
      "step": 53255
    },
    {
      "entropy": 5.885285377502441,
      "epoch": 4.143785255786812,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033790180479738205,
      "loss": 4.9391,
      "mean_token_accuracy": 0.21431785374879836,
      "num_tokens": 92384621.0,
      "step": 53260
    },
    {
      "entropy": 5.956968927383423,
      "epoch": 4.144174285158529,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003378751938311654,
      "loss": 4.9845,
      "mean_token_accuracy": 0.2113196611404419,
      "num_tokens": 92393170.0,
      "step": 53265
    },
    {
      "entropy": 5.946611928939819,
      "epoch": 4.144563314530247,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003378485819109177,
      "loss": 4.9807,
      "mean_token_accuracy": 0.2142006739974022,
      "num_tokens": 92401874.0,
      "step": 53270
    },
    {
      "entropy": 5.894551563262939,
      "epoch": 4.144952343901965,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033782196903704286,
      "loss": 4.9885,
      "mean_token_accuracy": 0.2128918468952179,
      "num_tokens": 92410735.0,
      "step": 53275
    },
    {
      "entropy": 5.8234357833862305,
      "epoch": 4.145341373273682,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003377953552099447,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22030453830957414,
      "num_tokens": 92419046.0,
      "step": 53280
    },
    {
      "entropy": 5.862298965454102,
      "epoch": 4.1457304026454,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003377687404300269,
      "loss": 4.903,
      "mean_token_accuracy": 0.22330783009529115,
      "num_tokens": 92428529.0,
      "step": 53285
    },
    {
      "entropy": 5.923554944992065,
      "epoch": 4.146119432017118,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003377421246976934,
      "loss": 4.9971,
      "mean_token_accuracy": 0.211757892370224,
      "num_tokens": 92437717.0,
      "step": 53290
    },
    {
      "entropy": 5.933090114593506,
      "epoch": 4.1465084613888346,
      "grad_norm": 1.5,
      "learning_rate": 0.00033771550801334806,
      "loss": 4.9079,
      "mean_token_accuracy": 0.2216992348432541,
      "num_tokens": 92446152.0,
      "step": 53295
    },
    {
      "entropy": 6.002976369857788,
      "epoch": 4.146897490760552,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033768889037739464,
      "loss": 5.0143,
      "mean_token_accuracy": 0.20954060703516006,
      "num_tokens": 92454671.0,
      "step": 53300
    },
    {
      "entropy": 5.904496097564698,
      "epoch": 4.14728652013227,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00033766227179023704,
      "loss": 5.02,
      "mean_token_accuracy": 0.2119699791073799,
      "num_tokens": 92463582.0,
      "step": 53305
    },
    {
      "entropy": 5.877216053009033,
      "epoch": 4.147675549503988,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033763565225227934,
      "loss": 4.9547,
      "mean_token_accuracy": 0.21369892209768296,
      "num_tokens": 92472431.0,
      "step": 53310
    },
    {
      "entropy": 5.915773344039917,
      "epoch": 4.148064578875705,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003376090317639252,
      "loss": 4.9291,
      "mean_token_accuracy": 0.21621883362531663,
      "num_tokens": 92480940.0,
      "step": 53315
    },
    {
      "entropy": 5.905050563812256,
      "epoch": 4.148453608247423,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00033758241032557875,
      "loss": 4.8885,
      "mean_token_accuracy": 0.22139197140932082,
      "num_tokens": 92488563.0,
      "step": 53320
    },
    {
      "entropy": 5.865471029281617,
      "epoch": 4.14884263761914,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033755578793764375,
      "loss": 4.9614,
      "mean_token_accuracy": 0.2141645148396492,
      "num_tokens": 92497733.0,
      "step": 53325
    },
    {
      "entropy": 5.885203218460083,
      "epoch": 4.1492316669908575,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003375291646005243,
      "loss": 4.889,
      "mean_token_accuracy": 0.2166086331009865,
      "num_tokens": 92505970.0,
      "step": 53330
    },
    {
      "entropy": 5.897160530090332,
      "epoch": 4.149620696362575,
      "grad_norm": 1.515625,
      "learning_rate": 0.00033750254031462425,
      "loss": 4.8164,
      "mean_token_accuracy": 0.21901627779006957,
      "num_tokens": 92514189.0,
      "step": 53335
    },
    {
      "entropy": 5.907404375076294,
      "epoch": 4.150009725734293,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033747591508034765,
      "loss": 4.991,
      "mean_token_accuracy": 0.21511359810829161,
      "num_tokens": 92523158.0,
      "step": 53340
    },
    {
      "entropy": 5.973261451721191,
      "epoch": 4.150398755106011,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003374492888980985,
      "loss": 4.9445,
      "mean_token_accuracy": 0.2234194129705429,
      "num_tokens": 92531762.0,
      "step": 53345
    },
    {
      "entropy": 5.932356548309326,
      "epoch": 4.150787784477728,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033742266176828073,
      "loss": 4.9612,
      "mean_token_accuracy": 0.21476418972015382,
      "num_tokens": 92540651.0,
      "step": 53350
    },
    {
      "entropy": 5.901476860046387,
      "epoch": 4.151176813849446,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033739603369129846,
      "loss": 5.0179,
      "mean_token_accuracy": 0.20615463554859162,
      "num_tokens": 92549515.0,
      "step": 53355
    },
    {
      "entropy": 5.896136379241943,
      "epoch": 4.151565843221163,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033736940466755566,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21683864295482635,
      "num_tokens": 92558574.0,
      "step": 53360
    },
    {
      "entropy": 5.947173452377319,
      "epoch": 4.1519548725928805,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033734277469745644,
      "loss": 4.939,
      "mean_token_accuracy": 0.2134280502796173,
      "num_tokens": 92566980.0,
      "step": 53365
    },
    {
      "entropy": 5.928539419174195,
      "epoch": 4.152343901964598,
      "grad_norm": 1.375,
      "learning_rate": 0.0003373161437814049,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20857737213373184,
      "num_tokens": 92575293.0,
      "step": 53370
    },
    {
      "entropy": 5.883195400238037,
      "epoch": 4.152732931336316,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033728951191980505,
      "loss": 4.9514,
      "mean_token_accuracy": 0.22458741068840027,
      "num_tokens": 92584347.0,
      "step": 53375
    },
    {
      "entropy": 5.8464783191680905,
      "epoch": 4.153121960708034,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033726287911306095,
      "loss": 4.8755,
      "mean_token_accuracy": 0.22081366926431656,
      "num_tokens": 92592614.0,
      "step": 53380
    },
    {
      "entropy": 5.954621696472168,
      "epoch": 4.153510990079751,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003372362453615768,
      "loss": 5.0296,
      "mean_token_accuracy": 0.20822788923978805,
      "num_tokens": 92601238.0,
      "step": 53385
    },
    {
      "entropy": 5.870832633972168,
      "epoch": 4.153900019451468,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003372096106657566,
      "loss": 4.9045,
      "mean_token_accuracy": 0.2165669247508049,
      "num_tokens": 92609764.0,
      "step": 53390
    },
    {
      "entropy": 5.901909828186035,
      "epoch": 4.154289048823186,
      "grad_norm": 1.453125,
      "learning_rate": 0.00033718297502600465,
      "loss": 4.9268,
      "mean_token_accuracy": 0.21533781588077544,
      "num_tokens": 92618266.0,
      "step": 53395
    },
    {
      "entropy": 5.918218517303467,
      "epoch": 4.1546780781949035,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033715633844272504,
      "loss": 4.9225,
      "mean_token_accuracy": 0.21469749510288239,
      "num_tokens": 92627213.0,
      "step": 53400
    },
    {
      "entropy": 5.8901652812957765,
      "epoch": 4.155067107566621,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033712970091632187,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2148434489965439,
      "num_tokens": 92635731.0,
      "step": 53405
    },
    {
      "entropy": 5.816846036911011,
      "epoch": 4.155456136938339,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033710306244719954,
      "loss": 4.8894,
      "mean_token_accuracy": 0.22217011600732803,
      "num_tokens": 92644282.0,
      "step": 53410
    },
    {
      "entropy": 5.929310417175293,
      "epoch": 4.155845166310057,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003370764230357619,
      "loss": 5.0562,
      "mean_token_accuracy": 0.20064301937818527,
      "num_tokens": 92652923.0,
      "step": 53415
    },
    {
      "entropy": 5.944867897033691,
      "epoch": 4.156234195681774,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003370497826824134,
      "loss": 4.9648,
      "mean_token_accuracy": 0.2130206808447838,
      "num_tokens": 92661849.0,
      "step": 53420
    },
    {
      "entropy": 5.933525943756104,
      "epoch": 4.156623225053491,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003370231413875582,
      "loss": 4.9746,
      "mean_token_accuracy": 0.2136972203850746,
      "num_tokens": 92670152.0,
      "step": 53425
    },
    {
      "entropy": 5.871288537979126,
      "epoch": 4.157012254425209,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033699649915160053,
      "loss": 4.9181,
      "mean_token_accuracy": 0.21274385303258897,
      "num_tokens": 92678363.0,
      "step": 53430
    },
    {
      "entropy": 5.921407270431518,
      "epoch": 4.1574012837969265,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033696985597494474,
      "loss": 4.9911,
      "mean_token_accuracy": 0.21379802525043487,
      "num_tokens": 92686648.0,
      "step": 53435
    },
    {
      "entropy": 5.974479341506958,
      "epoch": 4.157790313168644,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000336943211857995,
      "loss": 5.0599,
      "mean_token_accuracy": 0.2102423459291458,
      "num_tokens": 92694965.0,
      "step": 53440
    },
    {
      "entropy": 5.905294799804688,
      "epoch": 4.158179342540362,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003369165668011556,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21884283423423767,
      "num_tokens": 92703274.0,
      "step": 53445
    },
    {
      "entropy": 5.8830845832824705,
      "epoch": 4.15856837191208,
      "grad_norm": 1.375,
      "learning_rate": 0.0003368899208048308,
      "loss": 4.9339,
      "mean_token_accuracy": 0.22338421642780304,
      "num_tokens": 92711758.0,
      "step": 53450
    },
    {
      "entropy": 5.837594795227051,
      "epoch": 4.158957401283797,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033686327386942504,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2190188854932785,
      "num_tokens": 92720090.0,
      "step": 53455
    },
    {
      "entropy": 5.8626007556915285,
      "epoch": 4.159346430655514,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003368366259953425,
      "loss": 4.9265,
      "mean_token_accuracy": 0.21472094655036927,
      "num_tokens": 92728951.0,
      "step": 53460
    },
    {
      "entropy": 5.863406467437744,
      "epoch": 4.159735460027232,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033680997718298756,
      "loss": 4.9359,
      "mean_token_accuracy": 0.211899097263813,
      "num_tokens": 92738552.0,
      "step": 53465
    },
    {
      "entropy": 5.975723838806152,
      "epoch": 4.1601244893989495,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00033678332743276466,
      "loss": 5.0033,
      "mean_token_accuracy": 0.2125203177332878,
      "num_tokens": 92747060.0,
      "step": 53470
    },
    {
      "entropy": 5.907132816314697,
      "epoch": 4.160513518770667,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003367566767450781,
      "loss": 4.951,
      "mean_token_accuracy": 0.20979761183261872,
      "num_tokens": 92755490.0,
      "step": 53475
    },
    {
      "entropy": 5.9527746677398685,
      "epoch": 4.160902548142385,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033673002512033223,
      "loss": 5.0201,
      "mean_token_accuracy": 0.21719395369291306,
      "num_tokens": 92764285.0,
      "step": 53480
    },
    {
      "entropy": 5.940624380111695,
      "epoch": 4.161291577514103,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003367033725589315,
      "loss": 4.9675,
      "mean_token_accuracy": 0.21751809269189834,
      "num_tokens": 92772282.0,
      "step": 53485
    },
    {
      "entropy": 5.929183101654052,
      "epoch": 4.16168060688582,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003366767190612804,
      "loss": 4.9861,
      "mean_token_accuracy": 0.21386780142784118,
      "num_tokens": 92782441.0,
      "step": 53490
    },
    {
      "entropy": 5.972880077362061,
      "epoch": 4.162069636257537,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003366500646277831,
      "loss": 5.0099,
      "mean_token_accuracy": 0.2056671768426895,
      "num_tokens": 92791758.0,
      "step": 53495
    },
    {
      "entropy": 5.854140138626098,
      "epoch": 4.162458665629255,
      "grad_norm": 1.375,
      "learning_rate": 0.0003366234092588443,
      "loss": 4.9142,
      "mean_token_accuracy": 0.2104744866490364,
      "num_tokens": 92800484.0,
      "step": 53500
    },
    {
      "entropy": 5.911714506149292,
      "epoch": 4.1628476950009725,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033659675295486826,
      "loss": 5.0051,
      "mean_token_accuracy": 0.2139746367931366,
      "num_tokens": 92809640.0,
      "step": 53505
    },
    {
      "entropy": 5.872619581222534,
      "epoch": 4.16323672437269,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033657009571625964,
      "loss": 4.8781,
      "mean_token_accuracy": 0.20747690945863723,
      "num_tokens": 92817810.0,
      "step": 53510
    },
    {
      "entropy": 5.940346622467041,
      "epoch": 4.163625753744408,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003365434375434227,
      "loss": 4.9514,
      "mean_token_accuracy": 0.21364800333976747,
      "num_tokens": 92826719.0,
      "step": 53515
    },
    {
      "entropy": 5.910753297805786,
      "epoch": 4.164014783116126,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033651677843676213,
      "loss": 4.9928,
      "mean_token_accuracy": 0.2066065862774849,
      "num_tokens": 92835612.0,
      "step": 53520
    },
    {
      "entropy": 5.966928243637085,
      "epoch": 4.164403812487842,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003364901183966824,
      "loss": 4.9931,
      "mean_token_accuracy": 0.21317997574806213,
      "num_tokens": 92844980.0,
      "step": 53525
    },
    {
      "entropy": 5.934340190887451,
      "epoch": 4.16479284185956,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003364634574235879,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21803470253944396,
      "num_tokens": 92853733.0,
      "step": 53530
    },
    {
      "entropy": 5.879807043075561,
      "epoch": 4.165181871231278,
      "grad_norm": 1.53125,
      "learning_rate": 0.00033643679551788325,
      "loss": 4.9022,
      "mean_token_accuracy": 0.22135170549154282,
      "num_tokens": 92861944.0,
      "step": 53535
    },
    {
      "entropy": 5.88996958732605,
      "epoch": 4.1655709006029955,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033641013267997317,
      "loss": 4.9836,
      "mean_token_accuracy": 0.21443108916282655,
      "num_tokens": 92871177.0,
      "step": 53540
    },
    {
      "entropy": 5.868205213546753,
      "epoch": 4.165959929974713,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033638346891026196,
      "loss": 4.8813,
      "mean_token_accuracy": 0.21499197781085969,
      "num_tokens": 92879848.0,
      "step": 53545
    },
    {
      "entropy": 5.91450719833374,
      "epoch": 4.166348959346431,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033635680420915424,
      "loss": 4.8863,
      "mean_token_accuracy": 0.22100455462932586,
      "num_tokens": 92888381.0,
      "step": 53550
    },
    {
      "entropy": 5.843396329879761,
      "epoch": 4.166737988718149,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003363301385770548,
      "loss": 4.8757,
      "mean_token_accuracy": 0.21630217581987382,
      "num_tokens": 92897447.0,
      "step": 53555
    },
    {
      "entropy": 5.898119115829468,
      "epoch": 4.167127018089865,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033630347201436806,
      "loss": 5.0347,
      "mean_token_accuracy": 0.2091578245162964,
      "num_tokens": 92905820.0,
      "step": 53560
    },
    {
      "entropy": 5.892198753356934,
      "epoch": 4.167516047461583,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033627680452149873,
      "loss": 5.0041,
      "mean_token_accuracy": 0.21721869856119155,
      "num_tokens": 92913933.0,
      "step": 53565
    },
    {
      "entropy": 5.884625053405761,
      "epoch": 4.167905076833301,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003362501360988514,
      "loss": 4.8906,
      "mean_token_accuracy": 0.21452568471431732,
      "num_tokens": 92922730.0,
      "step": 53570
    },
    {
      "entropy": 5.904813957214356,
      "epoch": 4.1682941062050185,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003362234667468308,
      "loss": 4.8876,
      "mean_token_accuracy": 0.21633876115083694,
      "num_tokens": 92931367.0,
      "step": 53575
    },
    {
      "entropy": 5.914651393890381,
      "epoch": 4.168683135576736,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003361967964658415,
      "loss": 4.9877,
      "mean_token_accuracy": 0.20863874703645707,
      "num_tokens": 92939806.0,
      "step": 53580
    },
    {
      "entropy": 5.942535448074341,
      "epoch": 4.169072164948454,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0003361701252562882,
      "loss": 5.0572,
      "mean_token_accuracy": 0.21090054512023926,
      "num_tokens": 92948235.0,
      "step": 53585
    },
    {
      "entropy": 5.9426638126373295,
      "epoch": 4.169461194320172,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033614345311857566,
      "loss": 4.932,
      "mean_token_accuracy": 0.21251779049634933,
      "num_tokens": 92957091.0,
      "step": 53590
    },
    {
      "entropy": 5.886492538452148,
      "epoch": 4.169850223691888,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003361167800531085,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22059376537799835,
      "num_tokens": 92966792.0,
      "step": 53595
    },
    {
      "entropy": 5.9472582817077635,
      "epoch": 4.170239253063606,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003360901060602915,
      "loss": 4.9867,
      "mean_token_accuracy": 0.2082020729780197,
      "num_tokens": 92975005.0,
      "step": 53600
    },
    {
      "entropy": 5.827044057846069,
      "epoch": 4.170628282435324,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033606343114052936,
      "loss": 4.8734,
      "mean_token_accuracy": 0.22196402102708818,
      "num_tokens": 92983532.0,
      "step": 53605
    },
    {
      "entropy": 5.92602424621582,
      "epoch": 4.1710173118070415,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033603675529422685,
      "loss": 5.0078,
      "mean_token_accuracy": 0.20374674946069718,
      "num_tokens": 92992554.0,
      "step": 53610
    },
    {
      "entropy": 5.887905788421631,
      "epoch": 4.171406341178759,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003360100785217887,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22075749486684798,
      "num_tokens": 93001596.0,
      "step": 53615
    },
    {
      "entropy": 5.947287654876709,
      "epoch": 4.171795370550477,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003359834008236198,
      "loss": 4.9896,
      "mean_token_accuracy": 0.21106477677822114,
      "num_tokens": 93010767.0,
      "step": 53620
    },
    {
      "entropy": 5.806873798370361,
      "epoch": 4.172184399922194,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003359567222001248,
      "loss": 4.8956,
      "mean_token_accuracy": 0.22674275040626526,
      "num_tokens": 93019723.0,
      "step": 53625
    },
    {
      "entropy": 5.877566146850586,
      "epoch": 4.172573429293911,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003359300426517085,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22000922560691832,
      "num_tokens": 93028642.0,
      "step": 53630
    },
    {
      "entropy": 5.893415451049805,
      "epoch": 4.172962458665629,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003359033621787759,
      "loss": 4.916,
      "mean_token_accuracy": 0.21513424217700958,
      "num_tokens": 93037789.0,
      "step": 53635
    },
    {
      "entropy": 5.82288088798523,
      "epoch": 4.173351488037347,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003358766807817317,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22486298382282258,
      "num_tokens": 93046351.0,
      "step": 53640
    },
    {
      "entropy": 5.893827724456787,
      "epoch": 4.1737405174090645,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003358499984609807,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2190068021416664,
      "num_tokens": 93054628.0,
      "step": 53645
    },
    {
      "entropy": 5.900408697128296,
      "epoch": 4.174129546780782,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003358233152169278,
      "loss": 4.9521,
      "mean_token_accuracy": 0.21950263977050782,
      "num_tokens": 93063102.0,
      "step": 53650
    },
    {
      "entropy": 5.90348539352417,
      "epoch": 4.1745185761525,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033579663104997803,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21864834427833557,
      "num_tokens": 93071874.0,
      "step": 53655
    },
    {
      "entropy": 5.838825845718384,
      "epoch": 4.174907605524217,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033576994596053606,
      "loss": 4.8693,
      "mean_token_accuracy": 0.21873281002044678,
      "num_tokens": 93080353.0,
      "step": 53660
    },
    {
      "entropy": 5.899511384963989,
      "epoch": 4.175296634895934,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003357432599490069,
      "loss": 5.0528,
      "mean_token_accuracy": 0.20657528042793274,
      "num_tokens": 93088952.0,
      "step": 53665
    },
    {
      "entropy": 5.885079002380371,
      "epoch": 4.175685664267652,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033571657301579547,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21375426948070525,
      "num_tokens": 93096901.0,
      "step": 53670
    },
    {
      "entropy": 5.925373458862305,
      "epoch": 4.17607469363937,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003356898851613067,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21541971415281297,
      "num_tokens": 93105499.0,
      "step": 53675
    },
    {
      "entropy": 6.004173040390015,
      "epoch": 4.1764637230110875,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003356631963859455,
      "loss": 5.0501,
      "mean_token_accuracy": 0.20558395683765412,
      "num_tokens": 93114238.0,
      "step": 53680
    },
    {
      "entropy": 5.933343696594238,
      "epoch": 4.176852752382805,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003356365066901168,
      "loss": 5.0051,
      "mean_token_accuracy": 0.21496293395757676,
      "num_tokens": 93123539.0,
      "step": 53685
    },
    {
      "entropy": 5.882346773147583,
      "epoch": 4.177241781754523,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033560981607422567,
      "loss": 4.8846,
      "mean_token_accuracy": 0.22057074904441834,
      "num_tokens": 93131827.0,
      "step": 53690
    },
    {
      "entropy": 5.860810470581055,
      "epoch": 4.17763081112624,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033558312453867703,
      "loss": 4.8836,
      "mean_token_accuracy": 0.22262519896030425,
      "num_tokens": 93141389.0,
      "step": 53695
    },
    {
      "entropy": 5.865365076065063,
      "epoch": 4.178019840497957,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033555643208387593,
      "loss": 4.8252,
      "mean_token_accuracy": 0.21895772963762283,
      "num_tokens": 93149564.0,
      "step": 53700
    },
    {
      "entropy": 5.91902265548706,
      "epoch": 4.178408869869675,
      "grad_norm": 1.515625,
      "learning_rate": 0.00033552973871022737,
      "loss": 4.8853,
      "mean_token_accuracy": 0.2191261500120163,
      "num_tokens": 93157386.0,
      "step": 53705
    },
    {
      "entropy": 5.935545921325684,
      "epoch": 4.178797899241393,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033550304441813634,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21027239561080932,
      "num_tokens": 93166785.0,
      "step": 53710
    },
    {
      "entropy": 5.887098932266236,
      "epoch": 4.1791869286131105,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003354763492080079,
      "loss": 4.9491,
      "mean_token_accuracy": 0.21995031386613845,
      "num_tokens": 93174971.0,
      "step": 53715
    },
    {
      "entropy": 5.903947877883911,
      "epoch": 4.179575957984828,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003354496530802472,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21165534406900405,
      "num_tokens": 93182619.0,
      "step": 53720
    },
    {
      "entropy": 5.93675127029419,
      "epoch": 4.179964987356545,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003354229560352592,
      "loss": 4.9898,
      "mean_token_accuracy": 0.2147631898522377,
      "num_tokens": 93192200.0,
      "step": 53725
    },
    {
      "entropy": 5.921198463439941,
      "epoch": 4.180354016728263,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003353962580734489,
      "loss": 4.9371,
      "mean_token_accuracy": 0.20910192728042604,
      "num_tokens": 93200310.0,
      "step": 53730
    },
    {
      "entropy": 5.917319393157959,
      "epoch": 4.18074304609998,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033536955919522167,
      "loss": 5.0222,
      "mean_token_accuracy": 0.21281781494617463,
      "num_tokens": 93209112.0,
      "step": 53735
    },
    {
      "entropy": 5.937884330749512,
      "epoch": 4.181132075471698,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003353428594009824,
      "loss": 5.0207,
      "mean_token_accuracy": 0.21170932203531265,
      "num_tokens": 93217142.0,
      "step": 53740
    },
    {
      "entropy": 5.9074506759643555,
      "epoch": 4.181521104843416,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033531615869113624,
      "loss": 4.9157,
      "mean_token_accuracy": 0.21416405886411666,
      "num_tokens": 93225483.0,
      "step": 53745
    },
    {
      "entropy": 5.943867492675781,
      "epoch": 4.1819101342151335,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033528945706608844,
      "loss": 4.9924,
      "mean_token_accuracy": 0.20643789023160936,
      "num_tokens": 93234292.0,
      "step": 53750
    },
    {
      "entropy": 5.926475620269775,
      "epoch": 4.182299163586851,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033526275452624407,
      "loss": 4.9624,
      "mean_token_accuracy": 0.21229031234979628,
      "num_tokens": 93242837.0,
      "step": 53755
    },
    {
      "entropy": 5.905855894088745,
      "epoch": 4.182688192958568,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033523605107200826,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22321490645408631,
      "num_tokens": 93251763.0,
      "step": 53760
    },
    {
      "entropy": 5.8698139667510985,
      "epoch": 4.183077222330286,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033520934670378634,
      "loss": 4.8434,
      "mean_token_accuracy": 0.22865576595067977,
      "num_tokens": 93260262.0,
      "step": 53765
    },
    {
      "entropy": 5.915475130081177,
      "epoch": 4.183466251702003,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033518264142198335,
      "loss": 4.9522,
      "mean_token_accuracy": 0.21047503054141997,
      "num_tokens": 93268924.0,
      "step": 53770
    },
    {
      "entropy": 5.962980842590332,
      "epoch": 4.183855281073721,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003351559352270047,
      "loss": 4.9913,
      "mean_token_accuracy": 0.21400743424892427,
      "num_tokens": 93278073.0,
      "step": 53775
    },
    {
      "entropy": 5.911512422561645,
      "epoch": 4.184244310445439,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033512922811925535,
      "loss": 4.9297,
      "mean_token_accuracy": 0.21192835718393327,
      "num_tokens": 93287164.0,
      "step": 53780
    },
    {
      "entropy": 5.888061952590943,
      "epoch": 4.1846333398171565,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003351025200991406,
      "loss": 4.9979,
      "mean_token_accuracy": 0.21248734891414642,
      "num_tokens": 93295687.0,
      "step": 53785
    },
    {
      "entropy": 5.93458366394043,
      "epoch": 4.185022369188874,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003350758111670659,
      "loss": 4.9655,
      "mean_token_accuracy": 0.21652792543172836,
      "num_tokens": 93304334.0,
      "step": 53790
    },
    {
      "entropy": 5.917180824279785,
      "epoch": 4.185411398560591,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003350491013234364,
      "loss": 4.9023,
      "mean_token_accuracy": 0.2128070890903473,
      "num_tokens": 93312370.0,
      "step": 53795
    },
    {
      "entropy": 5.933994197845459,
      "epoch": 4.185800427932309,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003350223905686573,
      "loss": 5.0212,
      "mean_token_accuracy": 0.20935048907995224,
      "num_tokens": 93320310.0,
      "step": 53800
    },
    {
      "entropy": 5.916634845733642,
      "epoch": 4.186189457304026,
      "grad_norm": 1.5,
      "learning_rate": 0.000334995678903134,
      "loss": 4.8874,
      "mean_token_accuracy": 0.21783160269260407,
      "num_tokens": 93328480.0,
      "step": 53805
    },
    {
      "entropy": 5.90133228302002,
      "epoch": 4.186578486675744,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0003349689663272718,
      "loss": 5.008,
      "mean_token_accuracy": 0.21822841018438338,
      "num_tokens": 93337643.0,
      "step": 53810
    },
    {
      "entropy": 5.978509950637817,
      "epoch": 4.186967516047462,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003349422528414759,
      "loss": 4.989,
      "mean_token_accuracy": 0.2182777240872383,
      "num_tokens": 93346216.0,
      "step": 53815
    },
    {
      "entropy": 6.004340744018554,
      "epoch": 4.1873565454191795,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003349155384461517,
      "loss": 5.0473,
      "mean_token_accuracy": 0.20166324377059935,
      "num_tokens": 93354713.0,
      "step": 53820
    },
    {
      "entropy": 6.00973916053772,
      "epoch": 4.187745574790897,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00033488882314170463,
      "loss": 5.0542,
      "mean_token_accuracy": 0.2076243758201599,
      "num_tokens": 93363373.0,
      "step": 53825
    },
    {
      "entropy": 5.992966222763061,
      "epoch": 4.188134604162614,
      "grad_norm": 1.5625,
      "learning_rate": 0.00033486210692854,
      "loss": 4.9618,
      "mean_token_accuracy": 0.21450628489255905,
      "num_tokens": 93370960.0,
      "step": 53830
    },
    {
      "entropy": 5.9579085350036625,
      "epoch": 4.188523633534332,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003348353898070631,
      "loss": 4.9739,
      "mean_token_accuracy": 0.21519456803798676,
      "num_tokens": 93379807.0,
      "step": 53835
    },
    {
      "entropy": 5.829697704315185,
      "epoch": 4.188912662906049,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003348086717776795,
      "loss": 4.8287,
      "mean_token_accuracy": 0.22600819915533066,
      "num_tokens": 93388411.0,
      "step": 53840
    },
    {
      "entropy": 5.90871958732605,
      "epoch": 4.189301692277767,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00033478195284079447,
      "loss": 4.9608,
      "mean_token_accuracy": 0.21399974822998047,
      "num_tokens": 93396870.0,
      "step": 53845
    },
    {
      "entropy": 5.899751091003418,
      "epoch": 4.189690721649485,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003347552329968134,
      "loss": 4.8938,
      "mean_token_accuracy": 0.22584719955921173,
      "num_tokens": 93405065.0,
      "step": 53850
    },
    {
      "entropy": 5.886652231216431,
      "epoch": 4.1900797510212024,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003347285122461419,
      "loss": 4.8293,
      "mean_token_accuracy": 0.22612337321043013,
      "num_tokens": 93413842.0,
      "step": 53855
    },
    {
      "entropy": 5.904062795639038,
      "epoch": 4.190468780392919,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003347017905891852,
      "loss": 4.9102,
      "mean_token_accuracy": 0.21231829077005387,
      "num_tokens": 93423269.0,
      "step": 53860
    },
    {
      "entropy": 5.839717769622803,
      "epoch": 4.190857809764637,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003346750680263488,
      "loss": 4.9288,
      "mean_token_accuracy": 0.21997540444135666,
      "num_tokens": 93432294.0,
      "step": 53865
    },
    {
      "entropy": 5.881240177154541,
      "epoch": 4.191246839136355,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003346483445580384,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21836688816547395,
      "num_tokens": 93440791.0,
      "step": 53870
    },
    {
      "entropy": 5.925094842910767,
      "epoch": 4.191635868508072,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033462162018465917,
      "loss": 4.8938,
      "mean_token_accuracy": 0.21971093416213988,
      "num_tokens": 93449256.0,
      "step": 53875
    },
    {
      "entropy": 5.926153707504272,
      "epoch": 4.19202489787979,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033459489490661676,
      "loss": 4.9394,
      "mean_token_accuracy": 0.2190354883670807,
      "num_tokens": 93457680.0,
      "step": 53880
    },
    {
      "entropy": 5.881261587142944,
      "epoch": 4.192413927251508,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033456816872431673,
      "loss": 4.9293,
      "mean_token_accuracy": 0.22156398892402648,
      "num_tokens": 93465975.0,
      "step": 53885
    },
    {
      "entropy": 5.902797746658325,
      "epoch": 4.192802956623225,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003345414416381645,
      "loss": 4.9472,
      "mean_token_accuracy": 0.22233942598104478,
      "num_tokens": 93475522.0,
      "step": 53890
    },
    {
      "entropy": 5.882423782348633,
      "epoch": 4.193191985994942,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00033451471364856565,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21614208817481995,
      "num_tokens": 93483876.0,
      "step": 53895
    },
    {
      "entropy": 5.975253582000732,
      "epoch": 4.19358101536666,
      "grad_norm": 1.453125,
      "learning_rate": 0.00033448798475592585,
      "loss": 5.0129,
      "mean_token_accuracy": 0.2099709615111351,
      "num_tokens": 93491628.0,
      "step": 53900
    },
    {
      "entropy": 5.8995108127594,
      "epoch": 4.193970044738378,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003344612549606505,
      "loss": 4.9584,
      "mean_token_accuracy": 0.2150569438934326,
      "num_tokens": 93500513.0,
      "step": 53905
    },
    {
      "entropy": 5.863328742980957,
      "epoch": 4.194359074110095,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003344345242631453,
      "loss": 4.882,
      "mean_token_accuracy": 0.220084285736084,
      "num_tokens": 93509375.0,
      "step": 53910
    },
    {
      "entropy": 5.986605405807495,
      "epoch": 4.194748103481813,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003344077926638157,
      "loss": 4.9943,
      "mean_token_accuracy": 0.2117048218846321,
      "num_tokens": 93517824.0,
      "step": 53915
    },
    {
      "entropy": 5.922944688796997,
      "epoch": 4.195137132853531,
      "grad_norm": 1.515625,
      "learning_rate": 0.00033438106016306753,
      "loss": 4.8879,
      "mean_token_accuracy": 0.22066328078508377,
      "num_tokens": 93526224.0,
      "step": 53920
    },
    {
      "entropy": 5.8422276973724365,
      "epoch": 4.195526162225248,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003343543267613062,
      "loss": 4.9503,
      "mean_token_accuracy": 0.2202147737145424,
      "num_tokens": 93534817.0,
      "step": 53925
    },
    {
      "entropy": 5.897165727615357,
      "epoch": 4.195915191596965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003343275924589374,
      "loss": 5.032,
      "mean_token_accuracy": 0.20735633969306946,
      "num_tokens": 93544445.0,
      "step": 53930
    },
    {
      "entropy": 5.9483850479125975,
      "epoch": 4.196304220968683,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033430085725636685,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21511916518211366,
      "num_tokens": 93553756.0,
      "step": 53935
    },
    {
      "entropy": 5.893982887268066,
      "epoch": 4.196693250340401,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003342741211540002,
      "loss": 4.8816,
      "mean_token_accuracy": 0.21888074576854705,
      "num_tokens": 93562118.0,
      "step": 53940
    },
    {
      "entropy": 5.804582786560059,
      "epoch": 4.197082279712118,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003342473841522431,
      "loss": 4.8467,
      "mean_token_accuracy": 0.22164036929607392,
      "num_tokens": 93570594.0,
      "step": 53945
    },
    {
      "entropy": 5.8978447914123535,
      "epoch": 4.197471309083836,
      "grad_norm": 1.484375,
      "learning_rate": 0.00033422064625150124,
      "loss": 4.9421,
      "mean_token_accuracy": 0.21479247659444808,
      "num_tokens": 93579604.0,
      "step": 53950
    },
    {
      "entropy": 5.907360458374024,
      "epoch": 4.197860338455554,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003341939074521804,
      "loss": 4.9488,
      "mean_token_accuracy": 0.2215973913669586,
      "num_tokens": 93588489.0,
      "step": 53955
    },
    {
      "entropy": 5.912613201141357,
      "epoch": 4.1982493678272705,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003341671677546861,
      "loss": 4.946,
      "mean_token_accuracy": 0.2156466543674469,
      "num_tokens": 93596935.0,
      "step": 53960
    },
    {
      "entropy": 5.909515285491944,
      "epoch": 4.198638397198988,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00033414042715942423,
      "loss": 5.0025,
      "mean_token_accuracy": 0.21308566331863404,
      "num_tokens": 93605826.0,
      "step": 53965
    },
    {
      "entropy": 5.960588026046753,
      "epoch": 4.199027426570706,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003341136856668005,
      "loss": 4.9914,
      "mean_token_accuracy": 0.21759190112352372,
      "num_tokens": 93614077.0,
      "step": 53970
    },
    {
      "entropy": 5.910334491729737,
      "epoch": 4.199416455942424,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003340869432772207,
      "loss": 4.8742,
      "mean_token_accuracy": 0.2211216390132904,
      "num_tokens": 93622891.0,
      "step": 53975
    },
    {
      "entropy": 5.883740997314453,
      "epoch": 4.199805485314141,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033406019999109065,
      "loss": 4.9221,
      "mean_token_accuracy": 0.21255738884210587,
      "num_tokens": 93631454.0,
      "step": 53980
    },
    {
      "entropy": 5.868894004821778,
      "epoch": 4.200194514685859,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033403345580881606,
      "loss": 4.889,
      "mean_token_accuracy": 0.21944913268089294,
      "num_tokens": 93639707.0,
      "step": 53985
    },
    {
      "entropy": 5.896656131744384,
      "epoch": 4.200583544057577,
      "grad_norm": 1.53125,
      "learning_rate": 0.00033400671073080274,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21386097371578217,
      "num_tokens": 93648285.0,
      "step": 53990
    },
    {
      "entropy": 5.922615671157837,
      "epoch": 4.2009725734292935,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003339799647574564,
      "loss": 5.0525,
      "mean_token_accuracy": 0.2110658124089241,
      "num_tokens": 93656747.0,
      "step": 53995
    },
    {
      "entropy": 5.931103897094727,
      "epoch": 4.201361602801011,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003339532178891831,
      "loss": 4.9889,
      "mean_token_accuracy": 0.21166858077049255,
      "num_tokens": 93665650.0,
      "step": 54000
    },
    {
      "epoch": 4.201361602801011,
      "eval_entropy": 5.642706782824725,
      "eval_loss": 5.066500663757324,
      "eval_mean_token_accuracy": 0.21840525301920896,
      "eval_num_tokens": 93665650.0,
      "eval_runtime": 21.6617,
      "eval_samples_per_second": 1918.502,
      "eval_steps_per_second": 239.824,
      "step": 54000
    },
    {
      "entropy": 5.925408506393433,
      "epoch": 4.201750632172729,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003339264701263885,
      "loss": 5.0281,
      "mean_token_accuracy": 0.20960671454668045,
      "num_tokens": 93674057.0,
      "step": 54005
    },
    {
      "entropy": 5.881590080261231,
      "epoch": 4.202139661544447,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003338997214694784,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2158917173743248,
      "num_tokens": 93682392.0,
      "step": 54010
    },
    {
      "entropy": 5.970138740539551,
      "epoch": 4.202528690916164,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00033387297191885896,
      "loss": 5.0046,
      "mean_token_accuracy": 0.2105448827147484,
      "num_tokens": 93691477.0,
      "step": 54015
    },
    {
      "entropy": 5.901133489608765,
      "epoch": 4.202917720287882,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003338462214749358,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22796996533870698,
      "num_tokens": 93700398.0,
      "step": 54020
    },
    {
      "entropy": 5.831226778030396,
      "epoch": 4.2033067496596,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003338194701381148,
      "loss": 4.9253,
      "mean_token_accuracy": 0.21541440188884736,
      "num_tokens": 93709149.0,
      "step": 54025
    },
    {
      "entropy": 5.961451625823974,
      "epoch": 4.2036957790313165,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033379271790880203,
      "loss": 5.079,
      "mean_token_accuracy": 0.20318375825881957,
      "num_tokens": 93717355.0,
      "step": 54030
    },
    {
      "entropy": 5.951258134841919,
      "epoch": 4.204084808403034,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003337659647874034,
      "loss": 4.9909,
      "mean_token_accuracy": 0.20885471105575562,
      "num_tokens": 93725798.0,
      "step": 54035
    },
    {
      "entropy": 5.92584080696106,
      "epoch": 4.204473837774752,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003337392107743248,
      "loss": 5.0101,
      "mean_token_accuracy": 0.21253875643014908,
      "num_tokens": 93734526.0,
      "step": 54040
    },
    {
      "entropy": 5.9423479557037355,
      "epoch": 4.20486286714647,
      "grad_norm": 1.375,
      "learning_rate": 0.0003337124558699721,
      "loss": 5.0527,
      "mean_token_accuracy": 0.20558874756097795,
      "num_tokens": 93742611.0,
      "step": 54045
    },
    {
      "entropy": 5.870625734329224,
      "epoch": 4.205251896518187,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033368570007475133,
      "loss": 4.8966,
      "mean_token_accuracy": 0.21827151328325273,
      "num_tokens": 93750852.0,
      "step": 54050
    },
    {
      "entropy": 5.967768383026123,
      "epoch": 4.205640925889905,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003336589433890685,
      "loss": 5.0354,
      "mean_token_accuracy": 0.21154582798480986,
      "num_tokens": 93759382.0,
      "step": 54055
    },
    {
      "entropy": 5.922215700149536,
      "epoch": 4.206029955261622,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033363218581332956,
      "loss": 4.9486,
      "mean_token_accuracy": 0.22194952964782716,
      "num_tokens": 93768628.0,
      "step": 54060
    },
    {
      "entropy": 5.972916030883789,
      "epoch": 4.2064189846333395,
      "grad_norm": 1.5,
      "learning_rate": 0.0003336054273479405,
      "loss": 5.0597,
      "mean_token_accuracy": 0.20634615868330003,
      "num_tokens": 93776946.0,
      "step": 54065
    },
    {
      "entropy": 5.939820098876953,
      "epoch": 4.206808014005057,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033357866799330743,
      "loss": 5.0072,
      "mean_token_accuracy": 0.20996624678373338,
      "num_tokens": 93786844.0,
      "step": 54070
    },
    {
      "entropy": 5.945168447494507,
      "epoch": 4.207197043376775,
      "grad_norm": 1.25,
      "learning_rate": 0.00033355190774983626,
      "loss": 4.9389,
      "mean_token_accuracy": 0.22278840988874435,
      "num_tokens": 93795341.0,
      "step": 54075
    },
    {
      "entropy": 5.841437435150146,
      "epoch": 4.207586072748493,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033352514661793306,
      "loss": 4.8943,
      "mean_token_accuracy": 0.21833660900592805,
      "num_tokens": 93804035.0,
      "step": 54080
    },
    {
      "entropy": 5.90973653793335,
      "epoch": 4.20797510212021,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033349838459800394,
      "loss": 4.9321,
      "mean_token_accuracy": 0.21638155430555345,
      "num_tokens": 93813127.0,
      "step": 54085
    },
    {
      "entropy": 5.898960304260254,
      "epoch": 4.208364131491928,
      "grad_norm": 1.3125,
      "learning_rate": 0.000333471621690455,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21833309382200242,
      "num_tokens": 93821993.0,
      "step": 54090
    },
    {
      "entropy": 5.954837656021118,
      "epoch": 4.208753160863645,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003334448578956921,
      "loss": 4.9928,
      "mean_token_accuracy": 0.21387771368026734,
      "num_tokens": 93830944.0,
      "step": 54095
    },
    {
      "entropy": 5.9551682472229,
      "epoch": 4.2091421902353625,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003334180932141217,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21888377517461777,
      "num_tokens": 93839088.0,
      "step": 54100
    },
    {
      "entropy": 5.914205741882324,
      "epoch": 4.20953121960708,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003333913276461497,
      "loss": 4.9744,
      "mean_token_accuracy": 0.21496440470218658,
      "num_tokens": 93848426.0,
      "step": 54105
    },
    {
      "entropy": 5.887797212600708,
      "epoch": 4.209920248978798,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033336456119218214,
      "loss": 4.9328,
      "mean_token_accuracy": 0.21225731074810028,
      "num_tokens": 93857455.0,
      "step": 54110
    },
    {
      "entropy": 5.952798509597779,
      "epoch": 4.210309278350516,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033333779385262525,
      "loss": 5.0353,
      "mean_token_accuracy": 0.21044208705425263,
      "num_tokens": 93866331.0,
      "step": 54115
    },
    {
      "entropy": 5.916996765136719,
      "epoch": 4.210698307722233,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033331102562788527,
      "loss": 4.9415,
      "mean_token_accuracy": 0.21817522794008254,
      "num_tokens": 93875087.0,
      "step": 54120
    },
    {
      "entropy": 5.89392056465149,
      "epoch": 4.211087337093951,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003332842565183682,
      "loss": 5.0075,
      "mean_token_accuracy": 0.2094162881374359,
      "num_tokens": 93883248.0,
      "step": 54125
    },
    {
      "entropy": 5.954900693893433,
      "epoch": 4.211476366465668,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033325748652448037,
      "loss": 4.9161,
      "mean_token_accuracy": 0.22542114853858947,
      "num_tokens": 93892494.0,
      "step": 54130
    },
    {
      "entropy": 5.928280210494995,
      "epoch": 4.2118653958373855,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033323071564662796,
      "loss": 5.1005,
      "mean_token_accuracy": 0.20460900962352752,
      "num_tokens": 93901206.0,
      "step": 54135
    },
    {
      "entropy": 5.857502746582031,
      "epoch": 4.212254425209103,
      "grad_norm": 1.53125,
      "learning_rate": 0.00033320394388521707,
      "loss": 4.8844,
      "mean_token_accuracy": 0.22027831226587297,
      "num_tokens": 93909862.0,
      "step": 54140
    },
    {
      "entropy": 5.964752817153931,
      "epoch": 4.212643454580821,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000333177171240654,
      "loss": 4.9896,
      "mean_token_accuracy": 0.21682259440422058,
      "num_tokens": 93918744.0,
      "step": 54145
    },
    {
      "entropy": 6.002886199951172,
      "epoch": 4.213032483952539,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00033315039771334494,
      "loss": 5.0992,
      "mean_token_accuracy": 0.20871153175830842,
      "num_tokens": 93927382.0,
      "step": 54150
    },
    {
      "entropy": 5.95248851776123,
      "epoch": 4.213421513324256,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003331236233036961,
      "loss": 4.9315,
      "mean_token_accuracy": 0.22024007588624955,
      "num_tokens": 93936451.0,
      "step": 54155
    },
    {
      "entropy": 5.895279741287231,
      "epoch": 4.213810542695974,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003330968480121138,
      "loss": 4.9389,
      "mean_token_accuracy": 0.2089978352189064,
      "num_tokens": 93944995.0,
      "step": 54160
    },
    {
      "entropy": 5.871384525299073,
      "epoch": 4.214199572067691,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0003330700718390043,
      "loss": 4.902,
      "mean_token_accuracy": 0.22543398141860962,
      "num_tokens": 93953377.0,
      "step": 54165
    },
    {
      "entropy": 5.927482557296753,
      "epoch": 4.2145886014394085,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033304329478477395,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21452526599168778,
      "num_tokens": 93961687.0,
      "step": 54170
    },
    {
      "entropy": 5.868473100662231,
      "epoch": 4.214977630811126,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033301651684982894,
      "loss": 4.8801,
      "mean_token_accuracy": 0.21817435920238495,
      "num_tokens": 93970244.0,
      "step": 54175
    },
    {
      "entropy": 5.8881354331970215,
      "epoch": 4.215366660182844,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003329897380345757,
      "loss": 4.9813,
      "mean_token_accuracy": 0.21658388376235962,
      "num_tokens": 93979278.0,
      "step": 54180
    },
    {
      "entropy": 5.948899221420288,
      "epoch": 4.215755689554562,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003329629583394204,
      "loss": 4.9405,
      "mean_token_accuracy": 0.2112373948097229,
      "num_tokens": 93987651.0,
      "step": 54185
    },
    {
      "entropy": 5.8548589706420895,
      "epoch": 4.216144718926279,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003329361777647695,
      "loss": 4.8673,
      "mean_token_accuracy": 0.2209475204348564,
      "num_tokens": 93996043.0,
      "step": 54190
    },
    {
      "entropy": 5.859514045715332,
      "epoch": 4.216533748297996,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003329093963110293,
      "loss": 4.9828,
      "mean_token_accuracy": 0.21020440608263016,
      "num_tokens": 94005121.0,
      "step": 54195
    },
    {
      "entropy": 5.900605058670044,
      "epoch": 4.216922777669714,
      "grad_norm": 1.375,
      "learning_rate": 0.00033288261397860613,
      "loss": 4.9382,
      "mean_token_accuracy": 0.21281201541423797,
      "num_tokens": 94013944.0,
      "step": 54200
    },
    {
      "entropy": 5.913072967529297,
      "epoch": 4.2173118070414315,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033285583076790646,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21801962256431578,
      "num_tokens": 94022909.0,
      "step": 54205
    },
    {
      "entropy": 6.013250732421875,
      "epoch": 4.217700836413149,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033282904667933675,
      "loss": 5.0258,
      "mean_token_accuracy": 0.20392292886972427,
      "num_tokens": 94031042.0,
      "step": 54210
    },
    {
      "entropy": 5.86919412612915,
      "epoch": 4.218089865784867,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003328022617133032,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21241244077682495,
      "num_tokens": 94039695.0,
      "step": 54215
    },
    {
      "entropy": 5.774935579299926,
      "epoch": 4.218478895156585,
      "grad_norm": 1.375,
      "learning_rate": 0.0003327754758702124,
      "loss": 4.8901,
      "mean_token_accuracy": 0.21944576799869536,
      "num_tokens": 94048359.0,
      "step": 54220
    },
    {
      "entropy": 6.008325910568237,
      "epoch": 4.218867924528302,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033274868915047064,
      "loss": 5.073,
      "mean_token_accuracy": 0.20435744673013687,
      "num_tokens": 94057113.0,
      "step": 54225
    },
    {
      "entropy": 5.994015169143677,
      "epoch": 4.219256953900019,
      "grad_norm": 1.453125,
      "learning_rate": 0.00033272190155448446,
      "loss": 4.9844,
      "mean_token_accuracy": 0.21734994351863862,
      "num_tokens": 94065104.0,
      "step": 54230
    },
    {
      "entropy": 5.852358436584472,
      "epoch": 4.219645983271737,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003326951130826603,
      "loss": 4.895,
      "mean_token_accuracy": 0.21090505868196488,
      "num_tokens": 94072886.0,
      "step": 54235
    },
    {
      "entropy": 5.902312326431274,
      "epoch": 4.2200350126434545,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033266832373540463,
      "loss": 4.9473,
      "mean_token_accuracy": 0.21707794517278672,
      "num_tokens": 94081863.0,
      "step": 54240
    },
    {
      "entropy": 5.924773216247559,
      "epoch": 4.220424042015172,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003326415335131239,
      "loss": 4.9523,
      "mean_token_accuracy": 0.21685580760240555,
      "num_tokens": 94090640.0,
      "step": 54245
    },
    {
      "entropy": 5.881767749786377,
      "epoch": 4.22081307138689,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033261474241622475,
      "loss": 4.8746,
      "mean_token_accuracy": 0.2189313441514969,
      "num_tokens": 94099466.0,
      "step": 54250
    },
    {
      "entropy": 5.911907768249511,
      "epoch": 4.221202100758608,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003325879504451135,
      "loss": 4.9673,
      "mean_token_accuracy": 0.20509982258081436,
      "num_tokens": 94108515.0,
      "step": 54255
    },
    {
      "entropy": 5.882394552230835,
      "epoch": 4.221591130130324,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033256115760019677,
      "loss": 4.9157,
      "mean_token_accuracy": 0.21317753791809083,
      "num_tokens": 94116901.0,
      "step": 54260
    },
    {
      "entropy": 5.92430100440979,
      "epoch": 4.221980159502042,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033253436388188106,
      "loss": 4.962,
      "mean_token_accuracy": 0.21298840939998626,
      "num_tokens": 94125797.0,
      "step": 54265
    },
    {
      "entropy": 5.9713661670684814,
      "epoch": 4.22236918887376,
      "grad_norm": 1.265625,
      "learning_rate": 0.000332507569290573,
      "loss": 5.0341,
      "mean_token_accuracy": 0.21007930338382722,
      "num_tokens": 94135446.0,
      "step": 54270
    },
    {
      "entropy": 5.925769948959351,
      "epoch": 4.2227582182454775,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033248077382667913,
      "loss": 4.9134,
      "mean_token_accuracy": 0.2297309160232544,
      "num_tokens": 94144143.0,
      "step": 54275
    },
    {
      "entropy": 5.880003023147583,
      "epoch": 4.223147247617195,
      "grad_norm": 1.46875,
      "learning_rate": 0.000332453977490606,
      "loss": 4.8989,
      "mean_token_accuracy": 0.22096201181411743,
      "num_tokens": 94152831.0,
      "step": 54280
    },
    {
      "entropy": 5.836392784118653,
      "epoch": 4.223536276988913,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003324271802827602,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2187737450003624,
      "num_tokens": 94161405.0,
      "step": 54285
    },
    {
      "entropy": 5.920187377929688,
      "epoch": 4.223925306360631,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003324003822035483,
      "loss": 4.9777,
      "mean_token_accuracy": 0.20955915600061417,
      "num_tokens": 94170141.0,
      "step": 54290
    },
    {
      "entropy": 5.876787090301514,
      "epoch": 4.224314335732347,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003323735832533771,
      "loss": 4.9864,
      "mean_token_accuracy": 0.21559284925460814,
      "num_tokens": 94180023.0,
      "step": 54295
    },
    {
      "entropy": 5.978171539306641,
      "epoch": 4.224703365104065,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033234678343265297,
      "loss": 5.0316,
      "mean_token_accuracy": 0.20986644327640533,
      "num_tokens": 94189811.0,
      "step": 54300
    },
    {
      "entropy": 5.9358320236206055,
      "epoch": 4.225092394475783,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003323199827417827,
      "loss": 5.0001,
      "mean_token_accuracy": 0.20926810801029205,
      "num_tokens": 94199666.0,
      "step": 54305
    },
    {
      "entropy": 5.930307817459107,
      "epoch": 4.2254814238475005,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033229318118117293,
      "loss": 4.9429,
      "mean_token_accuracy": 0.21365652978420258,
      "num_tokens": 94208777.0,
      "step": 54310
    },
    {
      "entropy": 6.046535921096802,
      "epoch": 4.225870453219218,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00033226637875123026,
      "loss": 5.1664,
      "mean_token_accuracy": 0.20480982810258866,
      "num_tokens": 94217386.0,
      "step": 54315
    },
    {
      "entropy": 5.876422023773193,
      "epoch": 4.226259482590936,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033223957545236156,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22085874378681183,
      "num_tokens": 94226630.0,
      "step": 54320
    },
    {
      "entropy": 5.865003967285157,
      "epoch": 4.226648511962654,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003322127712849733,
      "loss": 4.9247,
      "mean_token_accuracy": 0.2125531405210495,
      "num_tokens": 94235420.0,
      "step": 54325
    },
    {
      "entropy": 5.966611862182617,
      "epoch": 4.22703754133437,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003321859662494724,
      "loss": 4.9855,
      "mean_token_accuracy": 0.2154558315873146,
      "num_tokens": 94243773.0,
      "step": 54330
    },
    {
      "entropy": 5.93405966758728,
      "epoch": 4.227426570706088,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003321591603462654,
      "loss": 4.9784,
      "mean_token_accuracy": 0.21020043641328812,
      "num_tokens": 94251975.0,
      "step": 54335
    },
    {
      "entropy": 5.903223705291748,
      "epoch": 4.227815600077806,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003321323535757592,
      "loss": 4.9662,
      "mean_token_accuracy": 0.2145107328891754,
      "num_tokens": 94261486.0,
      "step": 54340
    },
    {
      "entropy": 5.948007965087891,
      "epoch": 4.2282046294495235,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003321055459383604,
      "loss": 4.9776,
      "mean_token_accuracy": 0.2106636196374893,
      "num_tokens": 94269706.0,
      "step": 54345
    },
    {
      "entropy": 5.914697980880737,
      "epoch": 4.228593658821241,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033207873743447587,
      "loss": 4.9466,
      "mean_token_accuracy": 0.22060708552598954,
      "num_tokens": 94277861.0,
      "step": 54350
    },
    {
      "entropy": 5.906025409698486,
      "epoch": 4.228982688192959,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033205192806451227,
      "loss": 4.9685,
      "mean_token_accuracy": 0.21609849929809571,
      "num_tokens": 94286012.0,
      "step": 54355
    },
    {
      "entropy": 5.895386743545532,
      "epoch": 4.229371717564677,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033202511782887655,
      "loss": 4.9669,
      "mean_token_accuracy": 0.20952712297439574,
      "num_tokens": 94294404.0,
      "step": 54360
    },
    {
      "entropy": 6.004039335250854,
      "epoch": 4.229760746936393,
      "grad_norm": 1.5,
      "learning_rate": 0.00033199830672797545,
      "loss": 5.0367,
      "mean_token_accuracy": 0.20188789814710617,
      "num_tokens": 94302716.0,
      "step": 54365
    },
    {
      "entropy": 5.90697135925293,
      "epoch": 4.230149776308111,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003319714947622157,
      "loss": 4.917,
      "mean_token_accuracy": 0.21238841563463212,
      "num_tokens": 94311311.0,
      "step": 54370
    },
    {
      "entropy": 5.9348108768463135,
      "epoch": 4.230538805679829,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003319446819320043,
      "loss": 4.9832,
      "mean_token_accuracy": 0.20762578547000884,
      "num_tokens": 94319889.0,
      "step": 54375
    },
    {
      "entropy": 5.870354032516479,
      "epoch": 4.2309278350515465,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003319178682377479,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2257503256201744,
      "num_tokens": 94328378.0,
      "step": 54380
    },
    {
      "entropy": 5.906464862823486,
      "epoch": 4.231316864423264,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033189105367985343,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21812800467014312,
      "num_tokens": 94336828.0,
      "step": 54385
    },
    {
      "entropy": 5.8827025413513185,
      "epoch": 4.231705893794982,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033186423825872787,
      "loss": 4.9482,
      "mean_token_accuracy": 0.20942198783159255,
      "num_tokens": 94346574.0,
      "step": 54390
    },
    {
      "entropy": 5.935249185562133,
      "epoch": 4.232094923166699,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0003318374219747779,
      "loss": 4.9022,
      "mean_token_accuracy": 0.22116161584854127,
      "num_tokens": 94354159.0,
      "step": 54395
    },
    {
      "entropy": 5.9972052574157715,
      "epoch": 4.232483952538416,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033181060482841054,
      "loss": 5.0578,
      "mean_token_accuracy": 0.21034317165613176,
      "num_tokens": 94363396.0,
      "step": 54400
    },
    {
      "entropy": 5.972744655609131,
      "epoch": 4.232872981910134,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003317837868200327,
      "loss": 5.0284,
      "mean_token_accuracy": 0.20554822087287902,
      "num_tokens": 94371958.0,
      "step": 54405
    },
    {
      "entropy": 5.875565433502198,
      "epoch": 4.233262011281852,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033175696795005133,
      "loss": 4.9789,
      "mean_token_accuracy": 0.21215754449367524,
      "num_tokens": 94380574.0,
      "step": 54410
    },
    {
      "entropy": 5.841039943695068,
      "epoch": 4.2336510406535695,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003317301482188732,
      "loss": 4.9009,
      "mean_token_accuracy": 0.21745998561382293,
      "num_tokens": 94389102.0,
      "step": 54415
    },
    {
      "entropy": 5.917747497558594,
      "epoch": 4.234040070025287,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003317033276269054,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21753203123807907,
      "num_tokens": 94397125.0,
      "step": 54420
    },
    {
      "entropy": 5.927219486236572,
      "epoch": 4.234429099397005,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00033167650617455484,
      "loss": 4.9072,
      "mean_token_accuracy": 0.22607884407043458,
      "num_tokens": 94405075.0,
      "step": 54425
    },
    {
      "entropy": 5.89725923538208,
      "epoch": 4.234818128768722,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033164968386222856,
      "loss": 4.9452,
      "mean_token_accuracy": 0.21272364854812623,
      "num_tokens": 94414711.0,
      "step": 54430
    },
    {
      "entropy": 5.893005084991455,
      "epoch": 4.235207158140439,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003316228606903334,
      "loss": 4.9326,
      "mean_token_accuracy": 0.2200268566608429,
      "num_tokens": 94422832.0,
      "step": 54435
    },
    {
      "entropy": 5.840175485610962,
      "epoch": 4.235596187512157,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033159603665927655,
      "loss": 4.903,
      "mean_token_accuracy": 0.2191663846373558,
      "num_tokens": 94432503.0,
      "step": 54440
    },
    {
      "entropy": 5.922314548492432,
      "epoch": 4.235985216883875,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00033156921176946496,
      "loss": 4.9478,
      "mean_token_accuracy": 0.21343046724796294,
      "num_tokens": 94441059.0,
      "step": 54445
    },
    {
      "entropy": 5.9446971893310545,
      "epoch": 4.2363742462555924,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003315423860213057,
      "loss": 4.9801,
      "mean_token_accuracy": 0.2152017742395401,
      "num_tokens": 94450293.0,
      "step": 54450
    },
    {
      "entropy": 5.947464561462402,
      "epoch": 4.23676327562731,
      "grad_norm": 1.6875,
      "learning_rate": 0.0003315155594152055,
      "loss": 4.9878,
      "mean_token_accuracy": 0.2186127245426178,
      "num_tokens": 94458899.0,
      "step": 54455
    },
    {
      "entropy": 5.8924213409423825,
      "epoch": 4.237152304999028,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003314887319515717,
      "loss": 4.8917,
      "mean_token_accuracy": 0.21822055131196977,
      "num_tokens": 94467318.0,
      "step": 54460
    },
    {
      "entropy": 5.871268558502197,
      "epoch": 4.237541334370745,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003314619036308113,
      "loss": 4.9608,
      "mean_token_accuracy": 0.21997833102941514,
      "num_tokens": 94476464.0,
      "step": 54465
    },
    {
      "entropy": 5.917066144943237,
      "epoch": 4.237930363742462,
      "grad_norm": 1.5,
      "learning_rate": 0.00033143507445333145,
      "loss": 4.9196,
      "mean_token_accuracy": 0.2172902688384056,
      "num_tokens": 94484897.0,
      "step": 54470
    },
    {
      "entropy": 5.948601865768433,
      "epoch": 4.23831939311418,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033140824441953913,
      "loss": 4.9846,
      "mean_token_accuracy": 0.21147470623254777,
      "num_tokens": 94493726.0,
      "step": 54475
    },
    {
      "entropy": 5.904789400100708,
      "epoch": 4.238708422485898,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033138141352984145,
      "loss": 4.8819,
      "mean_token_accuracy": 0.22535137832164764,
      "num_tokens": 94502277.0,
      "step": 54480
    },
    {
      "entropy": 5.894386053085327,
      "epoch": 4.239097451857615,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033135458178464565,
      "loss": 4.9448,
      "mean_token_accuracy": 0.21768448650836944,
      "num_tokens": 94510714.0,
      "step": 54485
    },
    {
      "entropy": 5.922714900970459,
      "epoch": 4.239486481229333,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003313277491843587,
      "loss": 4.9664,
      "mean_token_accuracy": 0.21395401209592818,
      "num_tokens": 94518947.0,
      "step": 54490
    },
    {
      "entropy": 5.9754644393920895,
      "epoch": 4.239875510601051,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033130091572938786,
      "loss": 4.9926,
      "mean_token_accuracy": 0.21856670081615448,
      "num_tokens": 94527265.0,
      "step": 54495
    },
    {
      "entropy": 5.93578519821167,
      "epoch": 4.240264539972768,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003312740814201402,
      "loss": 4.9666,
      "mean_token_accuracy": 0.21449437290430068,
      "num_tokens": 94535047.0,
      "step": 54500
    },
    {
      "entropy": 5.864808464050293,
      "epoch": 4.240653569344485,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003312472462570229,
      "loss": 4.9069,
      "mean_token_accuracy": 0.21434569209814072,
      "num_tokens": 94543507.0,
      "step": 54505
    },
    {
      "entropy": 5.8731285572052006,
      "epoch": 4.241042598716203,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033122041024044325,
      "loss": 4.8835,
      "mean_token_accuracy": 0.21973054558038713,
      "num_tokens": 94552167.0,
      "step": 54510
    },
    {
      "entropy": 5.9693645477294925,
      "epoch": 4.241431628087921,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033119357337080827,
      "loss": 4.9741,
      "mean_token_accuracy": 0.21690181493759156,
      "num_tokens": 94560249.0,
      "step": 54515
    },
    {
      "entropy": 5.884009075164795,
      "epoch": 4.241820657459638,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003311667356485253,
      "loss": 4.89,
      "mean_token_accuracy": 0.21804728955030442,
      "num_tokens": 94568518.0,
      "step": 54520
    },
    {
      "entropy": 5.95898814201355,
      "epoch": 4.242209686831356,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003311398970740015,
      "loss": 5.0885,
      "mean_token_accuracy": 0.20031265318393707,
      "num_tokens": 94577334.0,
      "step": 54525
    },
    {
      "entropy": 5.971610069274902,
      "epoch": 4.242598716203073,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003311130576476441,
      "loss": 5.0402,
      "mean_token_accuracy": 0.20739947259426117,
      "num_tokens": 94585583.0,
      "step": 54530
    },
    {
      "entropy": 5.975155210494995,
      "epoch": 4.242987745574791,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003310862173698604,
      "loss": 4.9732,
      "mean_token_accuracy": 0.21989604234695434,
      "num_tokens": 94594492.0,
      "step": 54535
    },
    {
      "entropy": 5.920912456512451,
      "epoch": 4.243376774946508,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003310593762410576,
      "loss": 4.9532,
      "mean_token_accuracy": 0.2143271178007126,
      "num_tokens": 94603772.0,
      "step": 54540
    },
    {
      "entropy": 5.88807168006897,
      "epoch": 4.243765804318226,
      "grad_norm": 1.5625,
      "learning_rate": 0.000331032534261643,
      "loss": 4.9514,
      "mean_token_accuracy": 0.21967253386974334,
      "num_tokens": 94612049.0,
      "step": 54545
    },
    {
      "entropy": 6.009164524078369,
      "epoch": 4.244154833689944,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003310056914320239,
      "loss": 5.0416,
      "mean_token_accuracy": 0.20396026819944382,
      "num_tokens": 94620554.0,
      "step": 54550
    },
    {
      "entropy": 5.901394414901733,
      "epoch": 4.244543863061661,
      "grad_norm": 1.453125,
      "learning_rate": 0.00033097884775260754,
      "loss": 4.9114,
      "mean_token_accuracy": 0.21374585926532746,
      "num_tokens": 94628673.0,
      "step": 54555
    },
    {
      "entropy": 5.907778596878051,
      "epoch": 4.244932892433379,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033095200322380124,
      "loss": 5.0287,
      "mean_token_accuracy": 0.2163769021630287,
      "num_tokens": 94637804.0,
      "step": 54560
    },
    {
      "entropy": 5.905387735366821,
      "epoch": 4.245321921805096,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033092515784601245,
      "loss": 4.9826,
      "mean_token_accuracy": 0.21394221037626265,
      "num_tokens": 94647528.0,
      "step": 54565
    },
    {
      "entropy": 6.002347707748413,
      "epoch": 4.245710951176814,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003308983116196483,
      "loss": 4.965,
      "mean_token_accuracy": 0.2108680322766304,
      "num_tokens": 94656797.0,
      "step": 54570
    },
    {
      "entropy": 5.948218536376953,
      "epoch": 4.246099980548531,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003308714645451163,
      "loss": 5.0372,
      "mean_token_accuracy": 0.20686530470848083,
      "num_tokens": 94665462.0,
      "step": 54575
    },
    {
      "entropy": 5.952086353302002,
      "epoch": 4.246489009920249,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003308446166228237,
      "loss": 4.9612,
      "mean_token_accuracy": 0.21151547580957414,
      "num_tokens": 94673428.0,
      "step": 54580
    },
    {
      "entropy": 5.927017450332642,
      "epoch": 4.246878039291967,
      "grad_norm": 1.5,
      "learning_rate": 0.00033081776785317803,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21483672857284547,
      "num_tokens": 94682521.0,
      "step": 54585
    },
    {
      "entropy": 5.934924650192261,
      "epoch": 4.247267068663684,
      "grad_norm": 1.375,
      "learning_rate": 0.00033079091823658645,
      "loss": 4.9811,
      "mean_token_accuracy": 0.21247403919696808,
      "num_tokens": 94691317.0,
      "step": 54590
    },
    {
      "entropy": 5.938552713394165,
      "epoch": 4.247656098035401,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003307640677734566,
      "loss": 4.8734,
      "mean_token_accuracy": 0.2194083407521248,
      "num_tokens": 94700164.0,
      "step": 54595
    },
    {
      "entropy": 5.87406587600708,
      "epoch": 4.248045127407119,
      "grad_norm": 1.375,
      "learning_rate": 0.0003307372164641957,
      "loss": 4.87,
      "mean_token_accuracy": 0.2213505893945694,
      "num_tokens": 94708263.0,
      "step": 54600
    },
    {
      "entropy": 5.8182000637054445,
      "epoch": 4.248434156778837,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033071036430921133,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22415741682052612,
      "num_tokens": 94716953.0,
      "step": 54605
    },
    {
      "entropy": 5.929973316192627,
      "epoch": 4.248823186150554,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033068351130891076,
      "loss": 4.9312,
      "mean_token_accuracy": 0.21574445366859435,
      "num_tokens": 94725985.0,
      "step": 54610
    },
    {
      "entropy": 5.924100923538208,
      "epoch": 4.249212215522272,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003306566574637016,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21917377710342406,
      "num_tokens": 94733997.0,
      "step": 54615
    },
    {
      "entropy": 5.905703210830689,
      "epoch": 4.24960124489399,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033062980277399125,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21813781559467316,
      "num_tokens": 94742595.0,
      "step": 54620
    },
    {
      "entropy": 5.866001796722412,
      "epoch": 4.249990274265707,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003306029472401871,
      "loss": 4.8334,
      "mean_token_accuracy": 0.2281695544719696,
      "num_tokens": 94750150.0,
      "step": 54625
    },
    {
      "entropy": 5.910736179351806,
      "epoch": 4.250379303637424,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003305760908626968,
      "loss": 4.9709,
      "mean_token_accuracy": 0.21897337436676026,
      "num_tokens": 94758470.0,
      "step": 54630
    },
    {
      "entropy": 5.899111413955689,
      "epoch": 4.250768333009142,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003305492336419277,
      "loss": 5.0148,
      "mean_token_accuracy": 0.2048146277666092,
      "num_tokens": 94767372.0,
      "step": 54635
    },
    {
      "entropy": 5.870058393478393,
      "epoch": 4.25115736238086,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033052237557828763,
      "loss": 4.9417,
      "mean_token_accuracy": 0.21511054635047913,
      "num_tokens": 94775683.0,
      "step": 54640
    },
    {
      "entropy": 5.965483951568603,
      "epoch": 4.251546391752577,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003304955166721837,
      "loss": 4.9718,
      "mean_token_accuracy": 0.20918405950069427,
      "num_tokens": 94784076.0,
      "step": 54645
    },
    {
      "entropy": 5.913328647613525,
      "epoch": 4.251935421124295,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003304686569240237,
      "loss": 4.9442,
      "mean_token_accuracy": 0.213658644258976,
      "num_tokens": 94793512.0,
      "step": 54650
    },
    {
      "entropy": 5.913216924667358,
      "epoch": 4.252324450496013,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000330441796334215,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21754608154296876,
      "num_tokens": 94801919.0,
      "step": 54655
    },
    {
      "entropy": 5.977505207061768,
      "epoch": 4.25271347986773,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003304149349031653,
      "loss": 5.0169,
      "mean_token_accuracy": 0.20953661054372788,
      "num_tokens": 94811104.0,
      "step": 54660
    },
    {
      "entropy": 5.84434404373169,
      "epoch": 4.253102509239447,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003303880726312822,
      "loss": 4.9039,
      "mean_token_accuracy": 0.22365008890628815,
      "num_tokens": 94819836.0,
      "step": 54665
    },
    {
      "entropy": 5.971727991104126,
      "epoch": 4.253491538611165,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033036120951897334,
      "loss": 5.0316,
      "mean_token_accuracy": 0.20984099060297012,
      "num_tokens": 94829023.0,
      "step": 54670
    },
    {
      "entropy": 6.0348070621490475,
      "epoch": 4.253880567982883,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003303343455666462,
      "loss": 5.0372,
      "mean_token_accuracy": 0.20979625582695008,
      "num_tokens": 94837781.0,
      "step": 54675
    },
    {
      "entropy": 5.960360813140869,
      "epoch": 4.2542695973546,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00033030748077470846,
      "loss": 5.0084,
      "mean_token_accuracy": 0.22425804585218428,
      "num_tokens": 94846772.0,
      "step": 54680
    },
    {
      "entropy": 5.955118942260742,
      "epoch": 4.254658626726318,
      "grad_norm": 1.734375,
      "learning_rate": 0.00033028061514356766,
      "loss": 5.0298,
      "mean_token_accuracy": 0.21190354973077774,
      "num_tokens": 94855114.0,
      "step": 54685
    },
    {
      "entropy": 5.993079948425293,
      "epoch": 4.255047656098036,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003302537486736315,
      "loss": 5.0299,
      "mean_token_accuracy": 0.2111533299088478,
      "num_tokens": 94864144.0,
      "step": 54690
    },
    {
      "entropy": 5.940740346908569,
      "epoch": 4.255436685469753,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003302268813653077,
      "loss": 4.967,
      "mean_token_accuracy": 0.21686182916164398,
      "num_tokens": 94872497.0,
      "step": 54695
    },
    {
      "entropy": 5.950805282592773,
      "epoch": 4.25582571484147,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003302000132190039,
      "loss": 5.0033,
      "mean_token_accuracy": 0.22027105689048768,
      "num_tokens": 94881326.0,
      "step": 54700
    },
    {
      "entropy": 5.920584011077881,
      "epoch": 4.256214744213188,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003301731442351277,
      "loss": 4.9866,
      "mean_token_accuracy": 0.2093284621834755,
      "num_tokens": 94889887.0,
      "step": 54705
    },
    {
      "entropy": 5.88416600227356,
      "epoch": 4.256603773584906,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003301462744140869,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22125812768936157,
      "num_tokens": 94899210.0,
      "step": 54710
    },
    {
      "entropy": 5.9782342433929445,
      "epoch": 4.256992802956623,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003301194037562892,
      "loss": 5.0414,
      "mean_token_accuracy": 0.21191616654396056,
      "num_tokens": 94907303.0,
      "step": 54715
    },
    {
      "entropy": 5.879955768585205,
      "epoch": 4.257381832328341,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003300925322621422,
      "loss": 4.867,
      "mean_token_accuracy": 0.22361826449632644,
      "num_tokens": 94916069.0,
      "step": 54720
    },
    {
      "entropy": 5.844481182098389,
      "epoch": 4.257770861700059,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003300656599320537,
      "loss": 4.9301,
      "mean_token_accuracy": 0.2186798185110092,
      "num_tokens": 94926209.0,
      "step": 54725
    },
    {
      "entropy": 5.923324155807495,
      "epoch": 4.2581598910717755,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003300387867664315,
      "loss": 4.9651,
      "mean_token_accuracy": 0.21191472113132476,
      "num_tokens": 94934671.0,
      "step": 54730
    },
    {
      "entropy": 5.944418287277221,
      "epoch": 4.258548920443493,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003300119127656833,
      "loss": 5.0265,
      "mean_token_accuracy": 0.21447554975748062,
      "num_tokens": 94942992.0,
      "step": 54735
    },
    {
      "entropy": 5.873823451995849,
      "epoch": 4.258937949815211,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032998503793021694,
      "loss": 4.9248,
      "mean_token_accuracy": 0.2198515087366104,
      "num_tokens": 94951356.0,
      "step": 54740
    },
    {
      "entropy": 5.918998289108276,
      "epoch": 4.259326979186929,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003299581622604401,
      "loss": 4.9325,
      "mean_token_accuracy": 0.21402983069419862,
      "num_tokens": 94960412.0,
      "step": 54745
    },
    {
      "entropy": 5.896385574340821,
      "epoch": 4.259716008558646,
      "grad_norm": 1.484375,
      "learning_rate": 0.00032993128575676064,
      "loss": 4.9176,
      "mean_token_accuracy": 0.22058046609163284,
      "num_tokens": 94969005.0,
      "step": 54750
    },
    {
      "entropy": 5.931352996826172,
      "epoch": 4.260105037930364,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00032990440841958635,
      "loss": 5.0708,
      "mean_token_accuracy": 0.2082776367664337,
      "num_tokens": 94977142.0,
      "step": 54755
    },
    {
      "entropy": 5.9317127704620365,
      "epoch": 4.260494067302082,
      "grad_norm": 1.34375,
      "learning_rate": 0.000329877530249325,
      "loss": 4.9602,
      "mean_token_accuracy": 0.20847708582878113,
      "num_tokens": 94985936.0,
      "step": 54760
    },
    {
      "entropy": 5.927503395080566,
      "epoch": 4.2608830966737985,
      "grad_norm": 1.375,
      "learning_rate": 0.0003298506512463845,
      "loss": 4.9757,
      "mean_token_accuracy": 0.21178323924541473,
      "num_tokens": 94994187.0,
      "step": 54765
    },
    {
      "entropy": 5.913310241699219,
      "epoch": 4.261272126045516,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003298237714111727,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2158214658498764,
      "num_tokens": 95003921.0,
      "step": 54770
    },
    {
      "entropy": 5.93520975112915,
      "epoch": 4.261661155417234,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003297968907440974,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21660510152578355,
      "num_tokens": 95012752.0,
      "step": 54775
    },
    {
      "entropy": 5.901351070404052,
      "epoch": 4.262050184788952,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003297700092455665,
      "loss": 4.9648,
      "mean_token_accuracy": 0.21474601626396178,
      "num_tokens": 95021910.0,
      "step": 54780
    },
    {
      "entropy": 5.8472565650939945,
      "epoch": 4.262439214160669,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032974312691598794,
      "loss": 4.8447,
      "mean_token_accuracy": 0.22162685096263884,
      "num_tokens": 95030261.0,
      "step": 54785
    },
    {
      "entropy": 5.902706050872803,
      "epoch": 4.262828243532387,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032971624375576956,
      "loss": 4.8857,
      "mean_token_accuracy": 0.221635602414608,
      "num_tokens": 95038906.0,
      "step": 54790
    },
    {
      "entropy": 5.899225378036499,
      "epoch": 4.263217272904104,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003296893597653192,
      "loss": 4.9514,
      "mean_token_accuracy": 0.21382581442594528,
      "num_tokens": 95047690.0,
      "step": 54795
    },
    {
      "entropy": 5.9156729221344,
      "epoch": 4.2636063022758215,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003296624749450448,
      "loss": 4.9925,
      "mean_token_accuracy": 0.2123454675078392,
      "num_tokens": 95056294.0,
      "step": 54800
    },
    {
      "entropy": 5.94627366065979,
      "epoch": 4.263995331647539,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003296355892953545,
      "loss": 4.9712,
      "mean_token_accuracy": 0.2076500102877617,
      "num_tokens": 95065756.0,
      "step": 54805
    },
    {
      "entropy": 5.891101837158203,
      "epoch": 4.264384361019257,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000329608702816656,
      "loss": 4.9826,
      "mean_token_accuracy": 0.2092406392097473,
      "num_tokens": 95074459.0,
      "step": 54810
    },
    {
      "entropy": 5.931795120239258,
      "epoch": 4.264773390390975,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003295818155093574,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21877542436122893,
      "num_tokens": 95083163.0,
      "step": 54815
    },
    {
      "entropy": 5.909637928009033,
      "epoch": 4.265162419762692,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032955492737386663,
      "loss": 4.9228,
      "mean_token_accuracy": 0.22160348892211915,
      "num_tokens": 95092324.0,
      "step": 54820
    },
    {
      "entropy": 5.91535792350769,
      "epoch": 4.26555144913441,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003295280384105916,
      "loss": 4.8564,
      "mean_token_accuracy": 0.23067598789930344,
      "num_tokens": 95100250.0,
      "step": 54825
    },
    {
      "entropy": 5.873905181884766,
      "epoch": 4.265940478506128,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032950114861994047,
      "loss": 4.898,
      "mean_token_accuracy": 0.21934334188699722,
      "num_tokens": 95108641.0,
      "step": 54830
    },
    {
      "entropy": 5.920182943344116,
      "epoch": 4.2663295078778445,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00032947425800232107,
      "loss": 4.9516,
      "mean_token_accuracy": 0.21476654410362245,
      "num_tokens": 95116547.0,
      "step": 54835
    },
    {
      "entropy": 5.882655668258667,
      "epoch": 4.266718537249562,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003294473665581415,
      "loss": 4.8854,
      "mean_token_accuracy": 0.21966099441051484,
      "num_tokens": 95125334.0,
      "step": 54840
    },
    {
      "entropy": 5.919733715057373,
      "epoch": 4.26710756662128,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032942047428780985,
      "loss": 4.9511,
      "mean_token_accuracy": 0.20866973847150802,
      "num_tokens": 95134805.0,
      "step": 54845
    },
    {
      "entropy": 5.925727224349975,
      "epoch": 4.267496595992998,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003293935811917342,
      "loss": 4.9932,
      "mean_token_accuracy": 0.21327505111694336,
      "num_tokens": 95143539.0,
      "step": 54850
    },
    {
      "entropy": 5.876774311065674,
      "epoch": 4.267885625364715,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003293666872703224,
      "loss": 4.8892,
      "mean_token_accuracy": 0.2277057334780693,
      "num_tokens": 95152394.0,
      "step": 54855
    },
    {
      "entropy": 5.862449312210083,
      "epoch": 4.268274654736433,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003293397925239827,
      "loss": 4.9124,
      "mean_token_accuracy": 0.22262572050094603,
      "num_tokens": 95160666.0,
      "step": 54860
    },
    {
      "entropy": 5.9539327144622805,
      "epoch": 4.26866368410815,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003293128969531231,
      "loss": 5.0267,
      "mean_token_accuracy": 0.20479939579963685,
      "num_tokens": 95170007.0,
      "step": 54865
    },
    {
      "entropy": 6.0117440700531,
      "epoch": 4.2690527134798675,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003292860005581518,
      "loss": 5.0529,
      "mean_token_accuracy": 0.20532880127429962,
      "num_tokens": 95178169.0,
      "step": 54870
    },
    {
      "entropy": 5.947081136703491,
      "epoch": 4.269441742851585,
      "grad_norm": 1.5,
      "learning_rate": 0.00032925910333947694,
      "loss": 4.9804,
      "mean_token_accuracy": 0.21240086406469344,
      "num_tokens": 95186357.0,
      "step": 54875
    },
    {
      "entropy": 5.91890139579773,
      "epoch": 4.269830772223303,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003292322052975064,
      "loss": 4.9698,
      "mean_token_accuracy": 0.21924578100442887,
      "num_tokens": 95195678.0,
      "step": 54880
    },
    {
      "entropy": 5.942370223999023,
      "epoch": 4.270219801595021,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003292053064326485,
      "loss": 4.9648,
      "mean_token_accuracy": 0.20888793617486953,
      "num_tokens": 95204094.0,
      "step": 54885
    },
    {
      "entropy": 5.861639070510864,
      "epoch": 4.270608830966738,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003291784067453114,
      "loss": 4.9227,
      "mean_token_accuracy": 0.2161780431866646,
      "num_tokens": 95211976.0,
      "step": 54890
    },
    {
      "entropy": 5.864830160140992,
      "epoch": 4.270997860338456,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032915150623590304,
      "loss": 4.8382,
      "mean_token_accuracy": 0.22360936254262925,
      "num_tokens": 95220207.0,
      "step": 54895
    },
    {
      "entropy": 5.844212341308594,
      "epoch": 4.271386889710173,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032912460490483193,
      "loss": 4.8248,
      "mean_token_accuracy": 0.22407771944999694,
      "num_tokens": 95228226.0,
      "step": 54900
    },
    {
      "entropy": 5.839849042892456,
      "epoch": 4.2717759190818905,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003290977027525061,
      "loss": 4.8173,
      "mean_token_accuracy": 0.22076697647571564,
      "num_tokens": 95237159.0,
      "step": 54905
    },
    {
      "entropy": 5.833155775070191,
      "epoch": 4.272164948453608,
      "grad_norm": 1.46875,
      "learning_rate": 0.00032907079977933373,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2169177070260048,
      "num_tokens": 95245684.0,
      "step": 54910
    },
    {
      "entropy": 5.943340063095093,
      "epoch": 4.272553977825326,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032904389598572295,
      "loss": 5.0574,
      "mean_token_accuracy": 0.2081340253353119,
      "num_tokens": 95254097.0,
      "step": 54915
    },
    {
      "entropy": 5.964110374450684,
      "epoch": 4.272943007197044,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032901699137208215,
      "loss": 5.0401,
      "mean_token_accuracy": 0.20288221687078475,
      "num_tokens": 95263216.0,
      "step": 54920
    },
    {
      "entropy": 5.959118127822876,
      "epoch": 4.273332036568761,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003289900859388194,
      "loss": 4.987,
      "mean_token_accuracy": 0.2160157635807991,
      "num_tokens": 95272084.0,
      "step": 54925
    },
    {
      "entropy": 5.918903398513794,
      "epoch": 4.273721065940478,
      "grad_norm": 1.375,
      "learning_rate": 0.00032896317968634305,
      "loss": 4.931,
      "mean_token_accuracy": 0.2152649313211441,
      "num_tokens": 95280505.0,
      "step": 54930
    },
    {
      "entropy": 5.907107353210449,
      "epoch": 4.274110095312196,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003289362726150614,
      "loss": 4.8971,
      "mean_token_accuracy": 0.22444306910037995,
      "num_tokens": 95290037.0,
      "step": 54935
    },
    {
      "entropy": 5.968106031417847,
      "epoch": 4.2744991246839135,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032890936472538255,
      "loss": 4.9628,
      "mean_token_accuracy": 0.21619293093681335,
      "num_tokens": 95299370.0,
      "step": 54940
    },
    {
      "entropy": 5.9368799209594725,
      "epoch": 4.274888154055631,
      "grad_norm": 1.421875,
      "learning_rate": 0.000328882456017715,
      "loss": 4.8758,
      "mean_token_accuracy": 0.21603064984083176,
      "num_tokens": 95307349.0,
      "step": 54945
    },
    {
      "entropy": 5.866839027404785,
      "epoch": 4.275277183427349,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00032885554649246697,
      "loss": 4.9717,
      "mean_token_accuracy": 0.21191533505916596,
      "num_tokens": 95316084.0,
      "step": 54950
    },
    {
      "entropy": 5.9466125011444095,
      "epoch": 4.275666212799067,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0003288286361500467,
      "loss": 4.9481,
      "mean_token_accuracy": 0.2209256410598755,
      "num_tokens": 95323907.0,
      "step": 54955
    },
    {
      "entropy": 5.883621311187744,
      "epoch": 4.276055242170784,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032880172499086247,
      "loss": 4.9507,
      "mean_token_accuracy": 0.21169123649597169,
      "num_tokens": 95332850.0,
      "step": 54960
    },
    {
      "entropy": 5.901259708404541,
      "epoch": 4.276444271542501,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003287748130153228,
      "loss": 4.9802,
      "mean_token_accuracy": 0.21438556015491486,
      "num_tokens": 95341944.0,
      "step": 54965
    },
    {
      "entropy": 5.93082332611084,
      "epoch": 4.276833300914219,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003287479002238359,
      "loss": 5.0266,
      "mean_token_accuracy": 0.20667191445827485,
      "num_tokens": 95350224.0,
      "step": 54970
    },
    {
      "entropy": 5.9090526580810545,
      "epoch": 4.2772223302859365,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003287209866168102,
      "loss": 4.8844,
      "mean_token_accuracy": 0.21957225054502488,
      "num_tokens": 95359860.0,
      "step": 54975
    },
    {
      "entropy": 5.940339374542236,
      "epoch": 4.277611359657654,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000328694072194654,
      "loss": 4.9739,
      "mean_token_accuracy": 0.21604870408773422,
      "num_tokens": 95368734.0,
      "step": 54980
    },
    {
      "entropy": 5.976496410369873,
      "epoch": 4.278000389029372,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032866715695777565,
      "loss": 5.0442,
      "mean_token_accuracy": 0.2045149877667427,
      "num_tokens": 95378159.0,
      "step": 54985
    },
    {
      "entropy": 5.977135705947876,
      "epoch": 4.2783894184010896,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003286402409065837,
      "loss": 5.0384,
      "mean_token_accuracy": 0.20458727926015854,
      "num_tokens": 95386385.0,
      "step": 54990
    },
    {
      "entropy": 5.904051446914673,
      "epoch": 4.278778447772807,
      "grad_norm": 1.375,
      "learning_rate": 0.00032861332404148647,
      "loss": 4.963,
      "mean_token_accuracy": 0.21119810342788697,
      "num_tokens": 95395454.0,
      "step": 54995
    },
    {
      "entropy": 5.969295024871826,
      "epoch": 4.279167477144524,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032858640636289235,
      "loss": 5.0497,
      "mean_token_accuracy": 0.20861918926239015,
      "num_tokens": 95403832.0,
      "step": 55000
    },
    {
      "entropy": 5.901451444625854,
      "epoch": 4.279556506516242,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003285594878712098,
      "loss": 4.8967,
      "mean_token_accuracy": 0.22022674530744552,
      "num_tokens": 95413023.0,
      "step": 55005
    },
    {
      "entropy": 5.93023910522461,
      "epoch": 4.2799455358879595,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003285325685668473,
      "loss": 4.9841,
      "mean_token_accuracy": 0.21652329713106155,
      "num_tokens": 95421891.0,
      "step": 55010
    },
    {
      "entropy": 5.891927671432495,
      "epoch": 4.280334565259677,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032850564845021327,
      "loss": 5.0133,
      "mean_token_accuracy": 0.21354079842567444,
      "num_tokens": 95430515.0,
      "step": 55015
    },
    {
      "entropy": 5.915850734710693,
      "epoch": 4.280723594631395,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032847872752171615,
      "loss": 4.9071,
      "mean_token_accuracy": 0.21579401642084123,
      "num_tokens": 95440327.0,
      "step": 55020
    },
    {
      "entropy": 5.94127049446106,
      "epoch": 4.2811126240031125,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032845180578176456,
      "loss": 4.8901,
      "mean_token_accuracy": 0.21167974621057511,
      "num_tokens": 95448327.0,
      "step": 55025
    },
    {
      "entropy": 5.928249168395996,
      "epoch": 4.28150165337483,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032842488323076675,
      "loss": 4.9823,
      "mean_token_accuracy": 0.21034881025552749,
      "num_tokens": 95457994.0,
      "step": 55030
    },
    {
      "entropy": 5.882372188568115,
      "epoch": 4.281890682746547,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032839795986913146,
      "loss": 4.9265,
      "mean_token_accuracy": 0.21986666023731233,
      "num_tokens": 95466398.0,
      "step": 55035
    },
    {
      "entropy": 5.90666880607605,
      "epoch": 4.282279712118265,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00032837103569726713,
      "loss": 4.9795,
      "mean_token_accuracy": 0.2102600798010826,
      "num_tokens": 95475378.0,
      "step": 55040
    },
    {
      "entropy": 5.947157144546509,
      "epoch": 4.2826687414899824,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003283441107155822,
      "loss": 5.0386,
      "mean_token_accuracy": 0.2063481628894806,
      "num_tokens": 95484210.0,
      "step": 55045
    },
    {
      "entropy": 6.000141763687134,
      "epoch": 4.2830577708617,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003283171849244854,
      "loss": 4.9734,
      "mean_token_accuracy": 0.21504384875297547,
      "num_tokens": 95492572.0,
      "step": 55050
    },
    {
      "entropy": 5.921865510940552,
      "epoch": 4.283446800233418,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032829025832438507,
      "loss": 4.9303,
      "mean_token_accuracy": 0.22068173289299012,
      "num_tokens": 95501547.0,
      "step": 55055
    },
    {
      "entropy": 5.877391004562378,
      "epoch": 4.2838358296051355,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032826333091569,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21969527751207352,
      "num_tokens": 95509550.0,
      "step": 55060
    },
    {
      "entropy": 5.8734293460845945,
      "epoch": 4.284224858976852,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003282364026988087,
      "loss": 4.9653,
      "mean_token_accuracy": 0.21081321239471434,
      "num_tokens": 95518191.0,
      "step": 55065
    },
    {
      "entropy": 5.88088960647583,
      "epoch": 4.28461388834857,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032820947367414953,
      "loss": 4.9143,
      "mean_token_accuracy": 0.22237202376127244,
      "num_tokens": 95526187.0,
      "step": 55070
    },
    {
      "entropy": 5.764674520492553,
      "epoch": 4.285002917720288,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003281825438421215,
      "loss": 4.8408,
      "mean_token_accuracy": 0.22053172141313554,
      "num_tokens": 95534850.0,
      "step": 55075
    },
    {
      "entropy": 5.88139853477478,
      "epoch": 4.285391947092005,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00032815561320313286,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21655327677726746,
      "num_tokens": 95542916.0,
      "step": 55080
    },
    {
      "entropy": 5.9635316848754885,
      "epoch": 4.285780976463723,
      "grad_norm": 1.578125,
      "learning_rate": 0.00032812868175759245,
      "loss": 5.0245,
      "mean_token_accuracy": 0.21074368357658385,
      "num_tokens": 95551009.0,
      "step": 55085
    },
    {
      "entropy": 5.934903335571289,
      "epoch": 4.286170005835441,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003281017495059088,
      "loss": 5.0121,
      "mean_token_accuracy": 0.22104764878749847,
      "num_tokens": 95560964.0,
      "step": 55090
    },
    {
      "entropy": 5.917796945571899,
      "epoch": 4.2865590352071585,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003280748164484907,
      "loss": 4.8798,
      "mean_token_accuracy": 0.21807412952184677,
      "num_tokens": 95569902.0,
      "step": 55095
    },
    {
      "entropy": 5.965263080596924,
      "epoch": 4.286948064578875,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003280478825857468,
      "loss": 4.9431,
      "mean_token_accuracy": 0.21638462245464324,
      "num_tokens": 95578759.0,
      "step": 55100
    },
    {
      "entropy": 5.983382892608643,
      "epoch": 4.287337093950593,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032802094791808565,
      "loss": 5.0653,
      "mean_token_accuracy": 0.20726783722639083,
      "num_tokens": 95587288.0,
      "step": 55105
    },
    {
      "entropy": 5.924701261520386,
      "epoch": 4.287726123322311,
      "grad_norm": 1.625,
      "learning_rate": 0.000327994012445916,
      "loss": 4.9178,
      "mean_token_accuracy": 0.2199135586619377,
      "num_tokens": 95595698.0,
      "step": 55110
    },
    {
      "entropy": 5.878912782669067,
      "epoch": 4.288115152694028,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003279670761696465,
      "loss": 4.9169,
      "mean_token_accuracy": 0.21757130026817323,
      "num_tokens": 95603851.0,
      "step": 55115
    },
    {
      "entropy": 5.904960870742798,
      "epoch": 4.288504182065746,
      "grad_norm": 1.375,
      "learning_rate": 0.00032794013908968593,
      "loss": 4.977,
      "mean_token_accuracy": 0.20951825976371766,
      "num_tokens": 95612466.0,
      "step": 55120
    },
    {
      "entropy": 5.9170814037323,
      "epoch": 4.288893211437464,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003279132012064431,
      "loss": 4.9808,
      "mean_token_accuracy": 0.20918555110692977,
      "num_tokens": 95621194.0,
      "step": 55125
    },
    {
      "entropy": 5.933392333984375,
      "epoch": 4.289282240809181,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003278862625203266,
      "loss": 4.9852,
      "mean_token_accuracy": 0.2184272974729538,
      "num_tokens": 95630207.0,
      "step": 55130
    },
    {
      "entropy": 5.914591550827026,
      "epoch": 4.289671270180898,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00032785932303174524,
      "loss": 4.9211,
      "mean_token_accuracy": 0.21909142136573792,
      "num_tokens": 95638664.0,
      "step": 55135
    },
    {
      "entropy": 5.925998592376709,
      "epoch": 4.290060299552616,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032783238274110787,
      "loss": 4.962,
      "mean_token_accuracy": 0.21837485879659652,
      "num_tokens": 95647442.0,
      "step": 55140
    },
    {
      "entropy": 5.935156583786011,
      "epoch": 4.290449328924334,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00032780544164882317,
      "loss": 4.9955,
      "mean_token_accuracy": 0.2243955463171005,
      "num_tokens": 95655641.0,
      "step": 55145
    },
    {
      "entropy": 5.963480472564697,
      "epoch": 4.290838358296051,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003277784997552999,
      "loss": 4.9457,
      "mean_token_accuracy": 0.21119096577167512,
      "num_tokens": 95664107.0,
      "step": 55150
    },
    {
      "entropy": 5.9393861293792725,
      "epoch": 4.291227387667769,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032775155706094684,
      "loss": 4.9726,
      "mean_token_accuracy": 0.2143098384141922,
      "num_tokens": 95672936.0,
      "step": 55155
    },
    {
      "entropy": 5.855972814559936,
      "epoch": 4.291616417039487,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003277246135661729,
      "loss": 4.8567,
      "mean_token_accuracy": 0.2212250202894211,
      "num_tokens": 95681671.0,
      "step": 55160
    },
    {
      "entropy": 5.859768295288086,
      "epoch": 4.2920054464112045,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003276976692713869,
      "loss": 4.9042,
      "mean_token_accuracy": 0.21740612238645554,
      "num_tokens": 95690748.0,
      "step": 55165
    },
    {
      "entropy": 5.97273473739624,
      "epoch": 4.292394475782921,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003276707241769976,
      "loss": 5.0355,
      "mean_token_accuracy": 0.2081717312335968,
      "num_tokens": 95698948.0,
      "step": 55170
    },
    {
      "entropy": 5.952265453338623,
      "epoch": 4.292783505154639,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000327643778283414,
      "loss": 5.1128,
      "mean_token_accuracy": 0.20296162366867065,
      "num_tokens": 95707868.0,
      "step": 55175
    },
    {
      "entropy": 5.827863931655884,
      "epoch": 4.293172534526357,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032761683159104475,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22736350893974305,
      "num_tokens": 95716567.0,
      "step": 55180
    },
    {
      "entropy": 5.92322096824646,
      "epoch": 4.293561563898074,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003275898841002989,
      "loss": 4.9354,
      "mean_token_accuracy": 0.22188310325145721,
      "num_tokens": 95725435.0,
      "step": 55185
    },
    {
      "entropy": 5.966817188262939,
      "epoch": 4.293950593269792,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032756293581158517,
      "loss": 5.0767,
      "mean_token_accuracy": 0.2077276587486267,
      "num_tokens": 95734769.0,
      "step": 55190
    },
    {
      "entropy": 5.925601291656494,
      "epoch": 4.29433962264151,
      "grad_norm": 1.5,
      "learning_rate": 0.0003275359867253127,
      "loss": 4.969,
      "mean_token_accuracy": 0.2204594522714615,
      "num_tokens": 95742726.0,
      "step": 55195
    },
    {
      "entropy": 5.905518293380737,
      "epoch": 4.294728652013227,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003275090368418901,
      "loss": 4.9067,
      "mean_token_accuracy": 0.2223271682858467,
      "num_tokens": 95751086.0,
      "step": 55200
    },
    {
      "entropy": 5.92593903541565,
      "epoch": 4.295117681384944,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003274820861617265,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21023458391427993,
      "num_tokens": 95759420.0,
      "step": 55205
    },
    {
      "entropy": 5.934212684631348,
      "epoch": 4.295506710756662,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003274551346852308,
      "loss": 5.0368,
      "mean_token_accuracy": 0.21251590698957443,
      "num_tokens": 95768011.0,
      "step": 55210
    },
    {
      "entropy": 5.905106735229492,
      "epoch": 4.29589574012838,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003274281824128119,
      "loss": 4.9275,
      "mean_token_accuracy": 0.22029270827770234,
      "num_tokens": 95777138.0,
      "step": 55215
    },
    {
      "entropy": 5.8655658721923825,
      "epoch": 4.296284769500097,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032740122934487876,
      "loss": 4.8765,
      "mean_token_accuracy": 0.216781285405159,
      "num_tokens": 95785881.0,
      "step": 55220
    },
    {
      "entropy": 5.954648828506469,
      "epoch": 4.296673798871815,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003273742754818405,
      "loss": 4.9376,
      "mean_token_accuracy": 0.21818303763866426,
      "num_tokens": 95794533.0,
      "step": 55225
    },
    {
      "entropy": 5.891070938110351,
      "epoch": 4.297062828243533,
      "grad_norm": 1.25,
      "learning_rate": 0.0003273473208241058,
      "loss": 4.9834,
      "mean_token_accuracy": 0.20637192726135253,
      "num_tokens": 95804135.0,
      "step": 55230
    },
    {
      "entropy": 5.848950386047363,
      "epoch": 4.29745185761525,
      "grad_norm": 1.53125,
      "learning_rate": 0.00032732036537208386,
      "loss": 4.8927,
      "mean_token_accuracy": 0.21908506900072097,
      "num_tokens": 95813286.0,
      "step": 55235
    },
    {
      "entropy": 5.980234670639038,
      "epoch": 4.297840886986967,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003272934091261837,
      "loss": 5.0332,
      "mean_token_accuracy": 0.2116427019238472,
      "num_tokens": 95822234.0,
      "step": 55240
    },
    {
      "entropy": 5.84839358329773,
      "epoch": 4.298229916358685,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032726645208681427,
      "loss": 4.8867,
      "mean_token_accuracy": 0.21820137053728103,
      "num_tokens": 95830667.0,
      "step": 55245
    },
    {
      "entropy": 5.90915150642395,
      "epoch": 4.298618945730403,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032723949425438465,
      "loss": 5.0207,
      "mean_token_accuracy": 0.21370573490858077,
      "num_tokens": 95840347.0,
      "step": 55250
    },
    {
      "entropy": 5.951452159881592,
      "epoch": 4.29900797510212,
      "grad_norm": 1.578125,
      "learning_rate": 0.00032721253562930383,
      "loss": 5.0892,
      "mean_token_accuracy": 0.20923541933298112,
      "num_tokens": 95848237.0,
      "step": 55255
    },
    {
      "entropy": 5.987345218658447,
      "epoch": 4.299397004473838,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032718557621198087,
      "loss": 5.0981,
      "mean_token_accuracy": 0.2092142090201378,
      "num_tokens": 95857262.0,
      "step": 55260
    },
    {
      "entropy": 5.963543510437011,
      "epoch": 4.299786033845555,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003271586160028249,
      "loss": 5.0307,
      "mean_token_accuracy": 0.20898465812206268,
      "num_tokens": 95865807.0,
      "step": 55265
    },
    {
      "entropy": 5.906054162979126,
      "epoch": 4.300175063217273,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003271316550022449,
      "loss": 4.9546,
      "mean_token_accuracy": 0.21874375343322755,
      "num_tokens": 95874734.0,
      "step": 55270
    },
    {
      "entropy": 6.010351896286011,
      "epoch": 4.30056409258899,
      "grad_norm": 1.46875,
      "learning_rate": 0.00032710469321065,
      "loss": 4.9662,
      "mean_token_accuracy": 0.21215960532426834,
      "num_tokens": 95883328.0,
      "step": 55275
    },
    {
      "entropy": 5.866271924972534,
      "epoch": 4.300953121960708,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003270777306284493,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21694801449775697,
      "num_tokens": 95892152.0,
      "step": 55280
    },
    {
      "entropy": 5.947036552429199,
      "epoch": 4.301342151332426,
      "grad_norm": 1.484375,
      "learning_rate": 0.000327050767256052,
      "loss": 4.9657,
      "mean_token_accuracy": 0.21250649392604828,
      "num_tokens": 95900898.0,
      "step": 55285
    },
    {
      "entropy": 5.904823255538941,
      "epoch": 4.301731180704143,
      "grad_norm": 1.5625,
      "learning_rate": 0.00032702380309386714,
      "loss": 4.9241,
      "mean_token_accuracy": 0.21585486680269242,
      "num_tokens": 95909588.0,
      "step": 55290
    },
    {
      "entropy": 5.8983745098114015,
      "epoch": 4.302120210075861,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032699683814230385,
      "loss": 4.9196,
      "mean_token_accuracy": 0.21903885304927825,
      "num_tokens": 95918163.0,
      "step": 55295
    },
    {
      "entropy": 5.931772613525391,
      "epoch": 4.302509239447578,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003269698724017713,
      "loss": 5.003,
      "mean_token_accuracy": 0.2155136436223984,
      "num_tokens": 95926073.0,
      "step": 55300
    },
    {
      "entropy": 5.8874587059021,
      "epoch": 4.302898268819296,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032694290587267867,
      "loss": 4.9636,
      "mean_token_accuracy": 0.21912557631731033,
      "num_tokens": 95934136.0,
      "step": 55305
    },
    {
      "entropy": 5.875033712387085,
      "epoch": 4.303287298191013,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003269159385554352,
      "loss": 4.8911,
      "mean_token_accuracy": 0.22130660563707352,
      "num_tokens": 95942583.0,
      "step": 55310
    },
    {
      "entropy": 6.043048906326294,
      "epoch": 4.303676327562731,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003268889704504499,
      "loss": 5.0559,
      "mean_token_accuracy": 0.20804841965436935,
      "num_tokens": 95951267.0,
      "step": 55315
    },
    {
      "entropy": 5.981771087646484,
      "epoch": 4.304065356934449,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032686200155813196,
      "loss": 5.0913,
      "mean_token_accuracy": 0.20113102793693544,
      "num_tokens": 95959830.0,
      "step": 55320
    },
    {
      "entropy": 5.934607887268067,
      "epoch": 4.304454386306166,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003268350318788908,
      "loss": 5.013,
      "mean_token_accuracy": 0.21188382506370546,
      "num_tokens": 95967867.0,
      "step": 55325
    },
    {
      "entropy": 5.989916849136352,
      "epoch": 4.304843415677884,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003268080614131356,
      "loss": 5.1097,
      "mean_token_accuracy": 0.20354807525873184,
      "num_tokens": 95976521.0,
      "step": 55330
    },
    {
      "entropy": 5.9397862434387205,
      "epoch": 4.305232445049601,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003267810901612754,
      "loss": 4.9717,
      "mean_token_accuracy": 0.21563531458377838,
      "num_tokens": 95984640.0,
      "step": 55335
    },
    {
      "entropy": 5.872354888916016,
      "epoch": 4.305621474421319,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032675411812371975,
      "loss": 4.9431,
      "mean_token_accuracy": 0.21830699145793914,
      "num_tokens": 95992894.0,
      "step": 55340
    },
    {
      "entropy": 5.85751371383667,
      "epoch": 4.306010503793036,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00032672714530087763,
      "loss": 4.929,
      "mean_token_accuracy": 0.21145529448986053,
      "num_tokens": 96002084.0,
      "step": 55345
    },
    {
      "entropy": 5.922606086730957,
      "epoch": 4.306399533164754,
      "grad_norm": 1.25,
      "learning_rate": 0.00032670017169315836,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21180877089500427,
      "num_tokens": 96010864.0,
      "step": 55350
    },
    {
      "entropy": 5.937146091461182,
      "epoch": 4.306788562536472,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003266731973009713,
      "loss": 4.9629,
      "mean_token_accuracy": 0.2194577142596245,
      "num_tokens": 96020166.0,
      "step": 55355
    },
    {
      "entropy": 5.907773065567016,
      "epoch": 4.307177591908189,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003266462221247257,
      "loss": 4.9153,
      "mean_token_accuracy": 0.2150722861289978,
      "num_tokens": 96028329.0,
      "step": 55360
    },
    {
      "entropy": 5.976131439208984,
      "epoch": 4.307566621279907,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000326619246164831,
      "loss": 4.9991,
      "mean_token_accuracy": 0.2067098170518875,
      "num_tokens": 96036762.0,
      "step": 55365
    },
    {
      "entropy": 5.897979640960694,
      "epoch": 4.307955650651624,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003265922694216963,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23048423379659652,
      "num_tokens": 96044192.0,
      "step": 55370
    },
    {
      "entropy": 5.904801511764527,
      "epoch": 4.308344680023342,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032656529189573107,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22350123524665833,
      "num_tokens": 96052628.0,
      "step": 55375
    },
    {
      "entropy": 5.8809528827667235,
      "epoch": 4.308733709395059,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003265383135873446,
      "loss": 4.9906,
      "mean_token_accuracy": 0.208490788936615,
      "num_tokens": 96060766.0,
      "step": 55380
    },
    {
      "entropy": 5.901923131942749,
      "epoch": 4.309122738766777,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003265113344969462,
      "loss": 4.9666,
      "mean_token_accuracy": 0.21055087447166443,
      "num_tokens": 96068925.0,
      "step": 55385
    },
    {
      "entropy": 5.911430168151855,
      "epoch": 4.309511768138495,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003264843546249453,
      "loss": 4.9082,
      "mean_token_accuracy": 0.21956858336925505,
      "num_tokens": 96077318.0,
      "step": 55390
    },
    {
      "entropy": 5.899334144592285,
      "epoch": 4.309900797510212,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032645737397175135,
      "loss": 4.9392,
      "mean_token_accuracy": 0.2125054642558098,
      "num_tokens": 96086077.0,
      "step": 55395
    },
    {
      "entropy": 5.984917068481446,
      "epoch": 4.310289826881929,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032643039253777356,
      "loss": 5.0249,
      "mean_token_accuracy": 0.20374854207038878,
      "num_tokens": 96095155.0,
      "step": 55400
    },
    {
      "entropy": 5.965982151031494,
      "epoch": 4.310678856253647,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032640341032342147,
      "loss": 4.9368,
      "mean_token_accuracy": 0.21403609812259675,
      "num_tokens": 96103755.0,
      "step": 55405
    },
    {
      "entropy": 5.886755800247192,
      "epoch": 4.311067885625365,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032637642732910447,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22552481889724732,
      "num_tokens": 96112025.0,
      "step": 55410
    },
    {
      "entropy": 5.96072564125061,
      "epoch": 4.311456914997082,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032634944355523196,
      "loss": 5.0185,
      "mean_token_accuracy": 0.20774497985839843,
      "num_tokens": 96120043.0,
      "step": 55415
    },
    {
      "entropy": 5.933227682113648,
      "epoch": 4.3118459443688,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003263224590022133,
      "loss": 4.992,
      "mean_token_accuracy": 0.2078128382563591,
      "num_tokens": 96128751.0,
      "step": 55420
    },
    {
      "entropy": 5.959924125671387,
      "epoch": 4.312234973740518,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032629547367045805,
      "loss": 5.0432,
      "mean_token_accuracy": 0.21657536923885345,
      "num_tokens": 96138004.0,
      "step": 55425
    },
    {
      "entropy": 5.918430042266846,
      "epoch": 4.312624003112235,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032626848756037565,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21930391043424607,
      "num_tokens": 96146472.0,
      "step": 55430
    },
    {
      "entropy": 5.906902551651001,
      "epoch": 4.313013032483952,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032624150067237553,
      "loss": 4.9134,
      "mean_token_accuracy": 0.22621539533138274,
      "num_tokens": 96154389.0,
      "step": 55435
    },
    {
      "entropy": 5.947495841979981,
      "epoch": 4.31340206185567,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003262145130068672,
      "loss": 4.9322,
      "mean_token_accuracy": 0.21716121733188629,
      "num_tokens": 96162734.0,
      "step": 55440
    },
    {
      "entropy": 5.943421840667725,
      "epoch": 4.313791091227388,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032618752456426017,
      "loss": 5.0133,
      "mean_token_accuracy": 0.2141237810254097,
      "num_tokens": 96171148.0,
      "step": 55445
    },
    {
      "entropy": 5.969653177261352,
      "epoch": 4.314180120599105,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032616053534496387,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20969539880752563,
      "num_tokens": 96179810.0,
      "step": 55450
    },
    {
      "entropy": 5.906009864807129,
      "epoch": 4.314569149970823,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003261335453493879,
      "loss": 4.9709,
      "mean_token_accuracy": 0.20920396894216536,
      "num_tokens": 96189024.0,
      "step": 55455
    },
    {
      "entropy": 5.965884637832642,
      "epoch": 4.314958179342541,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032610655457794184,
      "loss": 5.0562,
      "mean_token_accuracy": 0.20895423144102096,
      "num_tokens": 96197848.0,
      "step": 55460
    },
    {
      "entropy": 5.959708786010742,
      "epoch": 4.3153472087142575,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032607956303103516,
      "loss": 4.9696,
      "mean_token_accuracy": 0.20848642140626908,
      "num_tokens": 96206220.0,
      "step": 55465
    },
    {
      "entropy": 5.989362335205078,
      "epoch": 4.315736238085975,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032605257070907734,
      "loss": 5.0453,
      "mean_token_accuracy": 0.20335549414157866,
      "num_tokens": 96214615.0,
      "step": 55470
    },
    {
      "entropy": 5.9546948909759525,
      "epoch": 4.316125267457693,
      "grad_norm": 1.546875,
      "learning_rate": 0.00032602557761247805,
      "loss": 4.9849,
      "mean_token_accuracy": 0.22163470089435577,
      "num_tokens": 96222884.0,
      "step": 55475
    },
    {
      "entropy": 5.865255689620971,
      "epoch": 4.316514296829411,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003259985837416468,
      "loss": 4.9038,
      "mean_token_accuracy": 0.22064450830221177,
      "num_tokens": 96232597.0,
      "step": 55480
    },
    {
      "entropy": 5.862288188934326,
      "epoch": 4.316903326201128,
      "grad_norm": 1.5,
      "learning_rate": 0.0003259715890969932,
      "loss": 4.8112,
      "mean_token_accuracy": 0.228871189057827,
      "num_tokens": 96241223.0,
      "step": 55485
    },
    {
      "entropy": 5.975696134567261,
      "epoch": 4.317292355572846,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003259445936789269,
      "loss": 5.0204,
      "mean_token_accuracy": 0.20990477204322816,
      "num_tokens": 96249253.0,
      "step": 55490
    },
    {
      "entropy": 5.9124062061309814,
      "epoch": 4.317681384944564,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032591759748785753,
      "loss": 4.8499,
      "mean_token_accuracy": 0.21776163280010224,
      "num_tokens": 96257748.0,
      "step": 55495
    },
    {
      "entropy": 5.9783851146698,
      "epoch": 4.3180704143162805,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003258906005241946,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21227151900529861,
      "num_tokens": 96266527.0,
      "step": 55500
    },
    {
      "entropy": 5.850212574005127,
      "epoch": 4.318459443687998,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003258636027883478,
      "loss": 4.8764,
      "mean_token_accuracy": 0.21739228516817094,
      "num_tokens": 96274873.0,
      "step": 55505
    },
    {
      "entropy": 5.816943168640137,
      "epoch": 4.318848473059716,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003258366042807269,
      "loss": 4.8705,
      "mean_token_accuracy": 0.22213295698165894,
      "num_tokens": 96282754.0,
      "step": 55510
    },
    {
      "entropy": 5.858556509017944,
      "epoch": 4.319237502431434,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003258096050017413,
      "loss": 4.9197,
      "mean_token_accuracy": 0.21789858490228653,
      "num_tokens": 96291833.0,
      "step": 55515
    },
    {
      "entropy": 5.983089256286621,
      "epoch": 4.319626531803151,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003257826049518008,
      "loss": 4.9683,
      "mean_token_accuracy": 0.21144139617681504,
      "num_tokens": 96300533.0,
      "step": 55520
    },
    {
      "entropy": 5.8526547908782955,
      "epoch": 4.320015561174869,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003257556041313153,
      "loss": 4.8344,
      "mean_token_accuracy": 0.22152081429958342,
      "num_tokens": 96309254.0,
      "step": 55525
    },
    {
      "entropy": 5.879026699066162,
      "epoch": 4.320404590546587,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003257286025406941,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21833261102437973,
      "num_tokens": 96317884.0,
      "step": 55530
    },
    {
      "entropy": 5.806975555419922,
      "epoch": 4.3207936199183035,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032570160018034717,
      "loss": 4.7671,
      "mean_token_accuracy": 0.2249694675207138,
      "num_tokens": 96326245.0,
      "step": 55535
    },
    {
      "entropy": 5.926566791534424,
      "epoch": 4.321182649290021,
      "grad_norm": 1.515625,
      "learning_rate": 0.00032567459705068415,
      "loss": 5.057,
      "mean_token_accuracy": 0.20405865609645843,
      "num_tokens": 96334952.0,
      "step": 55540
    },
    {
      "entropy": 5.881127786636353,
      "epoch": 4.321571678661739,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032564759315211475,
      "loss": 4.8782,
      "mean_token_accuracy": 0.2210049420595169,
      "num_tokens": 96343577.0,
      "step": 55545
    },
    {
      "entropy": 5.936071872711182,
      "epoch": 4.3219607080334566,
      "grad_norm": 1.515625,
      "learning_rate": 0.00032562058848504877,
      "loss": 5.0285,
      "mean_token_accuracy": 0.2048900917172432,
      "num_tokens": 96351837.0,
      "step": 55550
    },
    {
      "entropy": 5.914323234558106,
      "epoch": 4.322349737405174,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003255935830498959,
      "loss": 4.9768,
      "mean_token_accuracy": 0.21386099904775618,
      "num_tokens": 96359812.0,
      "step": 55555
    },
    {
      "entropy": 5.943222427368164,
      "epoch": 4.322738766776892,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000325566576847066,
      "loss": 5.0269,
      "mean_token_accuracy": 0.21511978507041932,
      "num_tokens": 96368441.0,
      "step": 55560
    },
    {
      "entropy": 5.924837303161621,
      "epoch": 4.32312779614861,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003255395698769687,
      "loss": 4.9129,
      "mean_token_accuracy": 0.22975900322198867,
      "num_tokens": 96376756.0,
      "step": 55565
    },
    {
      "entropy": 5.902131652832031,
      "epoch": 4.3235168255203265,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003255125621400139,
      "loss": 4.9263,
      "mean_token_accuracy": 0.21941958069801332,
      "num_tokens": 96385325.0,
      "step": 55570
    },
    {
      "entropy": 5.94778847694397,
      "epoch": 4.323905854892044,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003254855536366114,
      "loss": 4.959,
      "mean_token_accuracy": 0.22304253727197648,
      "num_tokens": 96393993.0,
      "step": 55575
    },
    {
      "entropy": 5.9074725151062015,
      "epoch": 4.324294884263762,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000325458544367171,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22325484901666642,
      "num_tokens": 96402221.0,
      "step": 55580
    },
    {
      "entropy": 5.8628215312957765,
      "epoch": 4.3246839136354795,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032543153433210243,
      "loss": 4.9125,
      "mean_token_accuracy": 0.21844890117645263,
      "num_tokens": 96411671.0,
      "step": 55585
    },
    {
      "entropy": 5.894718265533447,
      "epoch": 4.325072943007197,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003254045235318156,
      "loss": 4.9135,
      "mean_token_accuracy": 0.2181897684931755,
      "num_tokens": 96420294.0,
      "step": 55590
    },
    {
      "entropy": 5.989685916900635,
      "epoch": 4.325461972378915,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032537751196672033,
      "loss": 5.015,
      "mean_token_accuracy": 0.2142360508441925,
      "num_tokens": 96428817.0,
      "step": 55595
    },
    {
      "entropy": 6.028869009017944,
      "epoch": 4.325851001750632,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003253504996372266,
      "loss": 5.0328,
      "mean_token_accuracy": 0.20879340618848802,
      "num_tokens": 96437383.0,
      "step": 55600
    },
    {
      "entropy": 5.8979973793029785,
      "epoch": 4.3262400311223494,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003253234865437442,
      "loss": 4.8447,
      "mean_token_accuracy": 0.22378999441862107,
      "num_tokens": 96446451.0,
      "step": 55605
    },
    {
      "entropy": 5.961281394958496,
      "epoch": 4.326629060494067,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003252964726866829,
      "loss": 5.0641,
      "mean_token_accuracy": 0.2114281788468361,
      "num_tokens": 96455478.0,
      "step": 55610
    },
    {
      "entropy": 5.870669221878051,
      "epoch": 4.327018089865785,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00032526945806645265,
      "loss": 4.914,
      "mean_token_accuracy": 0.21908537447452545,
      "num_tokens": 96464393.0,
      "step": 55615
    },
    {
      "entropy": 5.930798864364624,
      "epoch": 4.3274071192375025,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003252424426834634,
      "loss": 4.8937,
      "mean_token_accuracy": 0.233104844391346,
      "num_tokens": 96473167.0,
      "step": 55620
    },
    {
      "entropy": 5.948755598068237,
      "epoch": 4.32779614860922,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003252154265381251,
      "loss": 4.9818,
      "mean_token_accuracy": 0.2136526256799698,
      "num_tokens": 96482355.0,
      "step": 55625
    },
    {
      "entropy": 5.949413633346557,
      "epoch": 4.328185177980938,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003251884096308476,
      "loss": 4.9941,
      "mean_token_accuracy": 0.2171738028526306,
      "num_tokens": 96490704.0,
      "step": 55630
    },
    {
      "entropy": 6.0008879661560055,
      "epoch": 4.328574207352655,
      "grad_norm": 1.53125,
      "learning_rate": 0.00032516139196204087,
      "loss": 4.9912,
      "mean_token_accuracy": 0.2173539087176323,
      "num_tokens": 96498469.0,
      "step": 55635
    },
    {
      "entropy": 6.015224504470825,
      "epoch": 4.328963236724372,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032513437353211483,
      "loss": 5.0291,
      "mean_token_accuracy": 0.21051740944385527,
      "num_tokens": 96506743.0,
      "step": 55640
    },
    {
      "entropy": 5.905644702911377,
      "epoch": 4.32935226609609,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003251073543414796,
      "loss": 4.9535,
      "mean_token_accuracy": 0.21195093989372255,
      "num_tokens": 96515084.0,
      "step": 55645
    },
    {
      "entropy": 5.878390264511109,
      "epoch": 4.329741295467808,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032508033439054493,
      "loss": 4.9885,
      "mean_token_accuracy": 0.21379218250513077,
      "num_tokens": 96523750.0,
      "step": 55650
    },
    {
      "entropy": 5.9908370018005375,
      "epoch": 4.3301303248395255,
      "grad_norm": 1.375,
      "learning_rate": 0.0003250533136797208,
      "loss": 5.0007,
      "mean_token_accuracy": 0.21402197182178498,
      "num_tokens": 96532939.0,
      "step": 55655
    },
    {
      "entropy": 5.95718994140625,
      "epoch": 4.330519354211243,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032502629220941743,
      "loss": 4.9552,
      "mean_token_accuracy": 0.2090720236301422,
      "num_tokens": 96541276.0,
      "step": 55660
    },
    {
      "entropy": 5.9001209259033205,
      "epoch": 4.33090838358296,
      "grad_norm": 1.5,
      "learning_rate": 0.00032499926998004464,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2145879954099655,
      "num_tokens": 96550068.0,
      "step": 55665
    },
    {
      "entropy": 5.838719654083252,
      "epoch": 4.331297412954678,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003249722469920126,
      "loss": 4.8398,
      "mean_token_accuracy": 0.22208221405744552,
      "num_tokens": 96558939.0,
      "step": 55670
    },
    {
      "entropy": 5.948802900314331,
      "epoch": 4.331686442326395,
      "grad_norm": 1.8125,
      "learning_rate": 0.0003249452232457312,
      "loss": 4.9286,
      "mean_token_accuracy": 0.2236321523785591,
      "num_tokens": 96567277.0,
      "step": 55675
    },
    {
      "entropy": 5.883264541625977,
      "epoch": 4.332075471698113,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032491819874161046,
      "loss": 4.9018,
      "mean_token_accuracy": 0.20881546139717103,
      "num_tokens": 96576013.0,
      "step": 55680
    },
    {
      "entropy": 5.971567010879516,
      "epoch": 4.332464501069831,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003248911734800606,
      "loss": 5.0475,
      "mean_token_accuracy": 0.20842851847410201,
      "num_tokens": 96585098.0,
      "step": 55685
    },
    {
      "entropy": 5.880332326889038,
      "epoch": 4.3328535304415485,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00032486414746149154,
      "loss": 4.8875,
      "mean_token_accuracy": 0.21800680905580522,
      "num_tokens": 96593770.0,
      "step": 55690
    },
    {
      "entropy": 5.961987257003784,
      "epoch": 4.333242559813266,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032483712068631345,
      "loss": 4.9915,
      "mean_token_accuracy": 0.20999549478292465,
      "num_tokens": 96602249.0,
      "step": 55695
    },
    {
      "entropy": 5.875831031799317,
      "epoch": 4.333631589184984,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003248100931549364,
      "loss": 4.9277,
      "mean_token_accuracy": 0.2208929792046547,
      "num_tokens": 96610267.0,
      "step": 55700
    },
    {
      "entropy": 5.936951494216919,
      "epoch": 4.334020618556701,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032478306486777037,
      "loss": 4.9842,
      "mean_token_accuracy": 0.21088332086801528,
      "num_tokens": 96618954.0,
      "step": 55705
    },
    {
      "entropy": 6.013401889801026,
      "epoch": 4.334409647928418,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003247560358252256,
      "loss": 5.0195,
      "mean_token_accuracy": 0.20819290429353715,
      "num_tokens": 96628130.0,
      "step": 55710
    },
    {
      "entropy": 5.938758325576782,
      "epoch": 4.334798677300136,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032472900602771224,
      "loss": 4.9558,
      "mean_token_accuracy": 0.217748261988163,
      "num_tokens": 96636671.0,
      "step": 55715
    },
    {
      "entropy": 5.915176200866699,
      "epoch": 4.335187706671854,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032470197547564036,
      "loss": 5.0419,
      "mean_token_accuracy": 0.21294764876365663,
      "num_tokens": 96646496.0,
      "step": 55720
    },
    {
      "entropy": 5.970814943313599,
      "epoch": 4.3355767360435715,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032467494416942006,
      "loss": 5.0272,
      "mean_token_accuracy": 0.20418087393045425,
      "num_tokens": 96654985.0,
      "step": 55725
    },
    {
      "entropy": 5.922613906860351,
      "epoch": 4.335965765415289,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003246479121094616,
      "loss": 4.8655,
      "mean_token_accuracy": 0.21308761984109878,
      "num_tokens": 96663533.0,
      "step": 55730
    },
    {
      "entropy": 5.98245267868042,
      "epoch": 4.336354794787006,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000324620879296175,
      "loss": 5.0766,
      "mean_token_accuracy": 0.20774131417274475,
      "num_tokens": 96672111.0,
      "step": 55735
    },
    {
      "entropy": 5.9694822311401365,
      "epoch": 4.336743824158724,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003245938457299706,
      "loss": 4.9261,
      "mean_token_accuracy": 0.22368149012327193,
      "num_tokens": 96680312.0,
      "step": 55740
    },
    {
      "entropy": 5.857763004302979,
      "epoch": 4.337132853530441,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032456681141125853,
      "loss": 4.7943,
      "mean_token_accuracy": 0.23133268803358079,
      "num_tokens": 96688881.0,
      "step": 55745
    },
    {
      "entropy": 5.8842949867248535,
      "epoch": 4.337521882902159,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000324539776340449,
      "loss": 4.9119,
      "mean_token_accuracy": 0.2148086115717888,
      "num_tokens": 96697899.0,
      "step": 55750
    },
    {
      "entropy": 5.9994312763214115,
      "epoch": 4.337910912273877,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003245127405179522,
      "loss": 5.0968,
      "mean_token_accuracy": 0.20263098180294037,
      "num_tokens": 96707124.0,
      "step": 55755
    },
    {
      "entropy": 5.90067687034607,
      "epoch": 4.3382999416455945,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003244857039441784,
      "loss": 4.8696,
      "mean_token_accuracy": 0.22495445013046264,
      "num_tokens": 96716441.0,
      "step": 55760
    },
    {
      "entropy": 5.996205139160156,
      "epoch": 4.338688971017312,
      "grad_norm": 1.46875,
      "learning_rate": 0.00032445866661953774,
      "loss": 5.0131,
      "mean_token_accuracy": 0.1987402081489563,
      "num_tokens": 96723700.0,
      "step": 55765
    },
    {
      "entropy": 5.886493587493897,
      "epoch": 4.339078000389029,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032443162854444064,
      "loss": 4.9238,
      "mean_token_accuracy": 0.21669620275497437,
      "num_tokens": 96732391.0,
      "step": 55770
    },
    {
      "entropy": 5.873948192596435,
      "epoch": 4.339467029760747,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0003244045897192972,
      "loss": 4.9663,
      "mean_token_accuracy": 0.21250430792570113,
      "num_tokens": 96740796.0,
      "step": 55775
    },
    {
      "entropy": 6.009280729293823,
      "epoch": 4.339856059132464,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032437755014451777,
      "loss": 4.9761,
      "mean_token_accuracy": 0.21308899372816087,
      "num_tokens": 96748976.0,
      "step": 55780
    },
    {
      "entropy": 6.019100904464722,
      "epoch": 4.340245088504182,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032435050982051246,
      "loss": 5.0724,
      "mean_token_accuracy": 0.20367142856121062,
      "num_tokens": 96758923.0,
      "step": 55785
    },
    {
      "entropy": 5.918279409408569,
      "epoch": 4.3406341178759,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032432346874769187,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21782004684209824,
      "num_tokens": 96768636.0,
      "step": 55790
    },
    {
      "entropy": 5.916307592391968,
      "epoch": 4.3410231472476175,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00032429642692646614,
      "loss": 4.911,
      "mean_token_accuracy": 0.2162455588579178,
      "num_tokens": 96777200.0,
      "step": 55795
    },
    {
      "entropy": 5.942195653915405,
      "epoch": 4.341412176619334,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032426938435724554,
      "loss": 5.0572,
      "mean_token_accuracy": 0.2048630326986313,
      "num_tokens": 96785775.0,
      "step": 55800
    },
    {
      "entropy": 5.958122587203979,
      "epoch": 4.341801205991052,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032424234104044055,
      "loss": 5.0446,
      "mean_token_accuracy": 0.20813334882259368,
      "num_tokens": 96793996.0,
      "step": 55805
    },
    {
      "entropy": 5.910165166854858,
      "epoch": 4.34219023536277,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003242152969764613,
      "loss": 4.9849,
      "mean_token_accuracy": 0.20624516308307647,
      "num_tokens": 96802433.0,
      "step": 55810
    },
    {
      "entropy": 5.968124580383301,
      "epoch": 4.342579264734487,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003241882521657183,
      "loss": 5.0079,
      "mean_token_accuracy": 0.20822871327400208,
      "num_tokens": 96812077.0,
      "step": 55815
    },
    {
      "entropy": 5.956984424591065,
      "epoch": 4.342968294106205,
      "grad_norm": 1.53125,
      "learning_rate": 0.00032416120660862187,
      "loss": 4.9494,
      "mean_token_accuracy": 0.21741608530282974,
      "num_tokens": 96820148.0,
      "step": 55820
    },
    {
      "entropy": 5.923762273788452,
      "epoch": 4.343357323477923,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003241341603055823,
      "loss": 4.9952,
      "mean_token_accuracy": 0.2139963001012802,
      "num_tokens": 96828341.0,
      "step": 55825
    },
    {
      "entropy": 5.899516487121582,
      "epoch": 4.3437463528496405,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003241071132570101,
      "loss": 4.9655,
      "mean_token_accuracy": 0.21616213023662567,
      "num_tokens": 96837076.0,
      "step": 55830
    },
    {
      "entropy": 5.952333402633667,
      "epoch": 4.344135382221357,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003240800654633157,
      "loss": 4.9831,
      "mean_token_accuracy": 0.2135088250041008,
      "num_tokens": 96845784.0,
      "step": 55835
    },
    {
      "entropy": 5.999456453323364,
      "epoch": 4.344524411593075,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003240530169249094,
      "loss": 4.9404,
      "mean_token_accuracy": 0.2231002628803253,
      "num_tokens": 96853927.0,
      "step": 55840
    },
    {
      "entropy": 5.876666069030762,
      "epoch": 4.344913440964793,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003240259676422016,
      "loss": 4.842,
      "mean_token_accuracy": 0.22084612846374513,
      "num_tokens": 96863286.0,
      "step": 55845
    },
    {
      "entropy": 5.973383140563965,
      "epoch": 4.34530247033651,
      "grad_norm": 1.484375,
      "learning_rate": 0.00032399891761560274,
      "loss": 4.9903,
      "mean_token_accuracy": 0.2116517648100853,
      "num_tokens": 96871168.0,
      "step": 55850
    },
    {
      "entropy": 5.894485187530518,
      "epoch": 4.345691499708228,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003239718668455233,
      "loss": 4.9164,
      "mean_token_accuracy": 0.21415430903434754,
      "num_tokens": 96880006.0,
      "step": 55855
    },
    {
      "entropy": 5.94235782623291,
      "epoch": 4.346080529079946,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003239448153323737,
      "loss": 4.988,
      "mean_token_accuracy": 0.21311260610818863,
      "num_tokens": 96888456.0,
      "step": 55860
    },
    {
      "entropy": 5.896415853500367,
      "epoch": 4.3464695584516635,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00032391776307656456,
      "loss": 4.8577,
      "mean_token_accuracy": 0.22564922720193864,
      "num_tokens": 96898027.0,
      "step": 55865
    },
    {
      "entropy": 5.9129640579223635,
      "epoch": 4.34685858782338,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032389071007850616,
      "loss": 4.9598,
      "mean_token_accuracy": 0.2193996161222458,
      "num_tokens": 96907091.0,
      "step": 55870
    },
    {
      "entropy": 5.972544431686401,
      "epoch": 4.347247617195098,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032386365633860914,
      "loss": 5.0069,
      "mean_token_accuracy": 0.20755694210529327,
      "num_tokens": 96915958.0,
      "step": 55875
    },
    {
      "entropy": 5.870822906494141,
      "epoch": 4.347636646566816,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003238366018572838,
      "loss": 4.914,
      "mean_token_accuracy": 0.2201402187347412,
      "num_tokens": 96924793.0,
      "step": 55880
    },
    {
      "entropy": 5.971823501586914,
      "epoch": 4.348025675938533,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003238095466349407,
      "loss": 5.0841,
      "mean_token_accuracy": 0.20929877012968062,
      "num_tokens": 96933763.0,
      "step": 55885
    },
    {
      "entropy": 5.999117994308472,
      "epoch": 4.348414705310251,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003237824906719905,
      "loss": 4.9756,
      "mean_token_accuracy": 0.2184861943125725,
      "num_tokens": 96942325.0,
      "step": 55890
    },
    {
      "entropy": 5.937520122528076,
      "epoch": 4.348803734681969,
      "grad_norm": 1.46875,
      "learning_rate": 0.00032375543396884367,
      "loss": 5.0165,
      "mean_token_accuracy": 0.21021803468465805,
      "num_tokens": 96951804.0,
      "step": 55895
    },
    {
      "entropy": 5.916098356246948,
      "epoch": 4.3491927640536865,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003237283765259107,
      "loss": 5.0015,
      "mean_token_accuracy": 0.21524731665849686,
      "num_tokens": 96960272.0,
      "step": 55900
    },
    {
      "entropy": 5.871624898910523,
      "epoch": 4.349581793425403,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032370131834360215,
      "loss": 4.8475,
      "mean_token_accuracy": 0.22052532732486724,
      "num_tokens": 96968979.0,
      "step": 55905
    },
    {
      "entropy": 5.9452403545379635,
      "epoch": 4.349970822797121,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003236742594223287,
      "loss": 4.9693,
      "mean_token_accuracy": 0.20842225849628448,
      "num_tokens": 96977836.0,
      "step": 55910
    },
    {
      "entropy": 5.935708332061767,
      "epoch": 4.350359852168839,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003236471997625008,
      "loss": 4.9152,
      "mean_token_accuracy": 0.21648208200931549,
      "num_tokens": 96986220.0,
      "step": 55915
    },
    {
      "entropy": 5.916415071487426,
      "epoch": 4.350748881540556,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003236201393645291,
      "loss": 4.9826,
      "mean_token_accuracy": 0.2132427453994751,
      "num_tokens": 96995261.0,
      "step": 55920
    },
    {
      "entropy": 5.93584566116333,
      "epoch": 4.351137910912274,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032359307822882403,
      "loss": 5.0202,
      "mean_token_accuracy": 0.21350504457950592,
      "num_tokens": 97004893.0,
      "step": 55925
    },
    {
      "entropy": 5.978117847442627,
      "epoch": 4.351526940283992,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032356601635579645,
      "loss": 4.9907,
      "mean_token_accuracy": 0.21074840128421785,
      "num_tokens": 97013244.0,
      "step": 55930
    },
    {
      "entropy": 6.0033854961395265,
      "epoch": 4.351915969655709,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003235389537458569,
      "loss": 5.0217,
      "mean_token_accuracy": 0.2141931548714638,
      "num_tokens": 97021305.0,
      "step": 55935
    },
    {
      "entropy": 5.939027738571167,
      "epoch": 4.352304999027426,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00032351189039941594,
      "loss": 4.9401,
      "mean_token_accuracy": 0.21352207213640212,
      "num_tokens": 97029666.0,
      "step": 55940
    },
    {
      "entropy": 5.865205955505371,
      "epoch": 4.352694028399144,
      "grad_norm": 1.546875,
      "learning_rate": 0.00032348482631688424,
      "loss": 4.9787,
      "mean_token_accuracy": 0.20683272927999496,
      "num_tokens": 97038270.0,
      "step": 55945
    },
    {
      "entropy": 5.941774463653564,
      "epoch": 4.353083057770862,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003234577614986725,
      "loss": 4.9683,
      "mean_token_accuracy": 0.21826982945203782,
      "num_tokens": 97047019.0,
      "step": 55950
    },
    {
      "entropy": 5.913543605804444,
      "epoch": 4.353472087142579,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003234306959451914,
      "loss": 4.9102,
      "mean_token_accuracy": 0.20883888900279998,
      "num_tokens": 97056193.0,
      "step": 55955
    },
    {
      "entropy": 5.901032114028931,
      "epoch": 4.353861116514297,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00032340362965685145,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22198074907064438,
      "num_tokens": 97064556.0,
      "step": 55960
    },
    {
      "entropy": 5.958832502365112,
      "epoch": 4.354250145886015,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003233765626340636,
      "loss": 5.0767,
      "mean_token_accuracy": 0.20854007005691527,
      "num_tokens": 97072923.0,
      "step": 55965
    },
    {
      "entropy": 5.980124139785767,
      "epoch": 4.354639175257732,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003233494948772384,
      "loss": 5.0441,
      "mean_token_accuracy": 0.2147770807147026,
      "num_tokens": 97081811.0,
      "step": 55970
    },
    {
      "entropy": 5.908018970489502,
      "epoch": 4.355028204629449,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00032332242638678647,
      "loss": 4.913,
      "mean_token_accuracy": 0.22189415395259857,
      "num_tokens": 97090051.0,
      "step": 55975
    },
    {
      "entropy": 5.976730489730835,
      "epoch": 4.355417234001167,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003232953571631187,
      "loss": 4.979,
      "mean_token_accuracy": 0.219314044713974,
      "num_tokens": 97098538.0,
      "step": 55980
    },
    {
      "entropy": 5.940883779525757,
      "epoch": 4.355806263372885,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003232682872066457,
      "loss": 4.9768,
      "mean_token_accuracy": 0.21777852922677993,
      "num_tokens": 97107163.0,
      "step": 55985
    },
    {
      "entropy": 5.905527877807617,
      "epoch": 4.356195292744602,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003232412165177783,
      "loss": 4.885,
      "mean_token_accuracy": 0.21814920604228974,
      "num_tokens": 97115855.0,
      "step": 55990
    },
    {
      "entropy": 5.872034978866577,
      "epoch": 4.35658432211632,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032321414509692726,
      "loss": 4.9141,
      "mean_token_accuracy": 0.21843722313642502,
      "num_tokens": 97125383.0,
      "step": 55995
    },
    {
      "entropy": 5.929622840881348,
      "epoch": 4.356973351488037,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032318707294450326,
      "loss": 5.0103,
      "mean_token_accuracy": 0.21229803264141084,
      "num_tokens": 97133932.0,
      "step": 56000
    },
    {
      "entropy": 5.972325420379638,
      "epoch": 4.357362380859755,
      "grad_norm": 1.5,
      "learning_rate": 0.0003231600000609171,
      "loss": 4.9904,
      "mean_token_accuracy": 0.21252066493034363,
      "num_tokens": 97142371.0,
      "step": 56005
    },
    {
      "entropy": 5.95780987739563,
      "epoch": 4.357751410231472,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00032313292644657967,
      "loss": 4.9888,
      "mean_token_accuracy": 0.20960887223482133,
      "num_tokens": 97150369.0,
      "step": 56010
    },
    {
      "entropy": 5.944254159927368,
      "epoch": 4.35814043960319,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003231058521019017,
      "loss": 5.011,
      "mean_token_accuracy": 0.20740724951028824,
      "num_tokens": 97158967.0,
      "step": 56015
    },
    {
      "entropy": 5.875265741348267,
      "epoch": 4.358529468974908,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003230787770272939,
      "loss": 4.9377,
      "mean_token_accuracy": 0.2161858409643173,
      "num_tokens": 97167420.0,
      "step": 56020
    },
    {
      "entropy": 5.941634321212769,
      "epoch": 4.358918498346625,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003230517012231673,
      "loss": 4.9429,
      "mean_token_accuracy": 0.20788084715604782,
      "num_tokens": 97175557.0,
      "step": 56025
    },
    {
      "entropy": 5.966920804977417,
      "epoch": 4.359307527718343,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032302462468993253,
      "loss": 4.9912,
      "mean_token_accuracy": 0.20945829451084136,
      "num_tokens": 97184206.0,
      "step": 56030
    },
    {
      "entropy": 5.963095378875733,
      "epoch": 4.359696557090061,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003229975474280006,
      "loss": 4.9504,
      "mean_token_accuracy": 0.21910752803087236,
      "num_tokens": 97193496.0,
      "step": 56035
    },
    {
      "entropy": 5.923631191253662,
      "epoch": 4.360085586461778,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032297046943778216,
      "loss": 4.9488,
      "mean_token_accuracy": 0.22001154273748397,
      "num_tokens": 97202608.0,
      "step": 56040
    },
    {
      "entropy": 5.94158091545105,
      "epoch": 4.360474615833495,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032294339071968833,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20597779601812363,
      "num_tokens": 97211692.0,
      "step": 56045
    },
    {
      "entropy": 5.958036088943482,
      "epoch": 4.360863645205213,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032291631127412976,
      "loss": 4.9382,
      "mean_token_accuracy": 0.21796551197767258,
      "num_tokens": 97220148.0,
      "step": 56050
    },
    {
      "entropy": 5.92380313873291,
      "epoch": 4.361252674576931,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003228892311015174,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22404381036758422,
      "num_tokens": 97229687.0,
      "step": 56055
    },
    {
      "entropy": 5.918945741653443,
      "epoch": 4.361641703948648,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032286215020226224,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2236122280359268,
      "num_tokens": 97238122.0,
      "step": 56060
    },
    {
      "entropy": 5.919385290145874,
      "epoch": 4.362030733320366,
      "grad_norm": 1.375,
      "learning_rate": 0.00032283506857677517,
      "loss": 4.9326,
      "mean_token_accuracy": 0.223294934630394,
      "num_tokens": 97246617.0,
      "step": 56065
    },
    {
      "entropy": 5.841565227508545,
      "epoch": 4.362419762692083,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003228079862254671,
      "loss": 4.8969,
      "mean_token_accuracy": 0.22199808210134506,
      "num_tokens": 97254863.0,
      "step": 56070
    },
    {
      "entropy": 5.902840948104858,
      "epoch": 4.362808792063801,
      "grad_norm": 1.375,
      "learning_rate": 0.00032278090314874887,
      "loss": 5.0231,
      "mean_token_accuracy": 0.19830291718244553,
      "num_tokens": 97263818.0,
      "step": 56075
    },
    {
      "entropy": 5.8855448246002195,
      "epoch": 4.363197821435518,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00032275381934703147,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21328169852495193,
      "num_tokens": 97271974.0,
      "step": 56080
    },
    {
      "entropy": 5.956129360198974,
      "epoch": 4.363586850807236,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003227267348207258,
      "loss": 4.9409,
      "mean_token_accuracy": 0.22072838842868805,
      "num_tokens": 97281192.0,
      "step": 56085
    },
    {
      "entropy": 5.967231941223145,
      "epoch": 4.363975880178954,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032269964957024305,
      "loss": 5.0186,
      "mean_token_accuracy": 0.2091701716184616,
      "num_tokens": 97290290.0,
      "step": 56090
    },
    {
      "entropy": 5.9358617782592775,
      "epoch": 4.364364909550671,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032267256359599396,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22697941660881044,
      "num_tokens": 97298422.0,
      "step": 56095
    },
    {
      "entropy": 5.978378868103027,
      "epoch": 4.364753938922389,
      "grad_norm": 1.484375,
      "learning_rate": 0.00032264547689838963,
      "loss": 4.9668,
      "mean_token_accuracy": 0.21296575516462327,
      "num_tokens": 97306741.0,
      "step": 56100
    },
    {
      "entropy": 5.89862642288208,
      "epoch": 4.365142968294106,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00032261838947784103,
      "loss": 4.9057,
      "mean_token_accuracy": 0.22273586243391036,
      "num_tokens": 97315585.0,
      "step": 56105
    },
    {
      "entropy": 5.938363122940063,
      "epoch": 4.365531997665824,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003225913013347591,
      "loss": 5.0001,
      "mean_token_accuracy": 0.20903493762016295,
      "num_tokens": 97324137.0,
      "step": 56110
    },
    {
      "entropy": 5.977956676483155,
      "epoch": 4.365921027037541,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032256421246955504,
      "loss": 4.9748,
      "mean_token_accuracy": 0.21186740547418595,
      "num_tokens": 97332701.0,
      "step": 56115
    },
    {
      "entropy": 5.925376462936401,
      "epoch": 4.366310056409259,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003225371228826396,
      "loss": 4.9313,
      "mean_token_accuracy": 0.22004767805337905,
      "num_tokens": 97342365.0,
      "step": 56120
    },
    {
      "entropy": 6.016350603103637,
      "epoch": 4.366699085780977,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003225100325744241,
      "loss": 4.9977,
      "mean_token_accuracy": 0.2063423216342926,
      "num_tokens": 97350031.0,
      "step": 56125
    },
    {
      "entropy": 5.904652500152588,
      "epoch": 4.367088115152694,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003224829415453194,
      "loss": 4.9547,
      "mean_token_accuracy": 0.21403586566448213,
      "num_tokens": 97358942.0,
      "step": 56130
    },
    {
      "entropy": 5.952938508987427,
      "epoch": 4.367477144524411,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003224558497957367,
      "loss": 5.0036,
      "mean_token_accuracy": 0.21067202240228652,
      "num_tokens": 97367019.0,
      "step": 56135
    },
    {
      "entropy": 5.902221918106079,
      "epoch": 4.367866173896129,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032242875732608697,
      "loss": 4.9232,
      "mean_token_accuracy": 0.2214179590344429,
      "num_tokens": 97375812.0,
      "step": 56140
    },
    {
      "entropy": 5.915839338302613,
      "epoch": 4.3682552032678466,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003224016641367814,
      "loss": 4.9817,
      "mean_token_accuracy": 0.21580185741186142,
      "num_tokens": 97385038.0,
      "step": 56145
    },
    {
      "entropy": 5.947696685791016,
      "epoch": 4.368644232639564,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032237457022823095,
      "loss": 4.9739,
      "mean_token_accuracy": 0.2163414478302002,
      "num_tokens": 97393695.0,
      "step": 56150
    },
    {
      "entropy": 5.952973318099976,
      "epoch": 4.369033262011282,
      "grad_norm": 1.5,
      "learning_rate": 0.00032234747560084675,
      "loss": 4.9613,
      "mean_token_accuracy": 0.21989718377590178,
      "num_tokens": 97401959.0,
      "step": 56155
    },
    {
      "entropy": 5.991323757171631,
      "epoch": 4.369422291383,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032232038025504015,
      "loss": 4.9763,
      "mean_token_accuracy": 0.2169852778315544,
      "num_tokens": 97410756.0,
      "step": 56160
    },
    {
      "entropy": 5.933845806121826,
      "epoch": 4.369811320754717,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032229328419122196,
      "loss": 4.9188,
      "mean_token_accuracy": 0.216201551258564,
      "num_tokens": 97418910.0,
      "step": 56165
    },
    {
      "entropy": 5.927053546905517,
      "epoch": 4.370200350126434,
      "grad_norm": 1.5,
      "learning_rate": 0.0003222661874098035,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20316251665353774,
      "num_tokens": 97427552.0,
      "step": 56170
    },
    {
      "entropy": 5.946160459518433,
      "epoch": 4.370589379498152,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003222390899111959,
      "loss": 4.9697,
      "mean_token_accuracy": 0.21411100178956985,
      "num_tokens": 97436242.0,
      "step": 56175
    },
    {
      "entropy": 5.980920314788818,
      "epoch": 4.3709784088698695,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003222119916958103,
      "loss": 4.9921,
      "mean_token_accuracy": 0.208387354016304,
      "num_tokens": 97444299.0,
      "step": 56180
    },
    {
      "entropy": 5.884223890304566,
      "epoch": 4.371367438241587,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032218489276405777,
      "loss": 4.9714,
      "mean_token_accuracy": 0.21913256794214248,
      "num_tokens": 97452768.0,
      "step": 56185
    },
    {
      "entropy": 5.9413690090179445,
      "epoch": 4.371756467613305,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00032215779311634967,
      "loss": 5.0843,
      "mean_token_accuracy": 0.2085232451558113,
      "num_tokens": 97461657.0,
      "step": 56190
    },
    {
      "entropy": 5.947444772720337,
      "epoch": 4.372145496985023,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032213069275309714,
      "loss": 4.9089,
      "mean_token_accuracy": 0.21769207268953322,
      "num_tokens": 97469720.0,
      "step": 56195
    },
    {
      "entropy": 5.872397708892822,
      "epoch": 4.37253452635674,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032210359167471136,
      "loss": 4.8818,
      "mean_token_accuracy": 0.21272236108779907,
      "num_tokens": 97478004.0,
      "step": 56200
    },
    {
      "entropy": 5.9213978290557865,
      "epoch": 4.372923555728457,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003220764898816035,
      "loss": 5.0119,
      "mean_token_accuracy": 0.20327015370130538,
      "num_tokens": 97487421.0,
      "step": 56205
    },
    {
      "entropy": 5.970058393478394,
      "epoch": 4.373312585100175,
      "grad_norm": 1.46875,
      "learning_rate": 0.00032204938737418495,
      "loss": 4.977,
      "mean_token_accuracy": 0.21098547875881196,
      "num_tokens": 97495533.0,
      "step": 56210
    },
    {
      "entropy": 5.984961605072021,
      "epoch": 4.3737016144718925,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00032202228415286664,
      "loss": 5.0702,
      "mean_token_accuracy": 0.20539511293172835,
      "num_tokens": 97504398.0,
      "step": 56215
    },
    {
      "entropy": 5.967674493789673,
      "epoch": 4.37409064384361,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003219951802180602,
      "loss": 5.0137,
      "mean_token_accuracy": 0.2133561208844185,
      "num_tokens": 97513024.0,
      "step": 56220
    },
    {
      "entropy": 5.926362609863281,
      "epoch": 4.374479673215328,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0003219680755701766,
      "loss": 4.8969,
      "mean_token_accuracy": 0.22283420115709304,
      "num_tokens": 97522514.0,
      "step": 56225
    },
    {
      "entropy": 5.969330358505249,
      "epoch": 4.374868702587046,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032194097020962726,
      "loss": 4.9915,
      "mean_token_accuracy": 0.2125408098101616,
      "num_tokens": 97530972.0,
      "step": 56230
    },
    {
      "entropy": 6.015900897979736,
      "epoch": 4.375257731958763,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003219138641368234,
      "loss": 5.0921,
      "mean_token_accuracy": 0.21186351627111435,
      "num_tokens": 97539551.0,
      "step": 56235
    },
    {
      "entropy": 5.910834169387817,
      "epoch": 4.37564676133048,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032188675735217627,
      "loss": 4.9582,
      "mean_token_accuracy": 0.2131665378808975,
      "num_tokens": 97548521.0,
      "step": 56240
    },
    {
      "entropy": 5.9350823879241945,
      "epoch": 4.376035790702198,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003218596498560973,
      "loss": 4.9648,
      "mean_token_accuracy": 0.21525688469409943,
      "num_tokens": 97556602.0,
      "step": 56245
    },
    {
      "entropy": 5.904653263092041,
      "epoch": 4.3764248200739155,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003218325416489976,
      "loss": 4.9271,
      "mean_token_accuracy": 0.2170117661356926,
      "num_tokens": 97564728.0,
      "step": 56250
    },
    {
      "entropy": 5.864369535446167,
      "epoch": 4.376813849445633,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003218054327312888,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2243138685822487,
      "num_tokens": 97574445.0,
      "step": 56255
    },
    {
      "entropy": 5.936367177963257,
      "epoch": 4.377202878817351,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032177832310338193,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22559448778629304,
      "num_tokens": 97583313.0,
      "step": 56260
    },
    {
      "entropy": 5.954859733581543,
      "epoch": 4.377591908189069,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032175121276568853,
      "loss": 4.9546,
      "mean_token_accuracy": 0.2172989308834076,
      "num_tokens": 97591429.0,
      "step": 56265
    },
    {
      "entropy": 5.838796997070313,
      "epoch": 4.377980937560785,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032172410171861987,
      "loss": 4.9016,
      "mean_token_accuracy": 0.2140534847974777,
      "num_tokens": 97600062.0,
      "step": 56270
    },
    {
      "entropy": 5.961314630508423,
      "epoch": 4.378369966932503,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032169698996258733,
      "loss": 4.8995,
      "mean_token_accuracy": 0.22232256829738617,
      "num_tokens": 97607893.0,
      "step": 56275
    },
    {
      "entropy": 5.908568859100342,
      "epoch": 4.378758996304221,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003216698774980023,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21947636753320693,
      "num_tokens": 97615965.0,
      "step": 56280
    },
    {
      "entropy": 5.914955949783325,
      "epoch": 4.3791480256759385,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00032164276432527605,
      "loss": 5.001,
      "mean_token_accuracy": 0.21439209431409836,
      "num_tokens": 97625191.0,
      "step": 56285
    },
    {
      "entropy": 5.941383790969849,
      "epoch": 4.379537055047656,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003216156504448202,
      "loss": 4.8994,
      "mean_token_accuracy": 0.2185932993888855,
      "num_tokens": 97633323.0,
      "step": 56290
    },
    {
      "entropy": 6.0086273670196535,
      "epoch": 4.379926084419374,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003215885358570461,
      "loss": 5.0124,
      "mean_token_accuracy": 0.21894364953041076,
      "num_tokens": 97642597.0,
      "step": 56295
    },
    {
      "entropy": 5.9069640159606935,
      "epoch": 4.380315113791092,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003215614205623651,
      "loss": 4.9546,
      "mean_token_accuracy": 0.22181523740291595,
      "num_tokens": 97651912.0,
      "step": 56300
    },
    {
      "entropy": 5.921082925796509,
      "epoch": 4.380704143162808,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003215343045611886,
      "loss": 5.015,
      "mean_token_accuracy": 0.21282305419445038,
      "num_tokens": 97660757.0,
      "step": 56305
    },
    {
      "entropy": 5.949446296691894,
      "epoch": 4.381093172534526,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003215071878539281,
      "loss": 4.9723,
      "mean_token_accuracy": 0.2135314866900444,
      "num_tokens": 97669310.0,
      "step": 56310
    },
    {
      "entropy": 5.918653106689453,
      "epoch": 4.381482201906244,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032148007044099507,
      "loss": 4.9999,
      "mean_token_accuracy": 0.21205658465623856,
      "num_tokens": 97678624.0,
      "step": 56315
    },
    {
      "entropy": 5.897065782546997,
      "epoch": 4.3818712312779615,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003214529523228009,
      "loss": 4.9326,
      "mean_token_accuracy": 0.20947071313858032,
      "num_tokens": 97687763.0,
      "step": 56320
    },
    {
      "entropy": 5.813525295257568,
      "epoch": 4.382260260649679,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003214258334997572,
      "loss": 4.82,
      "mean_token_accuracy": 0.22335976362228394,
      "num_tokens": 97696992.0,
      "step": 56325
    },
    {
      "entropy": 5.877807903289795,
      "epoch": 4.382649290021397,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003213987139722753,
      "loss": 4.9581,
      "mean_token_accuracy": 0.2174182191491127,
      "num_tokens": 97706544.0,
      "step": 56330
    },
    {
      "entropy": 5.90077919960022,
      "epoch": 4.383038319393114,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032137159374076693,
      "loss": 4.9311,
      "mean_token_accuracy": 0.20944517552852632,
      "num_tokens": 97715672.0,
      "step": 56335
    },
    {
      "entropy": 5.919169521331787,
      "epoch": 4.383427348764831,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003213444728056432,
      "loss": 4.9043,
      "mean_token_accuracy": 0.2251693934202194,
      "num_tokens": 97724526.0,
      "step": 56340
    },
    {
      "entropy": 5.859677505493164,
      "epoch": 4.383816378136549,
      "grad_norm": 1.375,
      "learning_rate": 0.00032131735116731604,
      "loss": 4.8768,
      "mean_token_accuracy": 0.21731405109167098,
      "num_tokens": 97733275.0,
      "step": 56345
    },
    {
      "entropy": 5.99262490272522,
      "epoch": 4.384205407508267,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003212902288261966,
      "loss": 5.0497,
      "mean_token_accuracy": 0.20619826167821884,
      "num_tokens": 97742490.0,
      "step": 56350
    },
    {
      "entropy": 5.93501615524292,
      "epoch": 4.3845944368799845,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003212631057826968,
      "loss": 4.9822,
      "mean_token_accuracy": 0.2118422955274582,
      "num_tokens": 97751596.0,
      "step": 56355
    },
    {
      "entropy": 5.887598752975464,
      "epoch": 4.384983466251702,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032123598203722786,
      "loss": 4.9583,
      "mean_token_accuracy": 0.204955492913723,
      "num_tokens": 97760236.0,
      "step": 56360
    },
    {
      "entropy": 5.932577323913574,
      "epoch": 4.38537249562342,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003212088575902016,
      "loss": 5.0442,
      "mean_token_accuracy": 0.2122553765773773,
      "num_tokens": 97769150.0,
      "step": 56365
    },
    {
      "entropy": 5.962339925765991,
      "epoch": 4.385761524995137,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00032118173244202947,
      "loss": 4.9646,
      "mean_token_accuracy": 0.2105802834033966,
      "num_tokens": 97777864.0,
      "step": 56370
    },
    {
      "entropy": 5.978813982009887,
      "epoch": 4.386150554366854,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032115460659312304,
      "loss": 4.9545,
      "mean_token_accuracy": 0.21664247065782546,
      "num_tokens": 97786417.0,
      "step": 56375
    },
    {
      "entropy": 5.958394527435303,
      "epoch": 4.386539583738572,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003211274800438939,
      "loss": 5.0502,
      "mean_token_accuracy": 0.2077595978975296,
      "num_tokens": 97795331.0,
      "step": 56380
    },
    {
      "entropy": 5.977413702011108,
      "epoch": 4.38692861311029,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003211003527947537,
      "loss": 5.0097,
      "mean_token_accuracy": 0.21154166162014007,
      "num_tokens": 97804099.0,
      "step": 56385
    },
    {
      "entropy": 5.949991369247437,
      "epoch": 4.3873176424820075,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032107322484611396,
      "loss": 4.9044,
      "mean_token_accuracy": 0.21822178810834886,
      "num_tokens": 97812401.0,
      "step": 56390
    },
    {
      "entropy": 5.906556510925293,
      "epoch": 4.387706671853725,
      "grad_norm": 1.375,
      "learning_rate": 0.00032104609619838635,
      "loss": 4.9729,
      "mean_token_accuracy": 0.2084787979722023,
      "num_tokens": 97821842.0,
      "step": 56395
    },
    {
      "entropy": 5.926098108291626,
      "epoch": 4.388095701225443,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003210189668519826,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2207949623465538,
      "num_tokens": 97829911.0,
      "step": 56400
    },
    {
      "entropy": 5.86351466178894,
      "epoch": 4.38848473059716,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003209918368073143,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2165749505162239,
      "num_tokens": 97838789.0,
      "step": 56405
    },
    {
      "entropy": 5.944088268280029,
      "epoch": 4.388873759968877,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000320964706064793,
      "loss": 4.9762,
      "mean_token_accuracy": 0.21002319306135178,
      "num_tokens": 97847440.0,
      "step": 56410
    },
    {
      "entropy": 5.9059055805206295,
      "epoch": 4.389262789340595,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032093757462483055,
      "loss": 4.882,
      "mean_token_accuracy": 0.2160208895802498,
      "num_tokens": 97856265.0,
      "step": 56415
    },
    {
      "entropy": 5.929558801651001,
      "epoch": 4.389651818712313,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032091044248783835,
      "loss": 4.9682,
      "mean_token_accuracy": 0.21483466476202012,
      "num_tokens": 97865226.0,
      "step": 56420
    },
    {
      "entropy": 5.9117718696594235,
      "epoch": 4.3900408480840305,
      "grad_norm": 1.375,
      "learning_rate": 0.00032088330965422836,
      "loss": 4.942,
      "mean_token_accuracy": 0.21430072337388992,
      "num_tokens": 97874396.0,
      "step": 56425
    },
    {
      "entropy": 5.876473236083984,
      "epoch": 4.390429877455748,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032085617612441215,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22171937972307204,
      "num_tokens": 97883349.0,
      "step": 56430
    },
    {
      "entropy": 5.926604795455932,
      "epoch": 4.390818906827466,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003208290418988015,
      "loss": 5.0066,
      "mean_token_accuracy": 0.20565370768308638,
      "num_tokens": 97891594.0,
      "step": 56435
    },
    {
      "entropy": 5.949440145492554,
      "epoch": 4.391207936199183,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000320801906977808,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21548758298158646,
      "num_tokens": 97900322.0,
      "step": 56440
    },
    {
      "entropy": 6.001616621017456,
      "epoch": 4.3915969655709,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00032077477136184356,
      "loss": 4.9548,
      "mean_token_accuracy": 0.21395036280155183,
      "num_tokens": 97909169.0,
      "step": 56445
    },
    {
      "entropy": 5.96880989074707,
      "epoch": 4.391985994942618,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0003207476350513198,
      "loss": 5.0152,
      "mean_token_accuracy": 0.2139275550842285,
      "num_tokens": 97916784.0,
      "step": 56450
    },
    {
      "entropy": 5.938837003707886,
      "epoch": 4.392375024314336,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003207204980466485,
      "loss": 5.0244,
      "mean_token_accuracy": 0.20933600813150405,
      "num_tokens": 97925899.0,
      "step": 56455
    },
    {
      "entropy": 5.88900318145752,
      "epoch": 4.3927640536860535,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032069336034824136,
      "loss": 4.9474,
      "mean_token_accuracy": 0.21331433951854706,
      "num_tokens": 97934632.0,
      "step": 56460
    },
    {
      "entropy": 5.9370969295501705,
      "epoch": 4.393153083057771,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032066622195651024,
      "loss": 4.9456,
      "mean_token_accuracy": 0.2151886910200119,
      "num_tokens": 97943381.0,
      "step": 56465
    },
    {
      "entropy": 5.938102436065674,
      "epoch": 4.393542112429488,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032063908287186687,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21777980923652648,
      "num_tokens": 97952667.0,
      "step": 56470
    },
    {
      "entropy": 5.895109510421753,
      "epoch": 4.393931141801206,
      "grad_norm": 1.59375,
      "learning_rate": 0.00032061194309472314,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2169489488005638,
      "num_tokens": 97960820.0,
      "step": 56475
    },
    {
      "entropy": 5.960391664505005,
      "epoch": 4.394320171172923,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032058480262549065,
      "loss": 4.972,
      "mean_token_accuracy": 0.21583024114370347,
      "num_tokens": 97969519.0,
      "step": 56480
    },
    {
      "entropy": 5.867330646514892,
      "epoch": 4.394709200544641,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003205576614645814,
      "loss": 4.9407,
      "mean_token_accuracy": 0.21840080469846726,
      "num_tokens": 97978838.0,
      "step": 56485
    },
    {
      "entropy": 5.865899848937988,
      "epoch": 4.395098229916359,
      "grad_norm": 1.546875,
      "learning_rate": 0.00032053051961240716,
      "loss": 4.855,
      "mean_token_accuracy": 0.22793754786252976,
      "num_tokens": 97986912.0,
      "step": 56490
    },
    {
      "entropy": 5.928740310668945,
      "epoch": 4.3954872592880765,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003205033770693797,
      "loss": 4.9949,
      "mean_token_accuracy": 0.21615003645420075,
      "num_tokens": 97995716.0,
      "step": 56495
    },
    {
      "entropy": 5.984345388412476,
      "epoch": 4.395876288659794,
      "grad_norm": 1.375,
      "learning_rate": 0.000320476233835911,
      "loss": 5.0281,
      "mean_token_accuracy": 0.21139513999223708,
      "num_tokens": 98004353.0,
      "step": 56500
    },
    {
      "entropy": 5.9611045837402346,
      "epoch": 4.396265318031511,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00032044908991241274,
      "loss": 4.9873,
      "mean_token_accuracy": 0.20275863260030746,
      "num_tokens": 98013091.0,
      "step": 56505
    },
    {
      "entropy": 5.9299938678741455,
      "epoch": 4.396654347403229,
      "grad_norm": 1.46875,
      "learning_rate": 0.00032042194529929694,
      "loss": 4.901,
      "mean_token_accuracy": 0.2232312634587288,
      "num_tokens": 98021296.0,
      "step": 56510
    },
    {
      "entropy": 5.927159929275513,
      "epoch": 4.397043376774946,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032039479999697545,
      "loss": 4.9551,
      "mean_token_accuracy": 0.22175492495298385,
      "num_tokens": 98029470.0,
      "step": 56515
    },
    {
      "entropy": 5.9263989448547365,
      "epoch": 4.397432406146664,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032036765400586005,
      "loss": 4.9636,
      "mean_token_accuracy": 0.21583996415138246,
      "num_tokens": 98037542.0,
      "step": 56520
    },
    {
      "entropy": 5.948272275924682,
      "epoch": 4.397821435518382,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032034050732636274,
      "loss": 4.9518,
      "mean_token_accuracy": 0.2176176503300667,
      "num_tokens": 98045683.0,
      "step": 56525
    },
    {
      "entropy": 5.92075834274292,
      "epoch": 4.3982104648900995,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003203133599588954,
      "loss": 4.9924,
      "mean_token_accuracy": 0.2161384180188179,
      "num_tokens": 98054147.0,
      "step": 56530
    },
    {
      "entropy": 5.952115154266357,
      "epoch": 4.398599494261817,
      "grad_norm": 1.515625,
      "learning_rate": 0.00032028621190386997,
      "loss": 5.0296,
      "mean_token_accuracy": 0.20907048732042313,
      "num_tokens": 98063140.0,
      "step": 56535
    },
    {
      "entropy": 5.883146333694458,
      "epoch": 4.398988523633534,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032025906316169835,
      "loss": 4.8856,
      "mean_token_accuracy": 0.22200715988874437,
      "num_tokens": 98071725.0,
      "step": 56540
    },
    {
      "entropy": 5.874318933486938,
      "epoch": 4.399377553005252,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003202319137327924,
      "loss": 4.9157,
      "mean_token_accuracy": 0.2241577312350273,
      "num_tokens": 98080810.0,
      "step": 56545
    },
    {
      "entropy": 5.958617401123047,
      "epoch": 4.399766582376969,
      "grad_norm": 1.484375,
      "learning_rate": 0.00032020476361756424,
      "loss": 4.9816,
      "mean_token_accuracy": 0.21890343874692916,
      "num_tokens": 98089420.0,
      "step": 56550
    },
    {
      "entropy": 5.873955011367798,
      "epoch": 4.400155611748687,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032017761281642564,
      "loss": 4.903,
      "mean_token_accuracy": 0.2219498112797737,
      "num_tokens": 98098517.0,
      "step": 56555
    },
    {
      "entropy": 5.926497507095337,
      "epoch": 4.400544641120405,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003201504613297888,
      "loss": 4.8868,
      "mean_token_accuracy": 0.21882407367229462,
      "num_tokens": 98106895.0,
      "step": 56560
    },
    {
      "entropy": 5.984807968139648,
      "epoch": 4.4009336704921225,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032012330915806553,
      "loss": 5.0722,
      "mean_token_accuracy": 0.2066210463643074,
      "num_tokens": 98116001.0,
      "step": 56565
    },
    {
      "entropy": 5.895428943634033,
      "epoch": 4.40132269986384,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032009615630166786,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21669747084379196,
      "num_tokens": 98124369.0,
      "step": 56570
    },
    {
      "entropy": 5.870689249038696,
      "epoch": 4.401711729235557,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003200690027610078,
      "loss": 4.8145,
      "mean_token_accuracy": 0.22550151646137237,
      "num_tokens": 98132875.0,
      "step": 56575
    },
    {
      "entropy": 5.971677017211914,
      "epoch": 4.402100758607275,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003200418485364974,
      "loss": 5.0586,
      "mean_token_accuracy": 0.21063570976257323,
      "num_tokens": 98141185.0,
      "step": 56580
    },
    {
      "entropy": 5.928086662292481,
      "epoch": 4.402489787978992,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003200146936285485,
      "loss": 4.9456,
      "mean_token_accuracy": 0.21375012993812562,
      "num_tokens": 98149944.0,
      "step": 56585
    },
    {
      "entropy": 5.965887975692749,
      "epoch": 4.40287881735071,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003199875380375734,
      "loss": 5.0293,
      "mean_token_accuracy": 0.20691123008728027,
      "num_tokens": 98157716.0,
      "step": 56590
    },
    {
      "entropy": 5.949393272399902,
      "epoch": 4.403267846722428,
      "grad_norm": 1.46875,
      "learning_rate": 0.000319960381763984,
      "loss": 4.9981,
      "mean_token_accuracy": 0.2118916094303131,
      "num_tokens": 98166108.0,
      "step": 56595
    },
    {
      "entropy": 5.845628452301026,
      "epoch": 4.4036568760941455,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003199332248081923,
      "loss": 4.8259,
      "mean_token_accuracy": 0.22670550048351287,
      "num_tokens": 98174182.0,
      "step": 56600
    },
    {
      "entropy": 5.8828497409820555,
      "epoch": 4.404045905465862,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031990606717061055,
      "loss": 4.9753,
      "mean_token_accuracy": 0.22398349046707153,
      "num_tokens": 98182955.0,
      "step": 56605
    },
    {
      "entropy": 5.918631601333618,
      "epoch": 4.40443493483758,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003198789088516506,
      "loss": 4.9078,
      "mean_token_accuracy": 0.2223304182291031,
      "num_tokens": 98192200.0,
      "step": 56610
    },
    {
      "entropy": 6.031404781341553,
      "epoch": 4.404823964209298,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00031985174985172467,
      "loss": 5.0063,
      "mean_token_accuracy": 0.21497859358787536,
      "num_tokens": 98199731.0,
      "step": 56615
    },
    {
      "entropy": 5.958570241928101,
      "epoch": 4.405212993581015,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003198245901712449,
      "loss": 4.9957,
      "mean_token_accuracy": 0.2112015664577484,
      "num_tokens": 98208535.0,
      "step": 56620
    },
    {
      "entropy": 5.970226526260376,
      "epoch": 4.405602022952733,
      "grad_norm": 1.375,
      "learning_rate": 0.0003197974298106232,
      "loss": 4.9738,
      "mean_token_accuracy": 0.22197719812393188,
      "num_tokens": 98217869.0,
      "step": 56625
    },
    {
      "entropy": 5.938514232635498,
      "epoch": 4.405991052324451,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00031977026877027186,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2174789860844612,
      "num_tokens": 98225795.0,
      "step": 56630
    },
    {
      "entropy": 5.940296030044555,
      "epoch": 4.4063800816961685,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031974310705060297,
      "loss": 4.9747,
      "mean_token_accuracy": 0.2140609696507454,
      "num_tokens": 98235367.0,
      "step": 56635
    },
    {
      "entropy": 5.9591333866119385,
      "epoch": 4.406769111067885,
      "grad_norm": 1.375,
      "learning_rate": 0.0003197159446520286,
      "loss": 5.009,
      "mean_token_accuracy": 0.21677228808403015,
      "num_tokens": 98244530.0,
      "step": 56640
    },
    {
      "entropy": 5.948100805282593,
      "epoch": 4.407158140439603,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031968878157496105,
      "loss": 4.939,
      "mean_token_accuracy": 0.21955470442771913,
      "num_tokens": 98254435.0,
      "step": 56645
    },
    {
      "entropy": 5.937676811218262,
      "epoch": 4.407547169811321,
      "grad_norm": 1.46875,
      "learning_rate": 0.00031966161781981224,
      "loss": 4.9754,
      "mean_token_accuracy": 0.21046240478754044,
      "num_tokens": 98263600.0,
      "step": 56650
    },
    {
      "entropy": 5.943836975097656,
      "epoch": 4.407936199183038,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00031963445338699446,
      "loss": 4.9133,
      "mean_token_accuracy": 0.21837540417909623,
      "num_tokens": 98271743.0,
      "step": 56655
    },
    {
      "entropy": 5.9307169914245605,
      "epoch": 4.408325228554756,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031960728827692,
      "loss": 4.9321,
      "mean_token_accuracy": 0.2249053806066513,
      "num_tokens": 98280244.0,
      "step": 56660
    },
    {
      "entropy": 5.977306175231933,
      "epoch": 4.408714257926474,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003195801224900009,
      "loss": 5.0341,
      "mean_token_accuracy": 0.20814796090126036,
      "num_tokens": 98289005.0,
      "step": 56665
    },
    {
      "entropy": 6.040990781784058,
      "epoch": 4.409103287298191,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031955295602664936,
      "loss": 5.0367,
      "mean_token_accuracy": 0.21656417548656465,
      "num_tokens": 98297687.0,
      "step": 56670
    },
    {
      "entropy": 5.9367907524108885,
      "epoch": 4.409492316669908,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031952578888727763,
      "loss": 4.9183,
      "mean_token_accuracy": 0.22128846794366835,
      "num_tokens": 98306159.0,
      "step": 56675
    },
    {
      "entropy": 5.890282392501831,
      "epoch": 4.409881346041626,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031949862107229796,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21719149500131607,
      "num_tokens": 98315114.0,
      "step": 56680
    },
    {
      "entropy": 5.960938262939453,
      "epoch": 4.410270375413344,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003194714525821225,
      "loss": 5.069,
      "mean_token_accuracy": 0.20947894901037217,
      "num_tokens": 98323413.0,
      "step": 56685
    },
    {
      "entropy": 5.996547269821167,
      "epoch": 4.410659404785061,
      "grad_norm": 1.5,
      "learning_rate": 0.0003194442834171636,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2121077820658684,
      "num_tokens": 98331946.0,
      "step": 56690
    },
    {
      "entropy": 5.95749945640564,
      "epoch": 4.411048434156779,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003194171135778334,
      "loss": 4.9666,
      "mean_token_accuracy": 0.20684940367937088,
      "num_tokens": 98341583.0,
      "step": 56695
    },
    {
      "entropy": 5.907864570617676,
      "epoch": 4.411437463528497,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003193899430645442,
      "loss": 4.8977,
      "mean_token_accuracy": 0.21910208016633986,
      "num_tokens": 98350403.0,
      "step": 56700
    },
    {
      "entropy": 5.9756568431854244,
      "epoch": 4.411826492900214,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031936277187770824,
      "loss": 5.0842,
      "mean_token_accuracy": 0.2027921810746193,
      "num_tokens": 98358803.0,
      "step": 56705
    },
    {
      "entropy": 5.970907163619995,
      "epoch": 4.412215522271931,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003193356000177378,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21828947365283966,
      "num_tokens": 98367971.0,
      "step": 56710
    },
    {
      "entropy": 5.958186340332031,
      "epoch": 4.412604551643649,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003193084274850453,
      "loss": 4.9892,
      "mean_token_accuracy": 0.22026481926441194,
      "num_tokens": 98377036.0,
      "step": 56715
    },
    {
      "entropy": 5.920546531677246,
      "epoch": 4.412993581015367,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031928125428004286,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21383829265832902,
      "num_tokens": 98385424.0,
      "step": 56720
    },
    {
      "entropy": 5.996419906616211,
      "epoch": 4.413382610387084,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003192540804031429,
      "loss": 4.9876,
      "mean_token_accuracy": 0.21340801566839218,
      "num_tokens": 98393621.0,
      "step": 56725
    },
    {
      "entropy": 6.013441896438598,
      "epoch": 4.413771639758802,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003192269058547577,
      "loss": 5.1052,
      "mean_token_accuracy": 0.20797307640314103,
      "num_tokens": 98402171.0,
      "step": 56730
    },
    {
      "entropy": 5.9125657081604,
      "epoch": 4.41416066913052,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003191997306352996,
      "loss": 4.9472,
      "mean_token_accuracy": 0.21520679891109468,
      "num_tokens": 98411111.0,
      "step": 56735
    },
    {
      "entropy": 5.933580684661865,
      "epoch": 4.4145496985022366,
      "grad_norm": 1.671875,
      "learning_rate": 0.000319172554745181,
      "loss": 4.8984,
      "mean_token_accuracy": 0.22040425091981888,
      "num_tokens": 98419658.0,
      "step": 56740
    },
    {
      "entropy": 5.972822523117065,
      "epoch": 4.414938727873954,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031914537818481404,
      "loss": 5.0934,
      "mean_token_accuracy": 0.20754118114709855,
      "num_tokens": 98428110.0,
      "step": 56745
    },
    {
      "entropy": 5.914547348022461,
      "epoch": 4.415327757245672,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003191182009546113,
      "loss": 4.9313,
      "mean_token_accuracy": 0.21953604221343995,
      "num_tokens": 98436871.0,
      "step": 56750
    },
    {
      "entropy": 5.855015230178833,
      "epoch": 4.41571678661739,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031909102305498514,
      "loss": 4.8626,
      "mean_token_accuracy": 0.21681714355945586,
      "num_tokens": 98444968.0,
      "step": 56755
    },
    {
      "entropy": 5.9820740699768065,
      "epoch": 4.416105815989107,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031906384448634784,
      "loss": 5.0398,
      "mean_token_accuracy": 0.21327763199806213,
      "num_tokens": 98454067.0,
      "step": 56760
    },
    {
      "entropy": 5.994354724884033,
      "epoch": 4.416494845360825,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031903666524911185,
      "loss": 5.0589,
      "mean_token_accuracy": 0.21250736117362976,
      "num_tokens": 98463604.0,
      "step": 56765
    },
    {
      "entropy": 5.900160026550293,
      "epoch": 4.416883874732543,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031900948534368954,
      "loss": 4.938,
      "mean_token_accuracy": 0.22009223103523254,
      "num_tokens": 98472790.0,
      "step": 56770
    },
    {
      "entropy": 5.962891483306885,
      "epoch": 4.4172729041042595,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031898230477049334,
      "loss": 5.0011,
      "mean_token_accuracy": 0.21671175062656403,
      "num_tokens": 98482382.0,
      "step": 56775
    },
    {
      "entropy": 5.867442226409912,
      "epoch": 4.417661933475977,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003189551235299357,
      "loss": 4.8896,
      "mean_token_accuracy": 0.2217137634754181,
      "num_tokens": 98491362.0,
      "step": 56780
    },
    {
      "entropy": 5.953394269943237,
      "epoch": 4.418050962847695,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003189279416224289,
      "loss": 4.9707,
      "mean_token_accuracy": 0.21368285715579988,
      "num_tokens": 98499293.0,
      "step": 56785
    },
    {
      "entropy": 5.996138286590576,
      "epoch": 4.418439992219413,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00031890075904838556,
      "loss": 5.034,
      "mean_token_accuracy": 0.2090111717581749,
      "num_tokens": 98507665.0,
      "step": 56790
    },
    {
      "entropy": 5.905490779876709,
      "epoch": 4.41882902159113,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031887357580821817,
      "loss": 4.8926,
      "mean_token_accuracy": 0.22003644555807114,
      "num_tokens": 98516662.0,
      "step": 56795
    },
    {
      "entropy": 5.924647855758667,
      "epoch": 4.419218050962848,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031884639190233905,
      "loss": 4.9819,
      "mean_token_accuracy": 0.21059444546699524,
      "num_tokens": 98524642.0,
      "step": 56800
    },
    {
      "entropy": 6.00779504776001,
      "epoch": 4.419607080334565,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003188192073311608,
      "loss": 5.0256,
      "mean_token_accuracy": 0.2092769429087639,
      "num_tokens": 98532805.0,
      "step": 56805
    },
    {
      "entropy": 5.914027500152588,
      "epoch": 4.4199961097062825,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031879202209509564,
      "loss": 4.94,
      "mean_token_accuracy": 0.22320854365825654,
      "num_tokens": 98541479.0,
      "step": 56810
    },
    {
      "entropy": 5.892930412292481,
      "epoch": 4.420385139078,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031876483619455633,
      "loss": 5.0243,
      "mean_token_accuracy": 0.21446819901466369,
      "num_tokens": 98550208.0,
      "step": 56815
    },
    {
      "entropy": 5.975541782379151,
      "epoch": 4.420774168449718,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003187376496299553,
      "loss": 5.0158,
      "mean_token_accuracy": 0.20977468490600587,
      "num_tokens": 98558890.0,
      "step": 56820
    },
    {
      "entropy": 5.988870096206665,
      "epoch": 4.421163197821436,
      "grad_norm": 1.34375,
      "learning_rate": 0.000318710462401705,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21382893323898317,
      "num_tokens": 98567269.0,
      "step": 56825
    },
    {
      "entropy": 6.020040273666382,
      "epoch": 4.421552227193153,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031868327451021816,
      "loss": 5.0745,
      "mean_token_accuracy": 0.2088913083076477,
      "num_tokens": 98576554.0,
      "step": 56830
    },
    {
      "entropy": 5.915755128860473,
      "epoch": 4.421941256564871,
      "grad_norm": 1.46875,
      "learning_rate": 0.00031865608595590706,
      "loss": 4.9614,
      "mean_token_accuracy": 0.21389412581920625,
      "num_tokens": 98584667.0,
      "step": 56835
    },
    {
      "entropy": 5.947736597061157,
      "epoch": 4.422330285936588,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031862889673918437,
      "loss": 4.9894,
      "mean_token_accuracy": 0.2134447604417801,
      "num_tokens": 98592976.0,
      "step": 56840
    },
    {
      "entropy": 5.942670202255249,
      "epoch": 4.4227193153083055,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003186017068604626,
      "loss": 4.9317,
      "mean_token_accuracy": 0.22419904321432113,
      "num_tokens": 98601406.0,
      "step": 56845
    },
    {
      "entropy": 5.9527791976928714,
      "epoch": 4.423108344680023,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003185745163201543,
      "loss": 4.9511,
      "mean_token_accuracy": 0.21429968178272246,
      "num_tokens": 98609797.0,
      "step": 56850
    },
    {
      "entropy": 5.862278652191162,
      "epoch": 4.423497374051741,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003185473251186721,
      "loss": 4.9305,
      "mean_token_accuracy": 0.217645063996315,
      "num_tokens": 98618428.0,
      "step": 56855
    },
    {
      "entropy": 5.8954509735107425,
      "epoch": 4.423886403423459,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003185201332564285,
      "loss": 4.9125,
      "mean_token_accuracy": 0.21673950403928757,
      "num_tokens": 98627294.0,
      "step": 56860
    },
    {
      "entropy": 5.943177127838135,
      "epoch": 4.424275432795176,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003184929407338362,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2123803973197937,
      "num_tokens": 98636721.0,
      "step": 56865
    },
    {
      "entropy": 5.9382078647613525,
      "epoch": 4.424664462166893,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00031846574755130773,
      "loss": 4.9457,
      "mean_token_accuracy": 0.22137642353773118,
      "num_tokens": 98645325.0,
      "step": 56870
    },
    {
      "entropy": 5.897609233856201,
      "epoch": 4.425053491538611,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00031843855370925574,
      "loss": 4.9061,
      "mean_token_accuracy": 0.22479104101657868,
      "num_tokens": 98653524.0,
      "step": 56875
    },
    {
      "entropy": 5.9612624168396,
      "epoch": 4.4254425209103285,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00031841135920809285,
      "loss": 4.9519,
      "mean_token_accuracy": 0.22124587893486022,
      "num_tokens": 98661483.0,
      "step": 56880
    },
    {
      "entropy": 5.9634778022766115,
      "epoch": 4.425831550282046,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003183841640482316,
      "loss": 5.013,
      "mean_token_accuracy": 0.21130256354808807,
      "num_tokens": 98670202.0,
      "step": 56885
    },
    {
      "entropy": 5.93214864730835,
      "epoch": 4.426220579653764,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031835696823008476,
      "loss": 4.9811,
      "mean_token_accuracy": 0.2149977296590805,
      "num_tokens": 98678632.0,
      "step": 56890
    },
    {
      "entropy": 5.975233268737793,
      "epoch": 4.426609609025482,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031832977175406496,
      "loss": 4.9963,
      "mean_token_accuracy": 0.2107300192117691,
      "num_tokens": 98686908.0,
      "step": 56895
    },
    {
      "entropy": 5.98721604347229,
      "epoch": 4.426998638397199,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003183025746205848,
      "loss": 5.0175,
      "mean_token_accuracy": 0.2167464703321457,
      "num_tokens": 98694954.0,
      "step": 56900
    },
    {
      "entropy": 5.919262647628784,
      "epoch": 4.427387667768917,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003182753768300571,
      "loss": 4.9186,
      "mean_token_accuracy": 0.22369845509529113,
      "num_tokens": 98703813.0,
      "step": 56905
    },
    {
      "entropy": 5.919897127151489,
      "epoch": 4.427776697140634,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003182481783828943,
      "loss": 4.9353,
      "mean_token_accuracy": 0.21661010086536409,
      "num_tokens": 98712198.0,
      "step": 56910
    },
    {
      "entropy": 6.01188383102417,
      "epoch": 4.4281657265123515,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031822097927950934,
      "loss": 5.015,
      "mean_token_accuracy": 0.21164233684539796,
      "num_tokens": 98721693.0,
      "step": 56915
    },
    {
      "entropy": 6.019493103027344,
      "epoch": 4.428554755884069,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031819377952031476,
      "loss": 5.0398,
      "mean_token_accuracy": 0.20627864301204682,
      "num_tokens": 98730238.0,
      "step": 56920
    },
    {
      "entropy": 5.92161226272583,
      "epoch": 4.428943785255787,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031816657910572327,
      "loss": 4.9391,
      "mean_token_accuracy": 0.21502457559108734,
      "num_tokens": 98739298.0,
      "step": 56925
    },
    {
      "entropy": 6.040129518508911,
      "epoch": 4.429332814627505,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00031813937803614774,
      "loss": 5.0714,
      "mean_token_accuracy": 0.2099572166800499,
      "num_tokens": 98748862.0,
      "step": 56930
    },
    {
      "entropy": 5.878735065460205,
      "epoch": 4.429721843999222,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003181121763120008,
      "loss": 4.8874,
      "mean_token_accuracy": 0.2254655569791794,
      "num_tokens": 98757445.0,
      "step": 56935
    },
    {
      "entropy": 5.961406612396241,
      "epoch": 4.430110873370939,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0003180849739336952,
      "loss": 4.9945,
      "mean_token_accuracy": 0.21152057200670243,
      "num_tokens": 98765856.0,
      "step": 56940
    },
    {
      "entropy": 5.93566484451294,
      "epoch": 4.430499902742657,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031805777090164364,
      "loss": 5.0282,
      "mean_token_accuracy": 0.21019337028265,
      "num_tokens": 98774819.0,
      "step": 56945
    },
    {
      "entropy": 5.983921051025391,
      "epoch": 4.4308889321143745,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000318030567216259,
      "loss": 4.9376,
      "mean_token_accuracy": 0.21509525775909424,
      "num_tokens": 98783532.0,
      "step": 56950
    },
    {
      "entropy": 5.975806999206543,
      "epoch": 4.431277961486092,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031800336287795395,
      "loss": 5.0368,
      "mean_token_accuracy": 0.20736061930656433,
      "num_tokens": 98792062.0,
      "step": 56955
    },
    {
      "entropy": 5.860547161102295,
      "epoch": 4.43166699085781,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031797615788714133,
      "loss": 4.8754,
      "mean_token_accuracy": 0.21118629723787308,
      "num_tokens": 98800558.0,
      "step": 56960
    },
    {
      "entropy": 5.884994125366211,
      "epoch": 4.432056020229528,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031794895224423396,
      "loss": 4.873,
      "mean_token_accuracy": 0.22376664727926254,
      "num_tokens": 98808612.0,
      "step": 56965
    },
    {
      "entropy": 5.9277924537658695,
      "epoch": 4.432445049601245,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003179217459496446,
      "loss": 4.9079,
      "mean_token_accuracy": 0.22111464589834212,
      "num_tokens": 98817077.0,
      "step": 56970
    },
    {
      "entropy": 5.839706897735596,
      "epoch": 4.432834078972962,
      "grad_norm": 1.421875,
      "learning_rate": 0.000317894539003786,
      "loss": 4.8691,
      "mean_token_accuracy": 0.22283138632774352,
      "num_tokens": 98825506.0,
      "step": 56975
    },
    {
      "entropy": 5.872729969024658,
      "epoch": 4.43322310834468,
      "grad_norm": 1.421875,
      "learning_rate": 0.000317867331407071,
      "loss": 4.9182,
      "mean_token_accuracy": 0.21669227927923201,
      "num_tokens": 98834112.0,
      "step": 56980
    },
    {
      "entropy": 5.965631628036499,
      "epoch": 4.4336121377163975,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003178401231599126,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21779697835445405,
      "num_tokens": 98843933.0,
      "step": 56985
    },
    {
      "entropy": 5.82295789718628,
      "epoch": 4.434001167088115,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031781291426272347,
      "loss": 4.8536,
      "mean_token_accuracy": 0.21710425466299058,
      "num_tokens": 98852646.0,
      "step": 56990
    },
    {
      "entropy": 5.904745721817017,
      "epoch": 4.434390196459833,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003177857047159165,
      "loss": 4.8556,
      "mean_token_accuracy": 0.23176491558551787,
      "num_tokens": 98860819.0,
      "step": 56995
    },
    {
      "entropy": 5.949557113647461,
      "epoch": 4.434779225831551,
      "grad_norm": 1.5,
      "learning_rate": 0.0003177584945199046,
      "loss": 5.0078,
      "mean_token_accuracy": 0.220664881169796,
      "num_tokens": 98869000.0,
      "step": 57000
    },
    {
      "epoch": 4.434779225831551,
      "eval_entropy": 5.618273828486276,
      "eval_loss": 5.064419746398926,
      "eval_mean_token_accuracy": 0.2197029515900098,
      "eval_num_tokens": 98869000.0,
      "eval_runtime": 21.6322,
      "eval_samples_per_second": 1921.116,
      "eval_steps_per_second": 240.151,
      "step": 57000
    },
    {
      "entropy": 5.905194711685181,
      "epoch": 4.435168255203267,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003177312836751005,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22685130983591079,
      "num_tokens": 98877174.0,
      "step": 57005
    },
    {
      "entropy": 5.896127843856812,
      "epoch": 4.435557284574985,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031770407218191727,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21606757640838622,
      "num_tokens": 98885934.0,
      "step": 57010
    },
    {
      "entropy": 5.87311053276062,
      "epoch": 4.435946313946703,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003176768600407677,
      "loss": 4.932,
      "mean_token_accuracy": 0.21097803711891175,
      "num_tokens": 98894824.0,
      "step": 57015
    },
    {
      "entropy": 5.970147895812988,
      "epoch": 4.4363353433184205,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003176496472520647,
      "loss": 4.9049,
      "mean_token_accuracy": 0.2207731530070305,
      "num_tokens": 98903173.0,
      "step": 57020
    },
    {
      "entropy": 5.927363061904908,
      "epoch": 4.436724372690138,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003176224338162212,
      "loss": 4.9602,
      "mean_token_accuracy": 0.2160036563873291,
      "num_tokens": 98912730.0,
      "step": 57025
    },
    {
      "entropy": 5.933333253860473,
      "epoch": 4.437113402061856,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031759521973365006,
      "loss": 5.0698,
      "mean_token_accuracy": 0.20470034182071686,
      "num_tokens": 98922182.0,
      "step": 57030
    },
    {
      "entropy": 5.9016188621521,
      "epoch": 4.437502431433574,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031756800500476435,
      "loss": 4.8616,
      "mean_token_accuracy": 0.22405731528997422,
      "num_tokens": 98930243.0,
      "step": 57035
    },
    {
      "entropy": 5.958464908599853,
      "epoch": 4.43789146080529,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031754078962997686,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21757351905107497,
      "num_tokens": 98938281.0,
      "step": 57040
    },
    {
      "entropy": 5.875172567367554,
      "epoch": 4.438280490177008,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003175135736097006,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21191619336605072,
      "num_tokens": 98946960.0,
      "step": 57045
    },
    {
      "entropy": 5.932411098480225,
      "epoch": 4.438669519548726,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031748635694434844,
      "loss": 4.8835,
      "mean_token_accuracy": 0.2192670226097107,
      "num_tokens": 98955322.0,
      "step": 57050
    },
    {
      "entropy": 5.945613670349121,
      "epoch": 4.4390585489204435,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003174591396343336,
      "loss": 5.0225,
      "mean_token_accuracy": 0.2130824714899063,
      "num_tokens": 98964517.0,
      "step": 57055
    },
    {
      "entropy": 5.975570154190064,
      "epoch": 4.439447578292161,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003174319216800688,
      "loss": 4.9784,
      "mean_token_accuracy": 0.21378668248653412,
      "num_tokens": 98973728.0,
      "step": 57060
    },
    {
      "entropy": 5.874374675750732,
      "epoch": 4.439836607663879,
      "grad_norm": 1.546875,
      "learning_rate": 0.00031740470308196724,
      "loss": 4.9535,
      "mean_token_accuracy": 0.2149740934371948,
      "num_tokens": 98981799.0,
      "step": 57065
    },
    {
      "entropy": 6.001570415496826,
      "epoch": 4.440225637035597,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003173774838404417,
      "loss": 4.9692,
      "mean_token_accuracy": 0.21336007118225098,
      "num_tokens": 98990929.0,
      "step": 57070
    },
    {
      "entropy": 5.973786735534668,
      "epoch": 4.440614666407313,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031735026395590535,
      "loss": 4.8956,
      "mean_token_accuracy": 0.22144406884908677,
      "num_tokens": 98998676.0,
      "step": 57075
    },
    {
      "entropy": 5.894015121459961,
      "epoch": 4.441003695779031,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000317323043428771,
      "loss": 4.8994,
      "mean_token_accuracy": 0.21539787352085113,
      "num_tokens": 99007557.0,
      "step": 57080
    },
    {
      "entropy": 5.914684152603149,
      "epoch": 4.441392725150749,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003172958222594519,
      "loss": 5.0115,
      "mean_token_accuracy": 0.21123959571123124,
      "num_tokens": 99016274.0,
      "step": 57085
    },
    {
      "entropy": 5.9649683952331545,
      "epoch": 4.4417817545224665,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003172686004483611,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21677168607711791,
      "num_tokens": 99024307.0,
      "step": 57090
    },
    {
      "entropy": 5.994546937942505,
      "epoch": 4.442170783894184,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031724137799591154,
      "loss": 5.067,
      "mean_token_accuracy": 0.20798503160476683,
      "num_tokens": 99033295.0,
      "step": 57095
    },
    {
      "entropy": 5.979236268997193,
      "epoch": 4.442559813265902,
      "grad_norm": 1.46875,
      "learning_rate": 0.00031721415490251634,
      "loss": 4.995,
      "mean_token_accuracy": 0.2075594261288643,
      "num_tokens": 99042028.0,
      "step": 57100
    },
    {
      "entropy": 5.925299263000488,
      "epoch": 4.44294884263762,
      "grad_norm": 1.375,
      "learning_rate": 0.0003171869311685885,
      "loss": 4.9342,
      "mean_token_accuracy": 0.21608500480651854,
      "num_tokens": 99050915.0,
      "step": 57105
    },
    {
      "entropy": 5.978424787521362,
      "epoch": 4.443337872009336,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003171597067945411,
      "loss": 4.9417,
      "mean_token_accuracy": 0.21003836244344712,
      "num_tokens": 99059474.0,
      "step": 57110
    },
    {
      "entropy": 5.867600440979004,
      "epoch": 4.443726901381054,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031713248178078717,
      "loss": 4.8383,
      "mean_token_accuracy": 0.22539416700601578,
      "num_tokens": 99068068.0,
      "step": 57115
    },
    {
      "entropy": 5.934176874160767,
      "epoch": 4.444115930752772,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031710525612774,
      "loss": 5.0635,
      "mean_token_accuracy": 0.20957596600055695,
      "num_tokens": 99077158.0,
      "step": 57120
    },
    {
      "entropy": 5.936445856094361,
      "epoch": 4.4445049601244895,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003170780298358126,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21439535766839982,
      "num_tokens": 99085270.0,
      "step": 57125
    },
    {
      "entropy": 5.980767869949341,
      "epoch": 4.444893989496207,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031705080290541797,
      "loss": 4.9804,
      "mean_token_accuracy": 0.2070990025997162,
      "num_tokens": 99094420.0,
      "step": 57130
    },
    {
      "entropy": 5.921824169158936,
      "epoch": 4.445283018867925,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031702357533696946,
      "loss": 4.9331,
      "mean_token_accuracy": 0.22231628000736237,
      "num_tokens": 99103399.0,
      "step": 57135
    },
    {
      "entropy": 5.825370931625367,
      "epoch": 4.445672048239642,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031699634713088,
      "loss": 4.7973,
      "mean_token_accuracy": 0.22457805275917053,
      "num_tokens": 99112417.0,
      "step": 57140
    },
    {
      "entropy": 5.909503602981568,
      "epoch": 4.446061077611359,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003169691182875628,
      "loss": 4.9323,
      "mean_token_accuracy": 0.21884176433086394,
      "num_tokens": 99120823.0,
      "step": 57145
    },
    {
      "entropy": 5.9147392272949215,
      "epoch": 4.446450106983077,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003169418888074311,
      "loss": 4.9204,
      "mean_token_accuracy": 0.2151761993765831,
      "num_tokens": 99129154.0,
      "step": 57150
    },
    {
      "entropy": 5.953570604324341,
      "epoch": 4.446839136354795,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031691465869089795,
      "loss": 4.9669,
      "mean_token_accuracy": 0.21604932248592376,
      "num_tokens": 99137294.0,
      "step": 57155
    },
    {
      "entropy": 5.945612049102783,
      "epoch": 4.4472281657265125,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003168874279383766,
      "loss": 5.0579,
      "mean_token_accuracy": 0.20919550210237503,
      "num_tokens": 99146219.0,
      "step": 57160
    },
    {
      "entropy": 5.935383558273315,
      "epoch": 4.44761719509823,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003168601965502802,
      "loss": 4.9913,
      "mean_token_accuracy": 0.2080627426505089,
      "num_tokens": 99154656.0,
      "step": 57165
    },
    {
      "entropy": 5.9690492153167725,
      "epoch": 4.448006224469948,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003168329645270219,
      "loss": 4.9603,
      "mean_token_accuracy": 0.21847521811723708,
      "num_tokens": 99163384.0,
      "step": 57170
    },
    {
      "entropy": 5.894696760177612,
      "epoch": 4.448395253841665,
      "grad_norm": 1.453125,
      "learning_rate": 0.000316805731869015,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21736711114645005,
      "num_tokens": 99172237.0,
      "step": 57175
    },
    {
      "entropy": 5.9461627960205075,
      "epoch": 4.448784283213382,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031677849857667266,
      "loss": 5.0726,
      "mean_token_accuracy": 0.21444928795099258,
      "num_tokens": 99181418.0,
      "step": 57180
    },
    {
      "entropy": 5.954098224639893,
      "epoch": 4.4491733125851,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003167512646504081,
      "loss": 4.9624,
      "mean_token_accuracy": 0.21858360916376113,
      "num_tokens": 99190683.0,
      "step": 57185
    },
    {
      "entropy": 5.88373703956604,
      "epoch": 4.449562341956818,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003167240300906345,
      "loss": 4.8663,
      "mean_token_accuracy": 0.2321986198425293,
      "num_tokens": 99199284.0,
      "step": 57190
    },
    {
      "entropy": 5.9756935119628904,
      "epoch": 4.4499513713285355,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003166967948977652,
      "loss": 5.1388,
      "mean_token_accuracy": 0.20770090222358703,
      "num_tokens": 99207616.0,
      "step": 57195
    },
    {
      "entropy": 5.93669958114624,
      "epoch": 4.450340400700253,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003166695590722134,
      "loss": 4.9407,
      "mean_token_accuracy": 0.21762867122888566,
      "num_tokens": 99216329.0,
      "step": 57200
    },
    {
      "entropy": 5.973982906341552,
      "epoch": 4.45072943007197,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031664232261439235,
      "loss": 4.9913,
      "mean_token_accuracy": 0.21524061560630797,
      "num_tokens": 99225161.0,
      "step": 57205
    },
    {
      "entropy": 5.9675816059112545,
      "epoch": 4.451118459443688,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003166150855247153,
      "loss": 5.008,
      "mean_token_accuracy": 0.20122611075639724,
      "num_tokens": 99234205.0,
      "step": 57210
    },
    {
      "entropy": 5.886396360397339,
      "epoch": 4.451507488815405,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003165878478035956,
      "loss": 4.94,
      "mean_token_accuracy": 0.22127435952425004,
      "num_tokens": 99243533.0,
      "step": 57215
    },
    {
      "entropy": 5.983232688903809,
      "epoch": 4.451896518187123,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003165606094514465,
      "loss": 4.9891,
      "mean_token_accuracy": 0.21591623723506928,
      "num_tokens": 99252563.0,
      "step": 57220
    },
    {
      "entropy": 5.925714063644409,
      "epoch": 4.452285547558841,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00031653337046868137,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21331370919942855,
      "num_tokens": 99260845.0,
      "step": 57225
    },
    {
      "entropy": 5.929655313491821,
      "epoch": 4.4526745769305585,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00031650613085571334,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22461340129375457,
      "num_tokens": 99269309.0,
      "step": 57230
    },
    {
      "entropy": 5.88968186378479,
      "epoch": 4.453063606302276,
      "grad_norm": 1.375,
      "learning_rate": 0.0003164788906129559,
      "loss": 4.9758,
      "mean_token_accuracy": 0.21726243197917938,
      "num_tokens": 99277659.0,
      "step": 57235
    },
    {
      "entropy": 5.906838607788086,
      "epoch": 4.453452635673994,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003164516497408223,
      "loss": 5.0225,
      "mean_token_accuracy": 0.2136093035340309,
      "num_tokens": 99286422.0,
      "step": 57240
    },
    {
      "entropy": 6.032088804244995,
      "epoch": 4.453841665045711,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003164244082397259,
      "loss": 5.0318,
      "mean_token_accuracy": 0.21278442740440368,
      "num_tokens": 99295140.0,
      "step": 57245
    },
    {
      "entropy": 5.963583087921142,
      "epoch": 4.454230694417428,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00031639716611008,
      "loss": 5.0092,
      "mean_token_accuracy": 0.2104533240199089,
      "num_tokens": 99304180.0,
      "step": 57250
    },
    {
      "entropy": 5.928808879852295,
      "epoch": 4.454619723789146,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003163699233522981,
      "loss": 4.9754,
      "mean_token_accuracy": 0.2114589810371399,
      "num_tokens": 99312765.0,
      "step": 57255
    },
    {
      "entropy": 5.912766981124878,
      "epoch": 4.455008753160864,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003163426799667934,
      "loss": 4.9648,
      "mean_token_accuracy": 0.21090328693389893,
      "num_tokens": 99321530.0,
      "step": 57260
    },
    {
      "entropy": 5.91319842338562,
      "epoch": 4.4553977825325815,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003163154359539793,
      "loss": 4.9833,
      "mean_token_accuracy": 0.20515942573547363,
      "num_tokens": 99329868.0,
      "step": 57265
    },
    {
      "entropy": 5.998158311843872,
      "epoch": 4.455786811904299,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031628819131426924,
      "loss": 4.9929,
      "mean_token_accuracy": 0.20849019587039946,
      "num_tokens": 99338780.0,
      "step": 57270
    },
    {
      "entropy": 5.943441390991211,
      "epoch": 4.456175841276016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003162609460480766,
      "loss": 4.913,
      "mean_token_accuracy": 0.21866649240255356,
      "num_tokens": 99348159.0,
      "step": 57275
    },
    {
      "entropy": 5.953143072128296,
      "epoch": 4.456564870647734,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003162337001558147,
      "loss": 4.9126,
      "mean_token_accuracy": 0.2165081098675728,
      "num_tokens": 99356136.0,
      "step": 57280
    },
    {
      "entropy": 5.995149993896485,
      "epoch": 4.456953900019451,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031620645363789713,
      "loss": 5.0375,
      "mean_token_accuracy": 0.20943720787763595,
      "num_tokens": 99364537.0,
      "step": 57285
    },
    {
      "entropy": 5.913239288330078,
      "epoch": 4.457342929391169,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003161792064947372,
      "loss": 4.8932,
      "mean_token_accuracy": 0.2224335864186287,
      "num_tokens": 99374263.0,
      "step": 57290
    },
    {
      "entropy": 5.969743251800537,
      "epoch": 4.457731958762887,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031615195872674836,
      "loss": 5.037,
      "mean_token_accuracy": 0.21639725118875502,
      "num_tokens": 99382534.0,
      "step": 57295
    },
    {
      "entropy": 5.897247552871704,
      "epoch": 4.4581209881346044,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031612471033434406,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2133727878332138,
      "num_tokens": 99390665.0,
      "step": 57300
    },
    {
      "entropy": 5.924976348876953,
      "epoch": 4.458510017506322,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003160974613179377,
      "loss": 4.9048,
      "mean_token_accuracy": 0.21887648850679398,
      "num_tokens": 99399700.0,
      "step": 57305
    },
    {
      "entropy": 6.004235458374024,
      "epoch": 4.458899046878039,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003160702116779428,
      "loss": 4.9828,
      "mean_token_accuracy": 0.20929044932127,
      "num_tokens": 99407505.0,
      "step": 57310
    },
    {
      "entropy": 5.9128087043762205,
      "epoch": 4.459288076249757,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003160429614147727,
      "loss": 4.9033,
      "mean_token_accuracy": 0.22038080394268036,
      "num_tokens": 99416454.0,
      "step": 57315
    },
    {
      "entropy": 5.882433223724365,
      "epoch": 4.459677105621474,
      "grad_norm": 1.5625,
      "learning_rate": 0.00031601571052884117,
      "loss": 4.9388,
      "mean_token_accuracy": 0.22060583531856537,
      "num_tokens": 99425134.0,
      "step": 57320
    },
    {
      "entropy": 5.918792247772217,
      "epoch": 4.460066134993192,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003159884590205613,
      "loss": 4.9429,
      "mean_token_accuracy": 0.21820741444826125,
      "num_tokens": 99433061.0,
      "step": 57325
    },
    {
      "entropy": 5.997221755981445,
      "epoch": 4.46045516436491,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000315961206890347,
      "loss": 4.9871,
      "mean_token_accuracy": 0.22209499627351761,
      "num_tokens": 99441814.0,
      "step": 57330
    },
    {
      "entropy": 5.988689422607422,
      "epoch": 4.460844193736627,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003159339541386115,
      "loss": 5.0239,
      "mean_token_accuracy": 0.21653041243553162,
      "num_tokens": 99450430.0,
      "step": 57335
    },
    {
      "entropy": 5.963972568511963,
      "epoch": 4.461233223108344,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031590670076576846,
      "loss": 5.0294,
      "mean_token_accuracy": 0.21142186969518661,
      "num_tokens": 99459408.0,
      "step": 57340
    },
    {
      "entropy": 5.95520396232605,
      "epoch": 4.461622252480062,
      "grad_norm": 1.5,
      "learning_rate": 0.00031587944677223133,
      "loss": 4.9078,
      "mean_token_accuracy": 0.2208254024386406,
      "num_tokens": 99468014.0,
      "step": 57345
    },
    {
      "entropy": 5.9869224548339846,
      "epoch": 4.46201128185178,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003158521921584136,
      "loss": 4.9875,
      "mean_token_accuracy": 0.22130175232887267,
      "num_tokens": 99476245.0,
      "step": 57350
    },
    {
      "entropy": 5.959428834915161,
      "epoch": 4.462400311223497,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00031582493692472886,
      "loss": 4.9144,
      "mean_token_accuracy": 0.2221628397703171,
      "num_tokens": 99484390.0,
      "step": 57355
    },
    {
      "entropy": 5.852032423019409,
      "epoch": 4.462789340595215,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003157976810715907,
      "loss": 4.8921,
      "mean_token_accuracy": 0.2217566654086113,
      "num_tokens": 99492885.0,
      "step": 57360
    },
    {
      "entropy": 5.941972541809082,
      "epoch": 4.463178369966933,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031577042459941275,
      "loss": 4.9897,
      "mean_token_accuracy": 0.21386562734842302,
      "num_tokens": 99501669.0,
      "step": 57365
    },
    {
      "entropy": 5.945519256591797,
      "epoch": 4.46356739933865,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031574316750860844,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2168682858347893,
      "num_tokens": 99509992.0,
      "step": 57370
    },
    {
      "entropy": 5.948888826370239,
      "epoch": 4.463956428710367,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003157159097995914,
      "loss": 4.9975,
      "mean_token_accuracy": 0.21237008571624755,
      "num_tokens": 99518867.0,
      "step": 57375
    },
    {
      "entropy": 5.909712076187134,
      "epoch": 4.464345458082085,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031568865147277525,
      "loss": 4.9665,
      "mean_token_accuracy": 0.21459220051765443,
      "num_tokens": 99528070.0,
      "step": 57380
    },
    {
      "entropy": 5.903021335601807,
      "epoch": 4.464734487453803,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031566139252857357,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2213527098298073,
      "num_tokens": 99536386.0,
      "step": 57385
    },
    {
      "entropy": 5.887178659439087,
      "epoch": 4.46512351682552,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003156341329674,
      "loss": 4.8943,
      "mean_token_accuracy": 0.2162710890173912,
      "num_tokens": 99544519.0,
      "step": 57390
    },
    {
      "entropy": 5.921792507171631,
      "epoch": 4.465512546197238,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003156068727896681,
      "loss": 5.006,
      "mean_token_accuracy": 0.21417365372180938,
      "num_tokens": 99553332.0,
      "step": 57395
    },
    {
      "entropy": 5.9782267093658445,
      "epoch": 4.465901575568956,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003155796119957915,
      "loss": 5.0279,
      "mean_token_accuracy": 0.2073320358991623,
      "num_tokens": 99561423.0,
      "step": 57400
    },
    {
      "entropy": 5.934919786453247,
      "epoch": 4.466290604940673,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031555235058618403,
      "loss": 4.923,
      "mean_token_accuracy": 0.21565380543470383,
      "num_tokens": 99570008.0,
      "step": 57405
    },
    {
      "entropy": 5.941976022720337,
      "epoch": 4.46667963431239,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00031552508856125906,
      "loss": 5.0435,
      "mean_token_accuracy": 0.2091601863503456,
      "num_tokens": 99579351.0,
      "step": 57410
    },
    {
      "entropy": 6.035035800933838,
      "epoch": 4.467068663684108,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003154978259214304,
      "loss": 5.085,
      "mean_token_accuracy": 0.19620254635810852,
      "num_tokens": 99588431.0,
      "step": 57415
    },
    {
      "entropy": 6.02313003540039,
      "epoch": 4.467457693055826,
      "grad_norm": 1.515625,
      "learning_rate": 0.00031547056266711173,
      "loss": 5.0021,
      "mean_token_accuracy": 0.2127077177166939,
      "num_tokens": 99596982.0,
      "step": 57420
    },
    {
      "entropy": 5.94403338432312,
      "epoch": 4.467846722427543,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003154432987987166,
      "loss": 5.0183,
      "mean_token_accuracy": 0.21316463649272918,
      "num_tokens": 99606315.0,
      "step": 57425
    },
    {
      "entropy": 5.988648700714111,
      "epoch": 4.468235751799261,
      "grad_norm": 1.5,
      "learning_rate": 0.0003154160343166589,
      "loss": 5.02,
      "mean_token_accuracy": 0.20944138765335082,
      "num_tokens": 99614617.0,
      "step": 57430
    },
    {
      "entropy": 6.008930635452271,
      "epoch": 4.468624781170979,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00031538876922135214,
      "loss": 5.0384,
      "mean_token_accuracy": 0.2118157386779785,
      "num_tokens": 99623369.0,
      "step": 57435
    },
    {
      "entropy": 5.9637336254119875,
      "epoch": 4.469013810542696,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003153615035132101,
      "loss": 4.9671,
      "mean_token_accuracy": 0.21535481810569762,
      "num_tokens": 99632859.0,
      "step": 57440
    },
    {
      "entropy": 5.878107261657715,
      "epoch": 4.469402839914413,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031533423719264645,
      "loss": 4.8325,
      "mean_token_accuracy": 0.23468345999717713,
      "num_tokens": 99641809.0,
      "step": 57445
    },
    {
      "entropy": 5.954575872421264,
      "epoch": 4.469791869286131,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000315306970260075,
      "loss": 4.9698,
      "mean_token_accuracy": 0.2241729184985161,
      "num_tokens": 99649335.0,
      "step": 57450
    },
    {
      "entropy": 5.902673864364624,
      "epoch": 4.470180898657849,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003152797027159094,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21339701861143112,
      "num_tokens": 99657961.0,
      "step": 57455
    },
    {
      "entropy": 5.903442430496216,
      "epoch": 4.470569928029566,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031525243456056345,
      "loss": 4.9231,
      "mean_token_accuracy": 0.22357354313135147,
      "num_tokens": 99667148.0,
      "step": 57460
    },
    {
      "entropy": 5.905039548873901,
      "epoch": 4.470958957401284,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003152251657944509,
      "loss": 4.9257,
      "mean_token_accuracy": 0.22159579694271087,
      "num_tokens": 99675110.0,
      "step": 57465
    },
    {
      "entropy": 5.896014785766601,
      "epoch": 4.471347986773002,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003151978964179854,
      "loss": 4.9484,
      "mean_token_accuracy": 0.219231516122818,
      "num_tokens": 99683173.0,
      "step": 57470
    },
    {
      "entropy": 5.9502325534820555,
      "epoch": 4.4717370161447185,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031517062643158085,
      "loss": 4.9945,
      "mean_token_accuracy": 0.21580644249916076,
      "num_tokens": 99691773.0,
      "step": 57475
    },
    {
      "entropy": 5.9457601547241214,
      "epoch": 4.472126045516436,
      "grad_norm": 1.46875,
      "learning_rate": 0.00031514335583565097,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2129787191748619,
      "num_tokens": 99699414.0,
      "step": 57480
    },
    {
      "entropy": 5.948827505111694,
      "epoch": 4.472515074888154,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031511608463060954,
      "loss": 4.9711,
      "mean_token_accuracy": 0.21007128208875656,
      "num_tokens": 99708343.0,
      "step": 57485
    },
    {
      "entropy": 5.959079456329346,
      "epoch": 4.472904104259872,
      "grad_norm": 1.515625,
      "learning_rate": 0.00031508881281687044,
      "loss": 5.0116,
      "mean_token_accuracy": 0.21384498924016954,
      "num_tokens": 99716393.0,
      "step": 57490
    },
    {
      "entropy": 5.821180582046509,
      "epoch": 4.473293133631589,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031506154039484736,
      "loss": 4.8045,
      "mean_token_accuracy": 0.22567864656448364,
      "num_tokens": 99724741.0,
      "step": 57495
    },
    {
      "entropy": 5.919459819793701,
      "epoch": 4.473682163003307,
      "grad_norm": 1.375,
      "learning_rate": 0.00031503426736495424,
      "loss": 4.9739,
      "mean_token_accuracy": 0.21527534276247023,
      "num_tokens": 99733429.0,
      "step": 57500
    },
    {
      "entropy": 5.95928840637207,
      "epoch": 4.474071192375025,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031500699372760476,
      "loss": 5.0005,
      "mean_token_accuracy": 0.20812428295612334,
      "num_tokens": 99742027.0,
      "step": 57505
    },
    {
      "entropy": 5.9797920227050785,
      "epoch": 4.4744602217467415,
      "grad_norm": 1.5,
      "learning_rate": 0.00031497971948321287,
      "loss": 4.9858,
      "mean_token_accuracy": 0.2149359717965126,
      "num_tokens": 99751375.0,
      "step": 57510
    },
    {
      "entropy": 5.948651266098023,
      "epoch": 4.474849251118459,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003149524446321923,
      "loss": 4.9375,
      "mean_token_accuracy": 0.21439621597528458,
      "num_tokens": 99759980.0,
      "step": 57515
    },
    {
      "entropy": 5.938612794876098,
      "epoch": 4.475238280490177,
      "grad_norm": 1.453125,
      "learning_rate": 0.000314925169174957,
      "loss": 4.9548,
      "mean_token_accuracy": 0.21585564464330673,
      "num_tokens": 99768570.0,
      "step": 57520
    },
    {
      "entropy": 5.962278413772583,
      "epoch": 4.475627309861895,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031489789311192086,
      "loss": 5.0441,
      "mean_token_accuracy": 0.20781535059213638,
      "num_tokens": 99777843.0,
      "step": 57525
    },
    {
      "entropy": 5.953825759887695,
      "epoch": 4.476016339233612,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003148706164434977,
      "loss": 4.9276,
      "mean_token_accuracy": 0.21425359696149826,
      "num_tokens": 99786873.0,
      "step": 57530
    },
    {
      "entropy": 5.956406593322754,
      "epoch": 4.47640536860533,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003148433391701014,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21408039331436157,
      "num_tokens": 99795854.0,
      "step": 57535
    },
    {
      "entropy": 5.943245077133179,
      "epoch": 4.476794397977047,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003148160612921457,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20961966812610627,
      "num_tokens": 99804927.0,
      "step": 57540
    },
    {
      "entropy": 6.021684980392456,
      "epoch": 4.4771834273487645,
      "grad_norm": 1.5625,
      "learning_rate": 0.00031478878281004474,
      "loss": 5.0714,
      "mean_token_accuracy": 0.21096121072769164,
      "num_tokens": 99812859.0,
      "step": 57545
    },
    {
      "entropy": 5.922207355499268,
      "epoch": 4.477572456720482,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00031476150372421234,
      "loss": 4.9428,
      "mean_token_accuracy": 0.21480560451745986,
      "num_tokens": 99821540.0,
      "step": 57550
    },
    {
      "entropy": 5.903932189941406,
      "epoch": 4.4779614860922,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003147342240350625,
      "loss": 4.9034,
      "mean_token_accuracy": 0.22199461460113526,
      "num_tokens": 99830399.0,
      "step": 57555
    },
    {
      "entropy": 5.883170127868652,
      "epoch": 4.478350515463918,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003147069437430089,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21566027253866196,
      "num_tokens": 99839936.0,
      "step": 57560
    },
    {
      "entropy": 5.925524091720581,
      "epoch": 4.478739544835635,
      "grad_norm": 1.375,
      "learning_rate": 0.00031467966284846573,
      "loss": 4.9969,
      "mean_token_accuracy": 0.21042490452528,
      "num_tokens": 99848760.0,
      "step": 57565
    },
    {
      "entropy": 6.052072048187256,
      "epoch": 4.479128574207353,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0003146523813518468,
      "loss": 5.0365,
      "mean_token_accuracy": 0.20962952971458435,
      "num_tokens": 99856639.0,
      "step": 57570
    },
    {
      "entropy": 5.923245096206665,
      "epoch": 4.47951760357907,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031462509925356616,
      "loss": 4.891,
      "mean_token_accuracy": 0.22599357217550278,
      "num_tokens": 99865213.0,
      "step": 57575
    },
    {
      "entropy": 5.859971141815185,
      "epoch": 4.4799066329507875,
      "grad_norm": 1.546875,
      "learning_rate": 0.00031459781655403756,
      "loss": 4.8874,
      "mean_token_accuracy": 0.2259276643395424,
      "num_tokens": 99873763.0,
      "step": 57580
    },
    {
      "entropy": 5.972398042678833,
      "epoch": 4.480295662322505,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003145705332536752,
      "loss": 5.0022,
      "mean_token_accuracy": 0.21115858405828475,
      "num_tokens": 99881786.0,
      "step": 57585
    },
    {
      "entropy": 5.950938367843628,
      "epoch": 4.480684691694223,
      "grad_norm": 1.515625,
      "learning_rate": 0.000314543249352893,
      "loss": 4.985,
      "mean_token_accuracy": 0.2106778532266617,
      "num_tokens": 99890224.0,
      "step": 57590
    },
    {
      "entropy": 5.891128444671631,
      "epoch": 4.481073721065941,
      "grad_norm": 1.5,
      "learning_rate": 0.00031451596485210486,
      "loss": 4.8383,
      "mean_token_accuracy": 0.21807336211204528,
      "num_tokens": 99898523.0,
      "step": 57595
    },
    {
      "entropy": 5.8926342010498045,
      "epoch": 4.481462750437658,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031448867975172493,
      "loss": 4.9129,
      "mean_token_accuracy": 0.2168262019753456,
      "num_tokens": 99906920.0,
      "step": 57600
    },
    {
      "entropy": 6.026831817626953,
      "epoch": 4.481851779809376,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031446139405216713,
      "loss": 5.0268,
      "mean_token_accuracy": 0.2158502906560898,
      "num_tokens": 99915813.0,
      "step": 57605
    },
    {
      "entropy": 5.899924325942993,
      "epoch": 4.482240809181093,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003144341077538455,
      "loss": 4.9103,
      "mean_token_accuracy": 0.21995767205953598,
      "num_tokens": 99924912.0,
      "step": 57610
    },
    {
      "entropy": 5.943276834487915,
      "epoch": 4.4826298385528105,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000314406820857174,
      "loss": 4.9738,
      "mean_token_accuracy": 0.218840192258358,
      "num_tokens": 99933107.0,
      "step": 57615
    },
    {
      "entropy": 5.868537378311157,
      "epoch": 4.483018867924528,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031437953336256667,
      "loss": 4.8777,
      "mean_token_accuracy": 0.230436509847641,
      "num_tokens": 99942205.0,
      "step": 57620
    },
    {
      "entropy": 5.928694581985473,
      "epoch": 4.483407897296246,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003143522452704377,
      "loss": 4.9621,
      "mean_token_accuracy": 0.21687428206205367,
      "num_tokens": 99950807.0,
      "step": 57625
    },
    {
      "entropy": 5.878244352340698,
      "epoch": 4.483796926667964,
      "grad_norm": 1.4375,
      "learning_rate": 0.000314324956581201,
      "loss": 4.9764,
      "mean_token_accuracy": 0.21714761704206467,
      "num_tokens": 99960030.0,
      "step": 57630
    },
    {
      "entropy": 5.966232776641846,
      "epoch": 4.484185956039681,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003142976672952707,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20714084059000015,
      "num_tokens": 99969607.0,
      "step": 57635
    },
    {
      "entropy": 5.966753673553467,
      "epoch": 4.484574985411399,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031427037741306083,
      "loss": 4.9619,
      "mean_token_accuracy": 0.21584053039550782,
      "num_tokens": 99978453.0,
      "step": 57640
    },
    {
      "entropy": 5.9479176044464115,
      "epoch": 4.484964014783116,
      "grad_norm": 1.375,
      "learning_rate": 0.0003142430869349855,
      "loss": 4.9296,
      "mean_token_accuracy": 0.22080231606960296,
      "num_tokens": 99987150.0,
      "step": 57645
    },
    {
      "entropy": 5.87459659576416,
      "epoch": 4.4853530441548335,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003142157958614588,
      "loss": 4.8868,
      "mean_token_accuracy": 0.21633867919445038,
      "num_tokens": 99996437.0,
      "step": 57650
    },
    {
      "entropy": 5.888665676116943,
      "epoch": 4.485742073526551,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003141885041928948,
      "loss": 4.8656,
      "mean_token_accuracy": 0.22074079215526582,
      "num_tokens": 100005159.0,
      "step": 57655
    },
    {
      "entropy": 5.955147457122803,
      "epoch": 4.486131102898269,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031416121192970754,
      "loss": 5.049,
      "mean_token_accuracy": 0.2093716561794281,
      "num_tokens": 100014611.0,
      "step": 57660
    },
    {
      "entropy": 5.937571477890015,
      "epoch": 4.486520132269987,
      "grad_norm": 1.515625,
      "learning_rate": 0.00031413391907231133,
      "loss": 4.9821,
      "mean_token_accuracy": 0.21225055456161498,
      "num_tokens": 100022708.0,
      "step": 57665
    },
    {
      "entropy": 5.928589868545532,
      "epoch": 4.486909161641704,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031410662562112015,
      "loss": 4.9952,
      "mean_token_accuracy": 0.2150104448199272,
      "num_tokens": 100031813.0,
      "step": 57670
    },
    {
      "entropy": 5.933185958862305,
      "epoch": 4.487298191013421,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031407933157654816,
      "loss": 5.0073,
      "mean_token_accuracy": 0.21717143207788467,
      "num_tokens": 100040781.0,
      "step": 57675
    },
    {
      "entropy": 5.900042343139648,
      "epoch": 4.487687220385139,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003140520369390095,
      "loss": 4.8892,
      "mean_token_accuracy": 0.21994658708572387,
      "num_tokens": 100049206.0,
      "step": 57680
    },
    {
      "entropy": 5.923494672775268,
      "epoch": 4.4880762497568565,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003140247417089184,
      "loss": 4.9632,
      "mean_token_accuracy": 0.22148555666208267,
      "num_tokens": 100059034.0,
      "step": 57685
    },
    {
      "entropy": 5.93550124168396,
      "epoch": 4.488465279128574,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003139974458866889,
      "loss": 4.9624,
      "mean_token_accuracy": 0.2210059255361557,
      "num_tokens": 100067093.0,
      "step": 57690
    },
    {
      "entropy": 5.985317611694336,
      "epoch": 4.488854308500292,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003139701494727353,
      "loss": 4.9673,
      "mean_token_accuracy": 0.21572909951210023,
      "num_tokens": 100075235.0,
      "step": 57695
    },
    {
      "entropy": 5.982493352890015,
      "epoch": 4.48924333787201,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003139428524674716,
      "loss": 5.026,
      "mean_token_accuracy": 0.2073841780424118,
      "num_tokens": 100083032.0,
      "step": 57700
    },
    {
      "entropy": 5.976552057266235,
      "epoch": 4.489632367243727,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003139155548713122,
      "loss": 5.0688,
      "mean_token_accuracy": 0.20331808179616928,
      "num_tokens": 100091613.0,
      "step": 57705
    },
    {
      "entropy": 5.921510934829712,
      "epoch": 4.490021396615444,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00031388825668467115,
      "loss": 4.932,
      "mean_token_accuracy": 0.21367980986833573,
      "num_tokens": 100100234.0,
      "step": 57710
    },
    {
      "entropy": 5.936568927764893,
      "epoch": 4.490410425987162,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00031386095790796265,
      "loss": 5.0035,
      "mean_token_accuracy": 0.21372120529413224,
      "num_tokens": 100108881.0,
      "step": 57715
    },
    {
      "entropy": 5.961302042007446,
      "epoch": 4.4907994553588795,
      "grad_norm": 1.5,
      "learning_rate": 0.00031383365854160104,
      "loss": 4.9916,
      "mean_token_accuracy": 0.2061997652053833,
      "num_tokens": 100117338.0,
      "step": 57720
    },
    {
      "entropy": 5.978952169418335,
      "epoch": 4.491188484730597,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003138063585860005,
      "loss": 4.944,
      "mean_token_accuracy": 0.22043062895536422,
      "num_tokens": 100125708.0,
      "step": 57725
    },
    {
      "entropy": 5.886020374298096,
      "epoch": 4.491577514102315,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003137790580415752,
      "loss": 4.9027,
      "mean_token_accuracy": 0.21921835839748383,
      "num_tokens": 100134039.0,
      "step": 57730
    },
    {
      "entropy": 5.939590215682983,
      "epoch": 4.491966543474033,
      "grad_norm": 1.515625,
      "learning_rate": 0.00031375175690873943,
      "loss": 5.0131,
      "mean_token_accuracy": 0.21159750521183013,
      "num_tokens": 100142204.0,
      "step": 57735
    },
    {
      "entropy": 5.926702833175659,
      "epoch": 4.49235557284575,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031372445518790737,
      "loss": 4.8939,
      "mean_token_accuracy": 0.22311621457338332,
      "num_tokens": 100150831.0,
      "step": 57740
    },
    {
      "entropy": 5.985374832153321,
      "epoch": 4.492744602217467,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031369715287949334,
      "loss": 5.0455,
      "mean_token_accuracy": 0.21759130656719208,
      "num_tokens": 100159584.0,
      "step": 57745
    },
    {
      "entropy": 5.88170428276062,
      "epoch": 4.493133631589185,
      "grad_norm": 1.59375,
      "learning_rate": 0.00031366984998391163,
      "loss": 4.9358,
      "mean_token_accuracy": 0.21962212771177292,
      "num_tokens": 100167907.0,
      "step": 57750
    },
    {
      "entropy": 5.937469005584717,
      "epoch": 4.4935226609609025,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00031364254650157646,
      "loss": 4.9449,
      "mean_token_accuracy": 0.2239786207675934,
      "num_tokens": 100176578.0,
      "step": 57755
    },
    {
      "entropy": 5.977820348739624,
      "epoch": 4.49391169033262,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003136152424329023,
      "loss": 5.0555,
      "mean_token_accuracy": 0.20882636606693267,
      "num_tokens": 100184514.0,
      "step": 57760
    },
    {
      "entropy": 5.942767858505249,
      "epoch": 4.494300719704338,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003135879377783032,
      "loss": 4.8988,
      "mean_token_accuracy": 0.21859647184610367,
      "num_tokens": 100192851.0,
      "step": 57765
    },
    {
      "entropy": 5.882703495025635,
      "epoch": 4.494689749076056,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031356063253819354,
      "loss": 4.8746,
      "mean_token_accuracy": 0.22332196533679963,
      "num_tokens": 100201874.0,
      "step": 57770
    },
    {
      "entropy": 5.939625453948975,
      "epoch": 4.495078778447773,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003135333267129877,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2137034296989441,
      "num_tokens": 100210032.0,
      "step": 57775
    },
    {
      "entropy": 5.998376655578613,
      "epoch": 4.49546780781949,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003135060203030999,
      "loss": 5.094,
      "mean_token_accuracy": 0.20147220492362977,
      "num_tokens": 100219780.0,
      "step": 57780
    },
    {
      "entropy": 6.040231800079345,
      "epoch": 4.495856837191208,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00031347871330894464,
      "loss": 5.0595,
      "mean_token_accuracy": 0.21232319325208665,
      "num_tokens": 100227951.0,
      "step": 57785
    },
    {
      "entropy": 5.977315950393677,
      "epoch": 4.4962458665629255,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000313451405730936,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21672005951404572,
      "num_tokens": 100236304.0,
      "step": 57790
    },
    {
      "entropy": 5.898017406463623,
      "epoch": 4.496634895934643,
      "grad_norm": 1.375,
      "learning_rate": 0.0003134240975694886,
      "loss": 4.9315,
      "mean_token_accuracy": 0.21468886584043503,
      "num_tokens": 100244889.0,
      "step": 57795
    },
    {
      "entropy": 5.918809700012207,
      "epoch": 4.497023925306361,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031339678882501667,
      "loss": 4.9297,
      "mean_token_accuracy": 0.22092988193035126,
      "num_tokens": 100253043.0,
      "step": 57800
    },
    {
      "entropy": 5.949669885635376,
      "epoch": 4.497412954678079,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031336947949793455,
      "loss": 4.9797,
      "mean_token_accuracy": 0.21337084323167801,
      "num_tokens": 100262184.0,
      "step": 57805
    },
    {
      "entropy": 5.934546279907226,
      "epoch": 4.497801984049795,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031334216958865663,
      "loss": 4.9102,
      "mean_token_accuracy": 0.2185825526714325,
      "num_tokens": 100270564.0,
      "step": 57810
    },
    {
      "entropy": 5.974032926559448,
      "epoch": 4.498191013421513,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031331485909759734,
      "loss": 4.9906,
      "mean_token_accuracy": 0.21082500070333482,
      "num_tokens": 100279406.0,
      "step": 57815
    },
    {
      "entropy": 5.955420875549317,
      "epoch": 4.498580042793231,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000313287548025171,
      "loss": 4.9847,
      "mean_token_accuracy": 0.21955261677503585,
      "num_tokens": 100288191.0,
      "step": 57820
    },
    {
      "entropy": 5.8724517822265625,
      "epoch": 4.4989690721649485,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003132602363717921,
      "loss": 4.8581,
      "mean_token_accuracy": 0.2180185616016388,
      "num_tokens": 100297269.0,
      "step": 57825
    },
    {
      "entropy": 5.9340564727783205,
      "epoch": 4.499358101536666,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031323292413787504,
      "loss": 4.9609,
      "mean_token_accuracy": 0.2196611136198044,
      "num_tokens": 100306009.0,
      "step": 57830
    },
    {
      "entropy": 5.995954370498657,
      "epoch": 4.499747130908384,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003132056113238341,
      "loss": 5.0578,
      "mean_token_accuracy": 0.2044097065925598,
      "num_tokens": 100314771.0,
      "step": 57835
    },
    {
      "entropy": 5.973256969451905,
      "epoch": 4.5001361602801015,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031317829793008403,
      "loss": 4.9461,
      "mean_token_accuracy": 0.2126249760389328,
      "num_tokens": 100323912.0,
      "step": 57840
    },
    {
      "entropy": 5.97506890296936,
      "epoch": 4.500525189651818,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003131509839570389,
      "loss": 4.9901,
      "mean_token_accuracy": 0.20717935711145402,
      "num_tokens": 100332351.0,
      "step": 57845
    },
    {
      "entropy": 5.941787815093994,
      "epoch": 4.500914219023536,
      "grad_norm": 1.46875,
      "learning_rate": 0.00031312366940511327,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22120385318994523,
      "num_tokens": 100340547.0,
      "step": 57850
    },
    {
      "entropy": 5.896755743026733,
      "epoch": 4.501303248395254,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003130963542747217,
      "loss": 4.9059,
      "mean_token_accuracy": 0.21704573631286622,
      "num_tokens": 100349067.0,
      "step": 57855
    },
    {
      "entropy": 5.872461891174316,
      "epoch": 4.5016922777669715,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00031306903856627866,
      "loss": 4.8989,
      "mean_token_accuracy": 0.22063470929861068,
      "num_tokens": 100357735.0,
      "step": 57860
    },
    {
      "entropy": 5.945851373672485,
      "epoch": 4.502081307138689,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031304172228019845,
      "loss": 4.9765,
      "mean_token_accuracy": 0.2170764222741127,
      "num_tokens": 100366389.0,
      "step": 57865
    },
    {
      "entropy": 5.9636180877685545,
      "epoch": 4.502470336510407,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031301440541689577,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22397948205471038,
      "num_tokens": 100374367.0,
      "step": 57870
    },
    {
      "entropy": 5.939149522781372,
      "epoch": 4.502859365882124,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003129870879767849,
      "loss": 4.939,
      "mean_token_accuracy": 0.2165229767560959,
      "num_tokens": 100382316.0,
      "step": 57875
    },
    {
      "entropy": 5.836052799224854,
      "epoch": 4.503248395253841,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0003129597699602804,
      "loss": 4.8859,
      "mean_token_accuracy": 0.2217896744608879,
      "num_tokens": 100390771.0,
      "step": 57880
    },
    {
      "entropy": 5.975125885009765,
      "epoch": 4.503637424625559,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003129324513677969,
      "loss": 5.0164,
      "mean_token_accuracy": 0.21014364659786225,
      "num_tokens": 100399777.0,
      "step": 57885
    },
    {
      "entropy": 5.951526355743408,
      "epoch": 4.504026453997277,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031290513219974875,
      "loss": 4.992,
      "mean_token_accuracy": 0.216563680768013,
      "num_tokens": 100408750.0,
      "step": 57890
    },
    {
      "entropy": 5.912357997894287,
      "epoch": 4.504415483368994,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031287781245655064,
      "loss": 4.9855,
      "mean_token_accuracy": 0.21447935253381728,
      "num_tokens": 100417278.0,
      "step": 57895
    },
    {
      "entropy": 5.946004152297974,
      "epoch": 4.504804512740712,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000312850492138617,
      "loss": 5.0126,
      "mean_token_accuracy": 0.21143608093261718,
      "num_tokens": 100425542.0,
      "step": 57900
    },
    {
      "entropy": 5.955247783660889,
      "epoch": 4.50519354211243,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003128231712463623,
      "loss": 4.9493,
      "mean_token_accuracy": 0.21840673983097075,
      "num_tokens": 100433492.0,
      "step": 57905
    },
    {
      "entropy": 5.927680110931396,
      "epoch": 4.5055825714841475,
      "grad_norm": 1.578125,
      "learning_rate": 0.0003127958497802012,
      "loss": 4.8906,
      "mean_token_accuracy": 0.2204967275261879,
      "num_tokens": 100442262.0,
      "step": 57910
    },
    {
      "entropy": 6.011193418502808,
      "epoch": 4.505971600855864,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003127685277405483,
      "loss": 5.0943,
      "mean_token_accuracy": 0.20572330206632614,
      "num_tokens": 100450886.0,
      "step": 57915
    },
    {
      "entropy": 5.952038145065307,
      "epoch": 4.506360630227582,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031274120512781806,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21237735450267792,
      "num_tokens": 100459379.0,
      "step": 57920
    },
    {
      "entropy": 5.924078464508057,
      "epoch": 4.5067496595993,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003127138819424252,
      "loss": 4.9341,
      "mean_token_accuracy": 0.21729949116706848,
      "num_tokens": 100467809.0,
      "step": 57925
    },
    {
      "entropy": 5.9069318771362305,
      "epoch": 4.507138688971017,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0003126865581847841,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2180008977651596,
      "num_tokens": 100476097.0,
      "step": 57930
    },
    {
      "entropy": 6.01404185295105,
      "epoch": 4.507527718342735,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031265923385530943,
      "loss": 5.1029,
      "mean_token_accuracy": 0.20472257137298583,
      "num_tokens": 100485403.0,
      "step": 57935
    },
    {
      "entropy": 5.955080318450928,
      "epoch": 4.507916747714452,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003126319089544159,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22053979933261872,
      "num_tokens": 100493782.0,
      "step": 57940
    },
    {
      "entropy": 6.044373989105225,
      "epoch": 4.50830577708617,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000312604583482518,
      "loss": 5.1237,
      "mean_token_accuracy": 0.20716950297355652,
      "num_tokens": 100503346.0,
      "step": 57945
    },
    {
      "entropy": 5.901104545593261,
      "epoch": 4.508694806457887,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003125772574400305,
      "loss": 4.9211,
      "mean_token_accuracy": 0.21207444220781327,
      "num_tokens": 100512015.0,
      "step": 57950
    },
    {
      "entropy": 5.883977031707763,
      "epoch": 4.509083835829605,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003125499308273679,
      "loss": 4.9874,
      "mean_token_accuracy": 0.21742089539766313,
      "num_tokens": 100520567.0,
      "step": 57955
    },
    {
      "entropy": 5.954219484329224,
      "epoch": 4.509472865201323,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00031252260364494477,
      "loss": 5.012,
      "mean_token_accuracy": 0.2082167685031891,
      "num_tokens": 100529004.0,
      "step": 57960
    },
    {
      "entropy": 5.940612649917602,
      "epoch": 4.50986189457304,
      "grad_norm": 1.375,
      "learning_rate": 0.00031249527589317596,
      "loss": 4.9897,
      "mean_token_accuracy": 0.21481944769620895,
      "num_tokens": 100538543.0,
      "step": 57965
    },
    {
      "entropy": 5.916070938110352,
      "epoch": 4.510250923944758,
      "grad_norm": 1.5,
      "learning_rate": 0.0003124679475724759,
      "loss": 4.9764,
      "mean_token_accuracy": 0.21857193112373352,
      "num_tokens": 100547456.0,
      "step": 57970
    },
    {
      "entropy": 5.909831190109253,
      "epoch": 4.510639953316476,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003124406186832595,
      "loss": 4.9193,
      "mean_token_accuracy": 0.21065421849489213,
      "num_tokens": 100556649.0,
      "step": 57975
    },
    {
      "entropy": 6.011018848419189,
      "epoch": 4.511028982688193,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003124132892259412,
      "loss": 5.0354,
      "mean_token_accuracy": 0.20909819304943084,
      "num_tokens": 100565323.0,
      "step": 57980
    },
    {
      "entropy": 5.9642504215240475,
      "epoch": 4.51141801205991,
      "grad_norm": 1.515625,
      "learning_rate": 0.00031238595920093584,
      "loss": 4.9114,
      "mean_token_accuracy": 0.22024666666984558,
      "num_tokens": 100574801.0,
      "step": 57985
    },
    {
      "entropy": 5.969989156723022,
      "epoch": 4.511807041431628,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031235862860865807,
      "loss": 4.9979,
      "mean_token_accuracy": 0.21524639427661896,
      "num_tokens": 100583879.0,
      "step": 57990
    },
    {
      "entropy": 5.907824373245239,
      "epoch": 4.512196070803346,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003123312974495226,
      "loss": 4.9109,
      "mean_token_accuracy": 0.22304484248161316,
      "num_tokens": 100592886.0,
      "step": 57995
    },
    {
      "entropy": 5.9389252185821535,
      "epoch": 4.512585100175063,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031230396572394406,
      "loss": 4.9769,
      "mean_token_accuracy": 0.22157772034406661,
      "num_tokens": 100601604.0,
      "step": 58000
    },
    {
      "entropy": 5.927848291397095,
      "epoch": 4.512974129546781,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003122766334323372,
      "loss": 5.0518,
      "mean_token_accuracy": 0.2080589011311531,
      "num_tokens": 100610436.0,
      "step": 58005
    },
    {
      "entropy": 5.980579471588134,
      "epoch": 4.513363158918498,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031224930057511673,
      "loss": 5.0112,
      "mean_token_accuracy": 0.21565478444099426,
      "num_tokens": 100618704.0,
      "step": 58010
    },
    {
      "entropy": 6.0168064594268795,
      "epoch": 4.513752188290216,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003122219671526974,
      "loss": 5.0224,
      "mean_token_accuracy": 0.2086917668581009,
      "num_tokens": 100627387.0,
      "step": 58015
    },
    {
      "entropy": 5.944635057449341,
      "epoch": 4.514141217661933,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031219463316549406,
      "loss": 4.8939,
      "mean_token_accuracy": 0.21469616293907165,
      "num_tokens": 100635862.0,
      "step": 58020
    },
    {
      "entropy": 5.907937288284302,
      "epoch": 4.514530247033651,
      "grad_norm": 1.546875,
      "learning_rate": 0.00031216729861392137,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20405325442552566,
      "num_tokens": 100644728.0,
      "step": 58025
    },
    {
      "entropy": 5.945425415039063,
      "epoch": 4.514919276405369,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003121399634983941,
      "loss": 4.9403,
      "mean_token_accuracy": 0.22173851430416108,
      "num_tokens": 100653684.0,
      "step": 58030
    },
    {
      "entropy": 5.986827421188354,
      "epoch": 4.515308305777086,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031211262781932693,
      "loss": 4.9578,
      "mean_token_accuracy": 0.22344642281532287,
      "num_tokens": 100662522.0,
      "step": 58035
    },
    {
      "entropy": 6.002403736114502,
      "epoch": 4.515697335148804,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031208529157713465,
      "loss": 5.0481,
      "mean_token_accuracy": 0.21523663252592087,
      "num_tokens": 100670810.0,
      "step": 58040
    },
    {
      "entropy": 5.921158218383789,
      "epoch": 4.516086364520521,
      "grad_norm": 1.46875,
      "learning_rate": 0.00031205795477223217,
      "loss": 4.9861,
      "mean_token_accuracy": 0.2157572850584984,
      "num_tokens": 100679749.0,
      "step": 58045
    },
    {
      "entropy": 5.958259773254395,
      "epoch": 4.516475393892239,
      "grad_norm": 1.375,
      "learning_rate": 0.00031203061740503414,
      "loss": 5.0412,
      "mean_token_accuracy": 0.20701769590377808,
      "num_tokens": 100689753.0,
      "step": 58050
    },
    {
      "entropy": 5.979867744445801,
      "epoch": 4.516864423263956,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003120032794759555,
      "loss": 5.0475,
      "mean_token_accuracy": 0.21024521440267563,
      "num_tokens": 100698278.0,
      "step": 58055
    },
    {
      "entropy": 5.926177263259888,
      "epoch": 4.517253452635674,
      "grad_norm": 1.46875,
      "learning_rate": 0.000311975940985411,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22328974157571793,
      "num_tokens": 100707354.0,
      "step": 58060
    },
    {
      "entropy": 5.9505815505981445,
      "epoch": 4.517642482007392,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003119486019338154,
      "loss": 4.9565,
      "mean_token_accuracy": 0.2207326591014862,
      "num_tokens": 100715946.0,
      "step": 58065
    },
    {
      "entropy": 6.044514751434326,
      "epoch": 4.518031511379109,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00031192126232158367,
      "loss": 5.0245,
      "mean_token_accuracy": 0.21429793983697892,
      "num_tokens": 100725288.0,
      "step": 58070
    },
    {
      "entropy": 5.952231979370117,
      "epoch": 4.518420540750826,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003118939221491305,
      "loss": 5.0001,
      "mean_token_accuracy": 0.21090563982725144,
      "num_tokens": 100733565.0,
      "step": 58075
    },
    {
      "entropy": 5.95211615562439,
      "epoch": 4.518809570122544,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003118665814168707,
      "loss": 4.982,
      "mean_token_accuracy": 0.21899991929531099,
      "num_tokens": 100741937.0,
      "step": 58080
    },
    {
      "entropy": 5.932010841369629,
      "epoch": 4.519198599494262,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00031183924012521925,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21757580041885377,
      "num_tokens": 100750581.0,
      "step": 58085
    },
    {
      "entropy": 5.894408798217773,
      "epoch": 4.519587628865979,
      "grad_norm": 1.421875,
      "learning_rate": 0.000311811898274591,
      "loss": 4.9383,
      "mean_token_accuracy": 0.22093418836593628,
      "num_tokens": 100758775.0,
      "step": 58090
    },
    {
      "entropy": 5.938787794113159,
      "epoch": 4.519976658237697,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031178455586540076,
      "loss": 4.9408,
      "mean_token_accuracy": 0.22057174891233444,
      "num_tokens": 100767002.0,
      "step": 58095
    },
    {
      "entropy": 6.046633338928222,
      "epoch": 4.520365687609415,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00031175721289806345,
      "loss": 5.0269,
      "mean_token_accuracy": 0.20922839045524597,
      "num_tokens": 100775428.0,
      "step": 58100
    },
    {
      "entropy": 5.992326736450195,
      "epoch": 4.520754716981132,
      "grad_norm": 1.5,
      "learning_rate": 0.00031172986937299396,
      "loss": 5.0104,
      "mean_token_accuracy": 0.212798373401165,
      "num_tokens": 100784164.0,
      "step": 58105
    },
    {
      "entropy": 5.99232497215271,
      "epoch": 4.52114374635285,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031170252529060715,
      "loss": 5.0049,
      "mean_token_accuracy": 0.21700541377067567,
      "num_tokens": 100792627.0,
      "step": 58110
    },
    {
      "entropy": 5.906730270385742,
      "epoch": 4.521532775724567,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003116751806513179,
      "loss": 4.9752,
      "mean_token_accuracy": 0.22187259495258332,
      "num_tokens": 100802179.0,
      "step": 58115
    },
    {
      "entropy": 5.9361485004425045,
      "epoch": 4.521921805096285,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003116478354555412,
      "loss": 4.905,
      "mean_token_accuracy": 0.21661875396966934,
      "num_tokens": 100810875.0,
      "step": 58120
    },
    {
      "entropy": 5.91546139717102,
      "epoch": 4.522310834468002,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031162048970369196,
      "loss": 4.996,
      "mean_token_accuracy": 0.21462104767560958,
      "num_tokens": 100820110.0,
      "step": 58125
    },
    {
      "entropy": 5.979083824157715,
      "epoch": 4.52269986383972,
      "grad_norm": 1.40625,
      "learning_rate": 0.000311593143396185,
      "loss": 5.0544,
      "mean_token_accuracy": 0.20374642312526703,
      "num_tokens": 100829573.0,
      "step": 58130
    },
    {
      "entropy": 6.031494140625,
      "epoch": 4.523088893211438,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031156579653343543,
      "loss": 5.0824,
      "mean_token_accuracy": 0.20815053582191467,
      "num_tokens": 100838430.0,
      "step": 58135
    },
    {
      "entropy": 5.926622438430786,
      "epoch": 4.523477922583155,
      "grad_norm": 1.28125,
      "learning_rate": 0.000311538449115858,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21772463023662567,
      "num_tokens": 100847400.0,
      "step": 58140
    },
    {
      "entropy": 5.908663272857666,
      "epoch": 4.523866951954872,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003115111011438678,
      "loss": 4.9401,
      "mean_token_accuracy": 0.21329007297754288,
      "num_tokens": 100855909.0,
      "step": 58145
    },
    {
      "entropy": 6.008129072189331,
      "epoch": 4.52425598132659,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003114837526178797,
      "loss": 5.052,
      "mean_token_accuracy": 0.21219624280929567,
      "num_tokens": 100864577.0,
      "step": 58150
    },
    {
      "entropy": 5.967572116851807,
      "epoch": 4.524645010698308,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003114564035383088,
      "loss": 5.0196,
      "mean_token_accuracy": 0.2084329307079315,
      "num_tokens": 100873059.0,
      "step": 58155
    },
    {
      "entropy": 5.947109651565552,
      "epoch": 4.525034040070025,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031142905390556995,
      "loss": 4.9895,
      "mean_token_accuracy": 0.21399738192558287,
      "num_tokens": 100881548.0,
      "step": 58160
    },
    {
      "entropy": 5.908093881607056,
      "epoch": 4.525423069441743,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003114017037200782,
      "loss": 4.9665,
      "mean_token_accuracy": 0.21674164682626723,
      "num_tokens": 100890732.0,
      "step": 58165
    },
    {
      "entropy": 5.9392472267150875,
      "epoch": 4.525812098813461,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031137435298224863,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21220220774412155,
      "num_tokens": 100899273.0,
      "step": 58170
    },
    {
      "entropy": 6.0436419486999515,
      "epoch": 4.526201128185178,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003113470016924961,
      "loss": 4.9742,
      "mean_token_accuracy": 0.21335437148809433,
      "num_tokens": 100907920.0,
      "step": 58175
    },
    {
      "entropy": 5.938825416564941,
      "epoch": 4.526590157556895,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031131964985123555,
      "loss": 4.9705,
      "mean_token_accuracy": 0.2114158809185028,
      "num_tokens": 100916754.0,
      "step": 58180
    },
    {
      "entropy": 5.89545168876648,
      "epoch": 4.526979186928613,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003112922974588822,
      "loss": 4.9573,
      "mean_token_accuracy": 0.2165951356291771,
      "num_tokens": 100925816.0,
      "step": 58185
    },
    {
      "entropy": 5.879301452636719,
      "epoch": 4.527368216300331,
      "grad_norm": 1.375,
      "learning_rate": 0.000311264944515851,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22944784313440322,
      "num_tokens": 100933833.0,
      "step": 58190
    },
    {
      "entropy": 5.993469715118408,
      "epoch": 4.527757245672048,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00031123759102255714,
      "loss": 5.0113,
      "mean_token_accuracy": 0.2101433530449867,
      "num_tokens": 100941945.0,
      "step": 58195
    },
    {
      "entropy": 5.9760418891906735,
      "epoch": 4.528146275043766,
      "grad_norm": 1.609375,
      "learning_rate": 0.0003112102369794153,
      "loss": 4.9141,
      "mean_token_accuracy": 0.2166390985250473,
      "num_tokens": 100949638.0,
      "step": 58200
    },
    {
      "entropy": 5.9400999546051025,
      "epoch": 4.528535304415484,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031118288238684086,
      "loss": 4.9995,
      "mean_token_accuracy": 0.21039969027042388,
      "num_tokens": 100959083.0,
      "step": 58205
    },
    {
      "entropy": 6.008056879043579,
      "epoch": 4.5289243337872005,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003111555272452486,
      "loss": 4.9852,
      "mean_token_accuracy": 0.21312781274318696,
      "num_tokens": 100967532.0,
      "step": 58210
    },
    {
      "entropy": 5.898654174804688,
      "epoch": 4.529313363158918,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031112817155505394,
      "loss": 4.917,
      "mean_token_accuracy": 0.2165583923459053,
      "num_tokens": 100975683.0,
      "step": 58215
    },
    {
      "entropy": 5.873921728134155,
      "epoch": 4.529702392530636,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00031110081531667184,
      "loss": 4.9096,
      "mean_token_accuracy": 0.2201073944568634,
      "num_tokens": 100984946.0,
      "step": 58220
    },
    {
      "entropy": 5.890289449691773,
      "epoch": 4.530091421902354,
      "grad_norm": 1.5,
      "learning_rate": 0.00031107345853051714,
      "loss": 4.9165,
      "mean_token_accuracy": 0.2166956514120102,
      "num_tokens": 100993925.0,
      "step": 58225
    },
    {
      "entropy": 6.00723934173584,
      "epoch": 4.530480451274071,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031104610119700525,
      "loss": 5.0107,
      "mean_token_accuracy": 0.21244004219770432,
      "num_tokens": 101002777.0,
      "step": 58230
    },
    {
      "entropy": 5.895885896682739,
      "epoch": 4.530869480645789,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003110187433165511,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21465131640434265,
      "num_tokens": 101010870.0,
      "step": 58235
    },
    {
      "entropy": 5.947065591812134,
      "epoch": 4.531258510017507,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003109913848895699,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2117588058114052,
      "num_tokens": 101020279.0,
      "step": 58240
    },
    {
      "entropy": 5.940305995941162,
      "epoch": 4.531647539389224,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003109640259164766,
      "loss": 4.991,
      "mean_token_accuracy": 0.21450170874595642,
      "num_tokens": 101028362.0,
      "step": 58245
    },
    {
      "entropy": 5.90500226020813,
      "epoch": 4.532036568760941,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031093666639768654,
      "loss": 4.8987,
      "mean_token_accuracy": 0.22023587077856063,
      "num_tokens": 101037143.0,
      "step": 58250
    },
    {
      "entropy": 5.960330820083618,
      "epoch": 4.532425598132659,
      "grad_norm": 1.375,
      "learning_rate": 0.0003109093063336147,
      "loss": 4.9785,
      "mean_token_accuracy": 0.21750691831111907,
      "num_tokens": 101046625.0,
      "step": 58255
    },
    {
      "entropy": 6.009812450408935,
      "epoch": 4.532814627504377,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003108819457246764,
      "loss": 5.1133,
      "mean_token_accuracy": 0.20294404029846191,
      "num_tokens": 101055960.0,
      "step": 58260
    },
    {
      "entropy": 5.963505506515503,
      "epoch": 4.533203656876094,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031085458457128667,
      "loss": 4.9898,
      "mean_token_accuracy": 0.20492182224988936,
      "num_tokens": 101064074.0,
      "step": 58265
    },
    {
      "entropy": 5.9894860744476315,
      "epoch": 4.533592686247812,
      "grad_norm": 1.375,
      "learning_rate": 0.00031082722287386065,
      "loss": 5.0364,
      "mean_token_accuracy": 0.2128014013171196,
      "num_tokens": 101072379.0,
      "step": 58270
    },
    {
      "entropy": 5.959867477416992,
      "epoch": 4.533981715619529,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003107998606328135,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21974776238203048,
      "num_tokens": 101081841.0,
      "step": 58275
    },
    {
      "entropy": 5.9559191226959225,
      "epoch": 4.5343707449912465,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031077249784856057,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20889939963817597,
      "num_tokens": 101091294.0,
      "step": 58280
    },
    {
      "entropy": 5.92326922416687,
      "epoch": 4.534759774362964,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031074513452151683,
      "loss": 4.9072,
      "mean_token_accuracy": 0.21585903018712999,
      "num_tokens": 101099677.0,
      "step": 58285
    },
    {
      "entropy": 5.978474521636963,
      "epoch": 4.535148803734682,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003107177706520975,
      "loss": 5.0258,
      "mean_token_accuracy": 0.21350251287221908,
      "num_tokens": 101108442.0,
      "step": 58290
    },
    {
      "entropy": 5.9244619846344,
      "epoch": 4.5355378331064,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000310690406240718,
      "loss": 5.0317,
      "mean_token_accuracy": 0.20912282466888427,
      "num_tokens": 101116663.0,
      "step": 58295
    },
    {
      "entropy": 5.990008115768433,
      "epoch": 4.535926862478117,
      "grad_norm": 1.515625,
      "learning_rate": 0.00031066304128779336,
      "loss": 5.0005,
      "mean_token_accuracy": 0.21326595693826675,
      "num_tokens": 101125202.0,
      "step": 58300
    },
    {
      "entropy": 5.981547594070435,
      "epoch": 4.536315891849835,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031063567579373876,
      "loss": 4.9788,
      "mean_token_accuracy": 0.20939721614122392,
      "num_tokens": 101134075.0,
      "step": 58305
    },
    {
      "entropy": 5.991951656341553,
      "epoch": 4.536704921221553,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00031060830975896954,
      "loss": 4.9897,
      "mean_token_accuracy": 0.21683793365955353,
      "num_tokens": 101142415.0,
      "step": 58310
    },
    {
      "entropy": 5.885130548477173,
      "epoch": 4.5370939505932695,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031058094318390085,
      "loss": 4.9327,
      "mean_token_accuracy": 0.2166779637336731,
      "num_tokens": 101151057.0,
      "step": 58315
    },
    {
      "entropy": 5.870819997787476,
      "epoch": 4.537482979964987,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031055357606894806,
      "loss": 4.9695,
      "mean_token_accuracy": 0.21005047410726546,
      "num_tokens": 101159604.0,
      "step": 58320
    },
    {
      "entropy": 5.963169288635254,
      "epoch": 4.537872009336705,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003105262084145263,
      "loss": 4.9063,
      "mean_token_accuracy": 0.2237996503710747,
      "num_tokens": 101167936.0,
      "step": 58325
    },
    {
      "entropy": 5.995974063873291,
      "epoch": 4.538261038708423,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003104988402210509,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21224923878908158,
      "num_tokens": 101176605.0,
      "step": 58330
    },
    {
      "entropy": 5.912528562545776,
      "epoch": 4.53865006808014,
      "grad_norm": 1.5,
      "learning_rate": 0.0003104714714889371,
      "loss": 4.9819,
      "mean_token_accuracy": 0.2128423348069191,
      "num_tokens": 101185669.0,
      "step": 58335
    },
    {
      "entropy": 5.9232262134552,
      "epoch": 4.539039097451858,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031044410221860017,
      "loss": 4.9307,
      "mean_token_accuracy": 0.22111738622188568,
      "num_tokens": 101194425.0,
      "step": 58340
    },
    {
      "entropy": 5.95557050704956,
      "epoch": 4.539428126823575,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003104167324104553,
      "loss": 4.9426,
      "mean_token_accuracy": 0.22172147780656815,
      "num_tokens": 101203232.0,
      "step": 58345
    },
    {
      "entropy": 5.9572936534881595,
      "epoch": 4.5398171561952925,
      "grad_norm": 1.484375,
      "learning_rate": 0.00031038936206491803,
      "loss": 4.9349,
      "mean_token_accuracy": 0.21463282704353331,
      "num_tokens": 101210948.0,
      "step": 58350
    },
    {
      "entropy": 5.820258140563965,
      "epoch": 4.54020618556701,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003103619911824035,
      "loss": 4.8454,
      "mean_token_accuracy": 0.22741685956716537,
      "num_tokens": 101220296.0,
      "step": 58355
    },
    {
      "entropy": 5.91034083366394,
      "epoch": 4.540595214938728,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000310334619763327,
      "loss": 5.0063,
      "mean_token_accuracy": 0.2147918686270714,
      "num_tokens": 101228908.0,
      "step": 58360
    },
    {
      "entropy": 5.940120697021484,
      "epoch": 4.540984244310446,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031030724780810385,
      "loss": 4.9821,
      "mean_token_accuracy": 0.215407295525074,
      "num_tokens": 101237919.0,
      "step": 58365
    },
    {
      "entropy": 6.046345853805542,
      "epoch": 4.541373273682163,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031027987531714944,
      "loss": 5.0199,
      "mean_token_accuracy": 0.20666933953762054,
      "num_tokens": 101246341.0,
      "step": 58370
    },
    {
      "entropy": 5.916973495483399,
      "epoch": 4.541762303053881,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00031025250229087915,
      "loss": 4.8988,
      "mean_token_accuracy": 0.21875904202461244,
      "num_tokens": 101254626.0,
      "step": 58375
    },
    {
      "entropy": 5.9798685073852536,
      "epoch": 4.542151332425598,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00031022512872970823,
      "loss": 5.0804,
      "mean_token_accuracy": 0.20609446316957475,
      "num_tokens": 101263390.0,
      "step": 58380
    },
    {
      "entropy": 5.868487071990967,
      "epoch": 4.5425403617973155,
      "grad_norm": 1.390625,
      "learning_rate": 0.000310197754634052,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21734785437583923,
      "num_tokens": 101272276.0,
      "step": 58385
    },
    {
      "entropy": 6.017352867126465,
      "epoch": 4.542929391169033,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003101703800043258,
      "loss": 4.998,
      "mean_token_accuracy": 0.21534724831581115,
      "num_tokens": 101280201.0,
      "step": 58390
    },
    {
      "entropy": 5.965345859527588,
      "epoch": 4.543318420540751,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031014300484094516,
      "loss": 4.9846,
      "mean_token_accuracy": 0.21885154098272325,
      "num_tokens": 101289170.0,
      "step": 58395
    },
    {
      "entropy": 5.924568462371826,
      "epoch": 4.5437074499124686,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003101156291443254,
      "loss": 4.9683,
      "mean_token_accuracy": 0.21679244488477706,
      "num_tokens": 101298687.0,
      "step": 58400
    },
    {
      "entropy": 6.063735151290894,
      "epoch": 4.544096479284186,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003100882529148817,
      "loss": 5.163,
      "mean_token_accuracy": 0.1990307942032814,
      "num_tokens": 101306898.0,
      "step": 58405
    },
    {
      "entropy": 6.020080089569092,
      "epoch": 4.544485508655903,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031006087615302976,
      "loss": 5.0271,
      "mean_token_accuracy": 0.2125757083296776,
      "num_tokens": 101316176.0,
      "step": 58410
    },
    {
      "entropy": 5.994750022888184,
      "epoch": 4.544874538027621,
      "grad_norm": 1.5,
      "learning_rate": 0.00031003349885918473,
      "loss": 4.9368,
      "mean_token_accuracy": 0.21369680911302566,
      "num_tokens": 101324264.0,
      "step": 58415
    },
    {
      "entropy": 5.965234851837158,
      "epoch": 4.5452635673993385,
      "grad_norm": 1.375,
      "learning_rate": 0.00031000612103376226,
      "loss": 4.947,
      "mean_token_accuracy": 0.21250844597816468,
      "num_tokens": 101333199.0,
      "step": 58420
    },
    {
      "entropy": 5.974844455718994,
      "epoch": 4.545652596771056,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003099787426771775,
      "loss": 4.9996,
      "mean_token_accuracy": 0.21571554392576217,
      "num_tokens": 101341851.0,
      "step": 58425
    },
    {
      "entropy": 5.918615293502808,
      "epoch": 4.546041626142774,
      "grad_norm": 1.34375,
      "learning_rate": 0.000309951363789846,
      "loss": 4.963,
      "mean_token_accuracy": 0.21680498421192168,
      "num_tokens": 101350595.0,
      "step": 58430
    },
    {
      "entropy": 5.891674900054932,
      "epoch": 4.5464306555144915,
      "grad_norm": 1.515625,
      "learning_rate": 0.00030992398437218314,
      "loss": 4.8804,
      "mean_token_accuracy": 0.22304200381040573,
      "num_tokens": 101359085.0,
      "step": 58435
    },
    {
      "entropy": 5.940161228179932,
      "epoch": 4.546819684886209,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003098966044246045,
      "loss": 4.9833,
      "mean_token_accuracy": 0.2095121443271637,
      "num_tokens": 101367965.0,
      "step": 58440
    },
    {
      "entropy": 5.977306604385376,
      "epoch": 4.547208714257927,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030986922394752534,
      "loss": 4.9847,
      "mean_token_accuracy": 0.22229924201965331,
      "num_tokens": 101376986.0,
      "step": 58445
    },
    {
      "entropy": 5.981124639511108,
      "epoch": 4.547597743629644,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00030984184294136134,
      "loss": 4.9742,
      "mean_token_accuracy": 0.2147505760192871,
      "num_tokens": 101385691.0,
      "step": 58450
    },
    {
      "entropy": 5.875857543945313,
      "epoch": 4.5479867730013614,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030981446140652773,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22326604276895523,
      "num_tokens": 101394328.0,
      "step": 58455
    },
    {
      "entropy": 5.981966829299926,
      "epoch": 4.548375802373079,
      "grad_norm": 1.53125,
      "learning_rate": 0.00030978707934344014,
      "loss": 4.9791,
      "mean_token_accuracy": 0.2144719272851944,
      "num_tokens": 101403165.0,
      "step": 58460
    },
    {
      "entropy": 5.94707145690918,
      "epoch": 4.548764831744797,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000309759696752514,
      "loss": 4.9401,
      "mean_token_accuracy": 0.21545272022485734,
      "num_tokens": 101411609.0,
      "step": 58465
    },
    {
      "entropy": 5.886676406860351,
      "epoch": 4.5491538611165145,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003097323136341647,
      "loss": 4.8934,
      "mean_token_accuracy": 0.22017668187618256,
      "num_tokens": 101420609.0,
      "step": 58470
    },
    {
      "entropy": 5.958272314071655,
      "epoch": 4.549542890488232,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003097049299888078,
      "loss": 5.0345,
      "mean_token_accuracy": 0.21841762363910674,
      "num_tokens": 101428931.0,
      "step": 58475
    },
    {
      "entropy": 5.964497995376587,
      "epoch": 4.549931919859949,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003096775458168589,
      "loss": 5.0108,
      "mean_token_accuracy": 0.20915891826152802,
      "num_tokens": 101438329.0,
      "step": 58480
    },
    {
      "entropy": 5.96292552947998,
      "epoch": 4.550320949231667,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00030965016111873345,
      "loss": 4.9305,
      "mean_token_accuracy": 0.21661896854639054,
      "num_tokens": 101446453.0,
      "step": 58485
    },
    {
      "entropy": 5.978559923171997,
      "epoch": 4.550709978603384,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030962277589484697,
      "loss": 4.9775,
      "mean_token_accuracy": 0.21155090183019637,
      "num_tokens": 101455368.0,
      "step": 58490
    },
    {
      "entropy": 5.937911319732666,
      "epoch": 4.551099007975102,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00030959539014561494,
      "loss": 4.9788,
      "mean_token_accuracy": 0.21017447263002395,
      "num_tokens": 101463573.0,
      "step": 58495
    },
    {
      "entropy": 5.899010419845581,
      "epoch": 4.55148803734682,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030956800387145294,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21537830233573912,
      "num_tokens": 101471534.0,
      "step": 58500
    },
    {
      "entropy": 5.942011976242066,
      "epoch": 4.5518770667185375,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030954061707277647,
      "loss": 4.9956,
      "mean_token_accuracy": 0.215328411757946,
      "num_tokens": 101479617.0,
      "step": 58505
    },
    {
      "entropy": 5.919935989379883,
      "epoch": 4.552266096090255,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003095132297500011,
      "loss": 4.931,
      "mean_token_accuracy": 0.22263543754816056,
      "num_tokens": 101487947.0,
      "step": 58510
    },
    {
      "entropy": 5.925440311431885,
      "epoch": 4.552655125461972,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030948584190354246,
      "loss": 4.9833,
      "mean_token_accuracy": 0.2216104283928871,
      "num_tokens": 101496542.0,
      "step": 58515
    },
    {
      "entropy": 5.95532078742981,
      "epoch": 4.55304415483369,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000309458453533816,
      "loss": 5.0086,
      "mean_token_accuracy": 0.21065550148487092,
      "num_tokens": 101505552.0,
      "step": 58520
    },
    {
      "entropy": 5.971432685852051,
      "epoch": 4.553433184205407,
      "grad_norm": 1.515625,
      "learning_rate": 0.00030943106464123744,
      "loss": 4.9357,
      "mean_token_accuracy": 0.21792164593935012,
      "num_tokens": 101514612.0,
      "step": 58525
    },
    {
      "entropy": 5.927151393890381,
      "epoch": 4.553822213577125,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030940367522622224,
      "loss": 4.9997,
      "mean_token_accuracy": 0.2211836576461792,
      "num_tokens": 101524316.0,
      "step": 58530
    },
    {
      "entropy": 5.947913932800293,
      "epoch": 4.554211242948843,
      "grad_norm": 1.421875,
      "learning_rate": 0.000309376285289186,
      "loss": 4.9813,
      "mean_token_accuracy": 0.21202642768621444,
      "num_tokens": 101532905.0,
      "step": 58535
    },
    {
      "entropy": 6.004165363311768,
      "epoch": 4.5546002723205605,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030934889483054433,
      "loss": 5.0436,
      "mean_token_accuracy": 0.20984185189008714,
      "num_tokens": 101541561.0,
      "step": 58540
    },
    {
      "entropy": 6.053962135314942,
      "epoch": 4.554989301692277,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003093215038507129,
      "loss": 5.108,
      "mean_token_accuracy": 0.19978517293930054,
      "num_tokens": 101550774.0,
      "step": 58545
    },
    {
      "entropy": 6.016974306106567,
      "epoch": 4.555378331063995,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003092941123501073,
      "loss": 5.0188,
      "mean_token_accuracy": 0.2169467940926552,
      "num_tokens": 101559562.0,
      "step": 58550
    },
    {
      "entropy": 6.0086188316345215,
      "epoch": 4.555767360435713,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030926672032914306,
      "loss": 4.9776,
      "mean_token_accuracy": 0.2118993103504181,
      "num_tokens": 101568458.0,
      "step": 58555
    },
    {
      "entropy": 5.920924043655395,
      "epoch": 4.55615638980743,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003092393277882359,
      "loss": 4.9403,
      "mean_token_accuracy": 0.2113639548420906,
      "num_tokens": 101577396.0,
      "step": 58560
    },
    {
      "entropy": 5.918664216995239,
      "epoch": 4.556545419179148,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030921193472780144,
      "loss": 4.9564,
      "mean_token_accuracy": 0.2137495383620262,
      "num_tokens": 101585464.0,
      "step": 58565
    },
    {
      "entropy": 5.87604513168335,
      "epoch": 4.556934448550866,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003091845411482554,
      "loss": 4.8547,
      "mean_token_accuracy": 0.22694648653268815,
      "num_tokens": 101593663.0,
      "step": 58570
    },
    {
      "entropy": 5.9865258693695065,
      "epoch": 4.5573234779225835,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030915714705001323,
      "loss": 5.0403,
      "mean_token_accuracy": 0.20996366292238236,
      "num_tokens": 101602678.0,
      "step": 58575
    },
    {
      "entropy": 6.012033605575562,
      "epoch": 4.557712507294301,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003091297524334908,
      "loss": 5.0315,
      "mean_token_accuracy": 0.2138388454914093,
      "num_tokens": 101610957.0,
      "step": 58580
    },
    {
      "entropy": 5.9539727687835695,
      "epoch": 4.558101536666018,
      "grad_norm": 1.5,
      "learning_rate": 0.0003091023572991036,
      "loss": 4.9373,
      "mean_token_accuracy": 0.2173429936170578,
      "num_tokens": 101619770.0,
      "step": 58585
    },
    {
      "entropy": 5.9686966896057125,
      "epoch": 4.558490566037736,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003090749616472675,
      "loss": 5.0585,
      "mean_token_accuracy": 0.2064809024333954,
      "num_tokens": 101628675.0,
      "step": 58590
    },
    {
      "entropy": 5.955616188049317,
      "epoch": 4.558879595409453,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003090475654783981,
      "loss": 4.9905,
      "mean_token_accuracy": 0.21637198328971863,
      "num_tokens": 101637861.0,
      "step": 58595
    },
    {
      "entropy": 5.839438772201538,
      "epoch": 4.559268624781171,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003090201687929111,
      "loss": 4.8908,
      "mean_token_accuracy": 0.2179200083017349,
      "num_tokens": 101645825.0,
      "step": 58600
    },
    {
      "entropy": 5.938732004165649,
      "epoch": 4.559657654152889,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00030899277159122213,
      "loss": 5.0206,
      "mean_token_accuracy": 0.2175292953848839,
      "num_tokens": 101654186.0,
      "step": 58605
    },
    {
      "entropy": 6.023279285430908,
      "epoch": 4.560046683524606,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030896537387374703,
      "loss": 5.1536,
      "mean_token_accuracy": 0.19467951506376266,
      "num_tokens": 101663195.0,
      "step": 58610
    },
    {
      "entropy": 5.9311261653900145,
      "epoch": 4.560435712896323,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00030893797564090136,
      "loss": 4.9585,
      "mean_token_accuracy": 0.21384785026311875,
      "num_tokens": 101672070.0,
      "step": 58615
    },
    {
      "entropy": 5.935377836227417,
      "epoch": 4.560824742268041,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030891057689310097,
      "loss": 4.9608,
      "mean_token_accuracy": 0.21435239166021347,
      "num_tokens": 101680738.0,
      "step": 58620
    },
    {
      "entropy": 5.950182008743286,
      "epoch": 4.561213771639759,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003088831776307616,
      "loss": 4.9857,
      "mean_token_accuracy": 0.2137000545859337,
      "num_tokens": 101688473.0,
      "step": 58625
    },
    {
      "entropy": 6.009902143478394,
      "epoch": 4.561602801011476,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030885577785429885,
      "loss": 5.0309,
      "mean_token_accuracy": 0.2093917101621628,
      "num_tokens": 101698129.0,
      "step": 58630
    },
    {
      "entropy": 5.942453670501709,
      "epoch": 4.561991830383194,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003088283775641286,
      "loss": 4.92,
      "mean_token_accuracy": 0.21736256182193756,
      "num_tokens": 101707096.0,
      "step": 58635
    },
    {
      "entropy": 5.981146049499512,
      "epoch": 4.562380859754912,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030880097676066646,
      "loss": 4.994,
      "mean_token_accuracy": 0.21264870315790177,
      "num_tokens": 101715173.0,
      "step": 58640
    },
    {
      "entropy": 5.9368593215942385,
      "epoch": 4.5627698891266295,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003087735754443285,
      "loss": 4.9945,
      "mean_token_accuracy": 0.2111407458782196,
      "num_tokens": 101723396.0,
      "step": 58645
    },
    {
      "entropy": 5.983474254608154,
      "epoch": 4.563158918498346,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003087461736155301,
      "loss": 5.0161,
      "mean_token_accuracy": 0.21227810978889466,
      "num_tokens": 101732143.0,
      "step": 58650
    },
    {
      "entropy": 5.957684421539307,
      "epoch": 4.563547947870064,
      "grad_norm": 1.5,
      "learning_rate": 0.0003087187712746873,
      "loss": 4.9653,
      "mean_token_accuracy": 0.2161446198821068,
      "num_tokens": 101740769.0,
      "step": 58655
    },
    {
      "entropy": 5.8591326713562015,
      "epoch": 4.563936977241782,
      "grad_norm": 1.5,
      "learning_rate": 0.0003086913684222159,
      "loss": 4.9152,
      "mean_token_accuracy": 0.22071201652288436,
      "num_tokens": 101749518.0,
      "step": 58660
    },
    {
      "entropy": 5.964906787872314,
      "epoch": 4.564326006613499,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003086639650585315,
      "loss": 4.9602,
      "mean_token_accuracy": 0.2245492309331894,
      "num_tokens": 101757891.0,
      "step": 58665
    },
    {
      "entropy": 5.954412221908569,
      "epoch": 4.564715035985217,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030863656118405007,
      "loss": 4.8801,
      "mean_token_accuracy": 0.21600429862737655,
      "num_tokens": 101766393.0,
      "step": 58670
    },
    {
      "entropy": 5.94890022277832,
      "epoch": 4.565104065356935,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030860915679918724,
      "loss": 4.9703,
      "mean_token_accuracy": 0.21318492144346238,
      "num_tokens": 101774896.0,
      "step": 58675
    },
    {
      "entropy": 5.912567281723023,
      "epoch": 4.565493094728652,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030858175190435913,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21803636401891707,
      "num_tokens": 101784175.0,
      "step": 58680
    },
    {
      "entropy": 5.929139471054077,
      "epoch": 4.565882124100369,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003085543464999813,
      "loss": 4.9386,
      "mean_token_accuracy": 0.2167488992214203,
      "num_tokens": 101792960.0,
      "step": 58685
    },
    {
      "entropy": 5.950344800949097,
      "epoch": 4.566271153472087,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003085269405864697,
      "loss": 4.9511,
      "mean_token_accuracy": 0.21659803986549378,
      "num_tokens": 101801777.0,
      "step": 58690
    },
    {
      "entropy": 5.943732500076294,
      "epoch": 4.566660182843805,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003084995341642401,
      "loss": 4.9764,
      "mean_token_accuracy": 0.21218783110380174,
      "num_tokens": 101810235.0,
      "step": 58695
    },
    {
      "entropy": 5.951359415054322,
      "epoch": 4.567049212215522,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003084721272337084,
      "loss": 4.9982,
      "mean_token_accuracy": 0.20874246954917908,
      "num_tokens": 101818203.0,
      "step": 58700
    },
    {
      "entropy": 5.865972805023193,
      "epoch": 4.56743824158724,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003084447197952904,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22032213509082793,
      "num_tokens": 101826438.0,
      "step": 58705
    },
    {
      "entropy": 5.992106819152832,
      "epoch": 4.567827270958958,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030841731184940205,
      "loss": 5.0464,
      "mean_token_accuracy": 0.2079189494252205,
      "num_tokens": 101835133.0,
      "step": 58710
    },
    {
      "entropy": 5.928923892974853,
      "epoch": 4.568216300330675,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00030838990339645915,
      "loss": 4.9076,
      "mean_token_accuracy": 0.219659623503685,
      "num_tokens": 101843539.0,
      "step": 58715
    },
    {
      "entropy": 5.92302827835083,
      "epoch": 4.568605329702392,
      "grad_norm": 1.484375,
      "learning_rate": 0.00030836249443687774,
      "loss": 4.9857,
      "mean_token_accuracy": 0.21148117631673813,
      "num_tokens": 101851469.0,
      "step": 58720
    },
    {
      "entropy": 5.93510684967041,
      "epoch": 4.56899435907411,
      "grad_norm": 1.484375,
      "learning_rate": 0.00030833508497107344,
      "loss": 5.0079,
      "mean_token_accuracy": 0.21083535254001617,
      "num_tokens": 101859346.0,
      "step": 58725
    },
    {
      "entropy": 5.960146522521972,
      "epoch": 4.569383388445828,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003083076749994623,
      "loss": 4.9788,
      "mean_token_accuracy": 0.21525996625423433,
      "num_tokens": 101867574.0,
      "step": 58730
    },
    {
      "entropy": 5.915790319442749,
      "epoch": 4.569772417817545,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003082802645224603,
      "loss": 5.0263,
      "mean_token_accuracy": 0.21086502969264984,
      "num_tokens": 101877147.0,
      "step": 58735
    },
    {
      "entropy": 5.910672283172607,
      "epoch": 4.570161447189263,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003082528535404832,
      "loss": 4.92,
      "mean_token_accuracy": 0.22247439175844191,
      "num_tokens": 101885486.0,
      "step": 58740
    },
    {
      "entropy": 5.967355728149414,
      "epoch": 4.57055047656098,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030822544205394686,
      "loss": 5.0241,
      "mean_token_accuracy": 0.20442670434713364,
      "num_tokens": 101893651.0,
      "step": 58745
    },
    {
      "entropy": 5.93948221206665,
      "epoch": 4.570939505932698,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030819803006326745,
      "loss": 4.9586,
      "mean_token_accuracy": 0.21547514647245408,
      "num_tokens": 101901709.0,
      "step": 58750
    },
    {
      "entropy": 5.971761322021484,
      "epoch": 4.571328535304415,
      "grad_norm": 1.5,
      "learning_rate": 0.00030817061756886077,
      "loss": 4.9178,
      "mean_token_accuracy": 0.22071709483861923,
      "num_tokens": 101909568.0,
      "step": 58755
    },
    {
      "entropy": 5.989965200424194,
      "epoch": 4.571717564676133,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030814320457114283,
      "loss": 5.049,
      "mean_token_accuracy": 0.2139080986380577,
      "num_tokens": 101918269.0,
      "step": 58760
    },
    {
      "entropy": 5.924484205245972,
      "epoch": 4.572106594047851,
      "grad_norm": 1.375,
      "learning_rate": 0.00030811579107052937,
      "loss": 4.9454,
      "mean_token_accuracy": 0.20906743854284288,
      "num_tokens": 101926666.0,
      "step": 58765
    },
    {
      "entropy": 5.946925401687622,
      "epoch": 4.572495623419568,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030808837706743666,
      "loss": 5.0573,
      "mean_token_accuracy": 0.20284990221261978,
      "num_tokens": 101936667.0,
      "step": 58770
    },
    {
      "entropy": 6.009098672866822,
      "epoch": 4.572884652791286,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003080609625622803,
      "loss": 4.9321,
      "mean_token_accuracy": 0.22043293863534927,
      "num_tokens": 101945909.0,
      "step": 58775
    },
    {
      "entropy": 5.96932282447815,
      "epoch": 4.573273682163004,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030803354755547653,
      "loss": 5.0081,
      "mean_token_accuracy": 0.21502549946308136,
      "num_tokens": 101954511.0,
      "step": 58780
    },
    {
      "entropy": 5.976952362060547,
      "epoch": 4.573662711534721,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003080061320474412,
      "loss": 4.9905,
      "mean_token_accuracy": 0.2058105483651161,
      "num_tokens": 101962879.0,
      "step": 58785
    },
    {
      "entropy": 5.910180568695068,
      "epoch": 4.574051740906438,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003079787160385905,
      "loss": 4.8725,
      "mean_token_accuracy": 0.23114990890026094,
      "num_tokens": 101971778.0,
      "step": 58790
    },
    {
      "entropy": 5.998329830169678,
      "epoch": 4.574440770278156,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003079512995293402,
      "loss": 5.0435,
      "mean_token_accuracy": 0.2102557361125946,
      "num_tokens": 101980483.0,
      "step": 58795
    },
    {
      "entropy": 5.93962779045105,
      "epoch": 4.574829799649874,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003079238825201064,
      "loss": 4.9293,
      "mean_token_accuracy": 0.21602404862642288,
      "num_tokens": 101989715.0,
      "step": 58800
    },
    {
      "entropy": 5.929000329971314,
      "epoch": 4.575218829021591,
      "grad_norm": 1.40625,
      "learning_rate": 0.000307896465011305,
      "loss": 5.015,
      "mean_token_accuracy": 0.2163167729973793,
      "num_tokens": 101998296.0,
      "step": 58805
    },
    {
      "entropy": 5.96214451789856,
      "epoch": 4.575607858393309,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00030786904700335217,
      "loss": 4.9869,
      "mean_token_accuracy": 0.2158162921667099,
      "num_tokens": 102006610.0,
      "step": 58810
    },
    {
      "entropy": 5.930808877944946,
      "epoch": 4.575996887765026,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003078416284966639,
      "loss": 4.8665,
      "mean_token_accuracy": 0.22697226107120513,
      "num_tokens": 102014168.0,
      "step": 58815
    },
    {
      "entropy": 5.913540029525757,
      "epoch": 4.576385917136744,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030781420949165607,
      "loss": 4.985,
      "mean_token_accuracy": 0.21210678964853286,
      "num_tokens": 102023292.0,
      "step": 58820
    },
    {
      "entropy": 5.976796007156372,
      "epoch": 4.576774946508461,
      "grad_norm": 1.375,
      "learning_rate": 0.00030778678998874494,
      "loss": 4.9979,
      "mean_token_accuracy": 0.2142992153763771,
      "num_tokens": 102032366.0,
      "step": 58825
    },
    {
      "entropy": 5.99838662147522,
      "epoch": 4.577163975880179,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003077593699883465,
      "loss": 4.9733,
      "mean_token_accuracy": 0.2138890191912651,
      "num_tokens": 102040705.0,
      "step": 58830
    },
    {
      "entropy": 5.987545204162598,
      "epoch": 4.577553005251897,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00030773194949087665,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21598704606294633,
      "num_tokens": 102049691.0,
      "step": 58835
    },
    {
      "entropy": 6.008904314041137,
      "epoch": 4.577942034623614,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003077045284967516,
      "loss": 5.051,
      "mean_token_accuracy": 0.20991770923137665,
      "num_tokens": 102058016.0,
      "step": 58840
    },
    {
      "entropy": 5.886815452575684,
      "epoch": 4.578331063995332,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003076771070063874,
      "loss": 4.9387,
      "mean_token_accuracy": 0.2133013591170311,
      "num_tokens": 102066814.0,
      "step": 58845
    },
    {
      "entropy": 6.04103422164917,
      "epoch": 4.578720093367049,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003076496850202001,
      "loss": 5.1665,
      "mean_token_accuracy": 0.20664051473140715,
      "num_tokens": 102075255.0,
      "step": 58850
    },
    {
      "entropy": 6.011362791061401,
      "epoch": 4.579109122738767,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00030762226253860584,
      "loss": 4.9568,
      "mean_token_accuracy": 0.21911944597959518,
      "num_tokens": 102083807.0,
      "step": 58855
    },
    {
      "entropy": 5.965685892105102,
      "epoch": 4.579498152110484,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0003075948395620206,
      "loss": 4.991,
      "mean_token_accuracy": 0.21085403859615326,
      "num_tokens": 102093185.0,
      "step": 58860
    },
    {
      "entropy": 6.017434978485108,
      "epoch": 4.579887181482202,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030756741609086057,
      "loss": 5.0813,
      "mean_token_accuracy": 0.21438451409339904,
      "num_tokens": 102101841.0,
      "step": 58865
    },
    {
      "entropy": 5.954827499389649,
      "epoch": 4.58027621085392,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00030753999212554183,
      "loss": 5.0078,
      "mean_token_accuracy": 0.21055535078048707,
      "num_tokens": 102111190.0,
      "step": 58870
    },
    {
      "entropy": 6.018886661529541,
      "epoch": 4.580665240225637,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003075125676664806,
      "loss": 5.0553,
      "mean_token_accuracy": 0.2144513726234436,
      "num_tokens": 102119870.0,
      "step": 58875
    },
    {
      "entropy": 5.96637864112854,
      "epoch": 4.581054269597354,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003074851427140928,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21294522136449814,
      "num_tokens": 102129372.0,
      "step": 58880
    },
    {
      "entropy": 5.908979606628418,
      "epoch": 4.581443298969072,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030745771726879467,
      "loss": 4.9073,
      "mean_token_accuracy": 0.21978216022253036,
      "num_tokens": 102137488.0,
      "step": 58885
    },
    {
      "entropy": 5.921315383911133,
      "epoch": 4.58183232834079,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003074302913310024,
      "loss": 5.0251,
      "mean_token_accuracy": 0.2060428589582443,
      "num_tokens": 102146245.0,
      "step": 58890
    },
    {
      "entropy": 5.924926662445069,
      "epoch": 4.582221357712507,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000307402864901132,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2214808776974678,
      "num_tokens": 102154950.0,
      "step": 58895
    },
    {
      "entropy": 5.949667119979859,
      "epoch": 4.582610387084225,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003073754379795998,
      "loss": 5.0362,
      "mean_token_accuracy": 0.21295692026615143,
      "num_tokens": 102163887.0,
      "step": 58900
    },
    {
      "entropy": 5.973671627044678,
      "epoch": 4.582999416455943,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030734801056682176,
      "loss": 5.0629,
      "mean_token_accuracy": 0.20917702168226243,
      "num_tokens": 102172864.0,
      "step": 58905
    },
    {
      "entropy": 5.946758460998535,
      "epoch": 4.58338844582766,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030732058266321415,
      "loss": 4.9094,
      "mean_token_accuracy": 0.22059286683797835,
      "num_tokens": 102181167.0,
      "step": 58910
    },
    {
      "entropy": 5.96749792098999,
      "epoch": 4.583777475199377,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0003072931542691932,
      "loss": 5.0316,
      "mean_token_accuracy": 0.21100189089775084,
      "num_tokens": 102189801.0,
      "step": 58915
    },
    {
      "entropy": 5.966647529602051,
      "epoch": 4.584166504571095,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000307265725385175,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2219487875699997,
      "num_tokens": 102198477.0,
      "step": 58920
    },
    {
      "entropy": 5.97558364868164,
      "epoch": 4.584555533942813,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003072382960115758,
      "loss": 4.9457,
      "mean_token_accuracy": 0.21573416143655777,
      "num_tokens": 102207763.0,
      "step": 58925
    },
    {
      "entropy": 6.023884677886963,
      "epoch": 4.58494456331453,
      "grad_norm": 1.484375,
      "learning_rate": 0.00030721086614881183,
      "loss": 4.9926,
      "mean_token_accuracy": 0.214885114133358,
      "num_tokens": 102216143.0,
      "step": 58930
    },
    {
      "entropy": 5.955588150024414,
      "epoch": 4.585333592686248,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003071834357972991,
      "loss": 4.9987,
      "mean_token_accuracy": 0.21656915545463562,
      "num_tokens": 102225127.0,
      "step": 58935
    },
    {
      "entropy": 5.909287405014038,
      "epoch": 4.585722622057966,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000307156004957454,
      "loss": 4.9061,
      "mean_token_accuracy": 0.21914063394069672,
      "num_tokens": 102234210.0,
      "step": 58940
    },
    {
      "entropy": 5.895959711074829,
      "epoch": 4.5861116514296825,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003071285736296928,
      "loss": 4.8759,
      "mean_token_accuracy": 0.2168147772550583,
      "num_tokens": 102242382.0,
      "step": 58945
    },
    {
      "entropy": 5.864328336715698,
      "epoch": 4.5865006808014,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003071011418144315,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21552136689424514,
      "num_tokens": 102251426.0,
      "step": 58950
    },
    {
      "entropy": 5.98614182472229,
      "epoch": 4.586889710173118,
      "grad_norm": 1.375,
      "learning_rate": 0.00030707370951208655,
      "loss": 5.0462,
      "mean_token_accuracy": 0.2112046629190445,
      "num_tokens": 102260246.0,
      "step": 58955
    },
    {
      "entropy": 6.004589509963989,
      "epoch": 4.587278739544836,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003070462767230741,
      "loss": 4.9955,
      "mean_token_accuracy": 0.2119741529226303,
      "num_tokens": 102268804.0,
      "step": 58960
    },
    {
      "entropy": 5.917537117004395,
      "epoch": 4.587667768916553,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003070188434478104,
      "loss": 4.8566,
      "mean_token_accuracy": 0.2214253842830658,
      "num_tokens": 102277188.0,
      "step": 58965
    },
    {
      "entropy": 5.981005239486694,
      "epoch": 4.588056798288271,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003069914096867117,
      "loss": 5.0414,
      "mean_token_accuracy": 0.21086664795875548,
      "num_tokens": 102286192.0,
      "step": 58970
    },
    {
      "entropy": 5.957050085067749,
      "epoch": 4.588445827659989,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003069639754401942,
      "loss": 5.0144,
      "mean_token_accuracy": 0.21152419000864028,
      "num_tokens": 102294687.0,
      "step": 58975
    },
    {
      "entropy": 5.8860499382019045,
      "epoch": 4.588834857031706,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003069365407086744,
      "loss": 4.9552,
      "mean_token_accuracy": 0.20674492716789244,
      "num_tokens": 102303490.0,
      "step": 58980
    },
    {
      "entropy": 6.054447841644287,
      "epoch": 4.589223886403423,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030690910549256836,
      "loss": 5.1261,
      "mean_token_accuracy": 0.19932192265987397,
      "num_tokens": 102311993.0,
      "step": 58985
    },
    {
      "entropy": 6.030044507980347,
      "epoch": 4.589612915775141,
      "grad_norm": 1.375,
      "learning_rate": 0.0003068816697922925,
      "loss": 4.9929,
      "mean_token_accuracy": 0.2106467977166176,
      "num_tokens": 102320544.0,
      "step": 58990
    },
    {
      "entropy": 5.97399320602417,
      "epoch": 4.5900019451468586,
      "grad_norm": 1.375,
      "learning_rate": 0.0003068542336082629,
      "loss": 5.0269,
      "mean_token_accuracy": 0.20406902432441712,
      "num_tokens": 102329192.0,
      "step": 58995
    },
    {
      "entropy": 5.9314109802246096,
      "epoch": 4.590390974518576,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030682679694089615,
      "loss": 4.9813,
      "mean_token_accuracy": 0.21335501372814178,
      "num_tokens": 102338146.0,
      "step": 59000
    },
    {
      "entropy": 5.908329677581787,
      "epoch": 4.590780003890294,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0003067993597906083,
      "loss": 5.0343,
      "mean_token_accuracy": 0.2059118077158928,
      "num_tokens": 102346883.0,
      "step": 59005
    },
    {
      "entropy": 5.98775086402893,
      "epoch": 4.591169033262012,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003067719221578158,
      "loss": 5.0255,
      "mean_token_accuracy": 0.21922040432691575,
      "num_tokens": 102355796.0,
      "step": 59010
    },
    {
      "entropy": 6.005100345611572,
      "epoch": 4.5915580626337285,
      "grad_norm": 1.5,
      "learning_rate": 0.000306744484042935,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2193056657910347,
      "num_tokens": 102364680.0,
      "step": 59015
    },
    {
      "entropy": 6.015369558334351,
      "epoch": 4.591947092005446,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003067170454463822,
      "loss": 5.0604,
      "mean_token_accuracy": 0.21222323477268218,
      "num_tokens": 102373218.0,
      "step": 59020
    },
    {
      "entropy": 5.991926383972168,
      "epoch": 4.592336121377164,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00030668960636857364,
      "loss": 5.0433,
      "mean_token_accuracy": 0.21496972143650056,
      "num_tokens": 102381839.0,
      "step": 59025
    },
    {
      "entropy": 6.023768758773803,
      "epoch": 4.5927251507488815,
      "grad_norm": 1.609375,
      "learning_rate": 0.0003066621668099258,
      "loss": 5.0982,
      "mean_token_accuracy": 0.20862198621034622,
      "num_tokens": 102390796.0,
      "step": 59030
    },
    {
      "entropy": 5.982073497772217,
      "epoch": 4.593114180120599,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030663472677085494,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21581994891166686,
      "num_tokens": 102399342.0,
      "step": 59035
    },
    {
      "entropy": 5.978408575057983,
      "epoch": 4.593503209492317,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030660728625177746,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2189146488904953,
      "num_tokens": 102408075.0,
      "step": 59040
    },
    {
      "entropy": 5.935489082336426,
      "epoch": 4.593892238864035,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003065798452531097,
      "loss": 4.9509,
      "mean_token_accuracy": 0.2179875984787941,
      "num_tokens": 102416484.0,
      "step": 59045
    },
    {
      "entropy": 5.981276273727417,
      "epoch": 4.5942812682357514,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030655240377526813,
      "loss": 5.0497,
      "mean_token_accuracy": 0.20332966446876527,
      "num_tokens": 102424545.0,
      "step": 59050
    },
    {
      "entropy": 5.968435192108155,
      "epoch": 4.594670297607469,
      "grad_norm": 1.515625,
      "learning_rate": 0.000306524961818669,
      "loss": 5.0453,
      "mean_token_accuracy": 0.21243750751018525,
      "num_tokens": 102433019.0,
      "step": 59055
    },
    {
      "entropy": 5.886138534545898,
      "epoch": 4.595059326979187,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003064975193837287,
      "loss": 4.8677,
      "mean_token_accuracy": 0.21428215205669404,
      "num_tokens": 102441350.0,
      "step": 59060
    },
    {
      "entropy": 5.992455196380615,
      "epoch": 4.5954483563509045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003064700764708638,
      "loss": 5.0358,
      "mean_token_accuracy": 0.21895599216222764,
      "num_tokens": 102449759.0,
      "step": 59065
    },
    {
      "entropy": 5.935897016525269,
      "epoch": 4.595837385722622,
      "grad_norm": 1.484375,
      "learning_rate": 0.00030644263308049046,
      "loss": 4.8599,
      "mean_token_accuracy": 0.22000134289264678,
      "num_tokens": 102457755.0,
      "step": 59070
    },
    {
      "entropy": 5.939029788970947,
      "epoch": 4.59622641509434,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003064151892130252,
      "loss": 4.935,
      "mean_token_accuracy": 0.22259055972099304,
      "num_tokens": 102466986.0,
      "step": 59075
    },
    {
      "entropy": 5.913162612915039,
      "epoch": 4.596615444466057,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003063877448688846,
      "loss": 4.9882,
      "mean_token_accuracy": 0.2092021018266678,
      "num_tokens": 102475625.0,
      "step": 59080
    },
    {
      "entropy": 5.970071315765381,
      "epoch": 4.597004473837774,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003063603000484848,
      "loss": 4.9949,
      "mean_token_accuracy": 0.20898510366678238,
      "num_tokens": 102484365.0,
      "step": 59085
    },
    {
      "entropy": 5.9684759140014645,
      "epoch": 4.597393503209492,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030633285475224244,
      "loss": 5.0358,
      "mean_token_accuracy": 0.2080918163061142,
      "num_tokens": 102492948.0,
      "step": 59090
    },
    {
      "entropy": 5.946693992614746,
      "epoch": 4.59778253258121,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030630540898057384,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21360691040754318,
      "num_tokens": 102501250.0,
      "step": 59095
    },
    {
      "entropy": 5.909185552597046,
      "epoch": 4.5981715619529275,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030627796273389546,
      "loss": 4.969,
      "mean_token_accuracy": 0.20937100797891617,
      "num_tokens": 102510258.0,
      "step": 59100
    },
    {
      "entropy": 5.893084001541138,
      "epoch": 4.598560591324645,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00030625051601262384,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21534937471151352,
      "num_tokens": 102518600.0,
      "step": 59105
    },
    {
      "entropy": 5.990673208236695,
      "epoch": 4.598949620696363,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003062230688171753,
      "loss": 5.0295,
      "mean_token_accuracy": 0.2044035792350769,
      "num_tokens": 102526690.0,
      "step": 59110
    },
    {
      "entropy": 5.962455177307129,
      "epoch": 4.599338650068081,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003061956211479665,
      "loss": 4.9554,
      "mean_token_accuracy": 0.21708112359046935,
      "num_tokens": 102535896.0,
      "step": 59115
    },
    {
      "entropy": 5.98532133102417,
      "epoch": 4.599727679439797,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003061681730054137,
      "loss": 5.0638,
      "mean_token_accuracy": 0.19663383662700654,
      "num_tokens": 102544442.0,
      "step": 59120
    },
    {
      "entropy": 5.938311576843262,
      "epoch": 4.600116708811515,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003061407243899335,
      "loss": 4.9143,
      "mean_token_accuracy": 0.21889810562133788,
      "num_tokens": 102552611.0,
      "step": 59125
    },
    {
      "entropy": 5.9978663444519045,
      "epoch": 4.600505738183233,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030611327530194243,
      "loss": 5.018,
      "mean_token_accuracy": 0.20889406502246857,
      "num_tokens": 102561009.0,
      "step": 59130
    },
    {
      "entropy": 5.943119049072266,
      "epoch": 4.6008947675549505,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003060858257418569,
      "loss": 5.0249,
      "mean_token_accuracy": 0.2078838512301445,
      "num_tokens": 102571001.0,
      "step": 59135
    },
    {
      "entropy": 5.924479007720947,
      "epoch": 4.601283796926668,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003060583757100934,
      "loss": 4.9225,
      "mean_token_accuracy": 0.22280605882406235,
      "num_tokens": 102579569.0,
      "step": 59140
    },
    {
      "entropy": 6.034378862380981,
      "epoch": 4.601672826298385,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003060309252070685,
      "loss": 5.1273,
      "mean_token_accuracy": 0.20755570530891418,
      "num_tokens": 102589120.0,
      "step": 59145
    },
    {
      "entropy": 5.96711049079895,
      "epoch": 4.602061855670103,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003060034742331987,
      "loss": 5.0506,
      "mean_token_accuracy": 0.20910661071538925,
      "num_tokens": 102598708.0,
      "step": 59150
    },
    {
      "entropy": 6.024827241897583,
      "epoch": 4.60245088504182,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003059760227889006,
      "loss": 5.0815,
      "mean_token_accuracy": 0.20881217420101167,
      "num_tokens": 102607053.0,
      "step": 59155
    },
    {
      "entropy": 5.860289096832275,
      "epoch": 4.602839914413538,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003059485708745906,
      "loss": 4.8367,
      "mean_token_accuracy": 0.22376374155282974,
      "num_tokens": 102615438.0,
      "step": 59160
    },
    {
      "entropy": 6.0010453224182125,
      "epoch": 4.603228943785256,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003059211184906853,
      "loss": 5.0177,
      "mean_token_accuracy": 0.20886463671922684,
      "num_tokens": 102623989.0,
      "step": 59165
    },
    {
      "entropy": 5.927425861358643,
      "epoch": 4.6036179731569735,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030589366563760125,
      "loss": 4.9463,
      "mean_token_accuracy": 0.2121858224272728,
      "num_tokens": 102633180.0,
      "step": 59170
    },
    {
      "entropy": 5.92069263458252,
      "epoch": 4.604007002528691,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00030586621231575494,
      "loss": 4.9244,
      "mean_token_accuracy": 0.22083283513784407,
      "num_tokens": 102642376.0,
      "step": 59175
    },
    {
      "entropy": 5.9898717403411865,
      "epoch": 4.604396031900409,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000305838758525563,
      "loss": 5.0214,
      "mean_token_accuracy": 0.20649200081825256,
      "num_tokens": 102651347.0,
      "step": 59180
    },
    {
      "entropy": 5.918342065811157,
      "epoch": 4.604785061272126,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003058113042674421,
      "loss": 4.9447,
      "mean_token_accuracy": 0.22265535145998,
      "num_tokens": 102659921.0,
      "step": 59185
    },
    {
      "entropy": 5.990040826797485,
      "epoch": 4.605174090643843,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030578384954180866,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20443608611822128,
      "num_tokens": 102668251.0,
      "step": 59190
    },
    {
      "entropy": 6.000902986526489,
      "epoch": 4.605563120015561,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003057563943490792,
      "loss": 5.0722,
      "mean_token_accuracy": 0.20436132848262786,
      "num_tokens": 102676858.0,
      "step": 59195
    },
    {
      "entropy": 5.973453426361084,
      "epoch": 4.605952149387279,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030572893868967044,
      "loss": 4.962,
      "mean_token_accuracy": 0.2142004609107971,
      "num_tokens": 102685283.0,
      "step": 59200
    },
    {
      "entropy": 6.042931365966797,
      "epoch": 4.6063411787589965,
      "grad_norm": 1.578125,
      "learning_rate": 0.000305701482563999,
      "loss": 5.0577,
      "mean_token_accuracy": 0.20779280811548234,
      "num_tokens": 102693500.0,
      "step": 59205
    },
    {
      "entropy": 5.941933441162109,
      "epoch": 4.606730208130714,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003056740259724814,
      "loss": 4.9521,
      "mean_token_accuracy": 0.22178158909082413,
      "num_tokens": 102702189.0,
      "step": 59210
    },
    {
      "entropy": 5.953090238571167,
      "epoch": 4.607119237502431,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003056465689155342,
      "loss": 5.025,
      "mean_token_accuracy": 0.21322276890277864,
      "num_tokens": 102711808.0,
      "step": 59215
    },
    {
      "entropy": 5.957609462738037,
      "epoch": 4.607508266874149,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030561911139357424,
      "loss": 5.034,
      "mean_token_accuracy": 0.20782601237297058,
      "num_tokens": 102720389.0,
      "step": 59220
    },
    {
      "entropy": 5.904064035415649,
      "epoch": 4.607897296245866,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030559165340701797,
      "loss": 4.8163,
      "mean_token_accuracy": 0.2218530982732773,
      "num_tokens": 102728696.0,
      "step": 59225
    },
    {
      "entropy": 5.966795253753662,
      "epoch": 4.608286325617584,
      "grad_norm": 1.359375,
      "learning_rate": 0.000305564194956282,
      "loss": 5.0477,
      "mean_token_accuracy": 0.20276754796504975,
      "num_tokens": 102737552.0,
      "step": 59230
    },
    {
      "entropy": 5.917603492736816,
      "epoch": 4.608675354989302,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00030553673604178305,
      "loss": 4.9305,
      "mean_token_accuracy": 0.21193026304244994,
      "num_tokens": 102746263.0,
      "step": 59235
    },
    {
      "entropy": 5.940702486038208,
      "epoch": 4.6090643843610195,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00030550927666393767,
      "loss": 5.0344,
      "mean_token_accuracy": 0.20888848453760148,
      "num_tokens": 102754774.0,
      "step": 59240
    },
    {
      "entropy": 5.943389415740967,
      "epoch": 4.609453413732737,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030548181682316263,
      "loss": 4.8567,
      "mean_token_accuracy": 0.2195924311876297,
      "num_tokens": 102763049.0,
      "step": 59245
    },
    {
      "entropy": 5.930915307998657,
      "epoch": 4.609842443104454,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030545435651987453,
      "loss": 4.9576,
      "mean_token_accuracy": 0.21346021294593812,
      "num_tokens": 102771995.0,
      "step": 59250
    },
    {
      "entropy": 5.893159675598144,
      "epoch": 4.610231472476172,
      "grad_norm": 1.5,
      "learning_rate": 0.0003054268957544901,
      "loss": 4.9813,
      "mean_token_accuracy": 0.20838552564382554,
      "num_tokens": 102780336.0,
      "step": 59255
    },
    {
      "entropy": 5.969234418869019,
      "epoch": 4.610620501847889,
      "grad_norm": 1.375,
      "learning_rate": 0.00030539943452742585,
      "loss": 5.0224,
      "mean_token_accuracy": 0.21511559188365936,
      "num_tokens": 102788881.0,
      "step": 59260
    },
    {
      "entropy": 5.94543719291687,
      "epoch": 4.611009531219607,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030537197283909875,
      "loss": 4.9053,
      "mean_token_accuracy": 0.2175610989332199,
      "num_tokens": 102797514.0,
      "step": 59265
    },
    {
      "entropy": 5.940406084060669,
      "epoch": 4.611398560591325,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030534451068992506,
      "loss": 5.0422,
      "mean_token_accuracy": 0.21846988946199417,
      "num_tokens": 102805567.0,
      "step": 59270
    },
    {
      "entropy": 5.91773247718811,
      "epoch": 4.6117875899630425,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030531704808032183,
      "loss": 4.9767,
      "mean_token_accuracy": 0.21391172409057618,
      "num_tokens": 102814073.0,
      "step": 59275
    },
    {
      "entropy": 5.89242639541626,
      "epoch": 4.612176619334759,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003052895850107057,
      "loss": 4.8341,
      "mean_token_accuracy": 0.22623540610074996,
      "num_tokens": 102822787.0,
      "step": 59280
    },
    {
      "entropy": 5.890420532226562,
      "epoch": 4.612565648706477,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030526212148149326,
      "loss": 4.8611,
      "mean_token_accuracy": 0.2226916655898094,
      "num_tokens": 102831411.0,
      "step": 59285
    },
    {
      "entropy": 5.938187980651856,
      "epoch": 4.612954678078195,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003052346574931013,
      "loss": 4.9602,
      "mean_token_accuracy": 0.2177382379770279,
      "num_tokens": 102839608.0,
      "step": 59290
    },
    {
      "entropy": 5.8607666015625,
      "epoch": 4.613343707449912,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003052071930459465,
      "loss": 4.9028,
      "mean_token_accuracy": 0.21924732774496078,
      "num_tokens": 102848335.0,
      "step": 59295
    },
    {
      "entropy": 5.938835525512696,
      "epoch": 4.61373273682163,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003051797281404457,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21824652701616287,
      "num_tokens": 102857376.0,
      "step": 59300
    },
    {
      "entropy": 6.000780773162842,
      "epoch": 4.614121766193348,
      "grad_norm": 1.40625,
      "learning_rate": 0.00030515226277701553,
      "loss": 5.0123,
      "mean_token_accuracy": 0.2085227370262146,
      "num_tokens": 102865832.0,
      "step": 59305
    },
    {
      "entropy": 6.012026023864746,
      "epoch": 4.6145107955650655,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003051247969560728,
      "loss": 5.0498,
      "mean_token_accuracy": 0.21119601428508758,
      "num_tokens": 102874555.0,
      "step": 59310
    },
    {
      "entropy": 5.930571174621582,
      "epoch": 4.614899824936783,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003050973306780342,
      "loss": 4.9693,
      "mean_token_accuracy": 0.21370158344507217,
      "num_tokens": 102883325.0,
      "step": 59315
    },
    {
      "entropy": 5.957516431808472,
      "epoch": 4.6152888543085,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030506986394331656,
      "loss": 5.0707,
      "mean_token_accuracy": 0.20858072638511657,
      "num_tokens": 102892712.0,
      "step": 59320
    },
    {
      "entropy": 5.91046109199524,
      "epoch": 4.615677883680218,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030504239675233655,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2140487924218178,
      "num_tokens": 102901502.0,
      "step": 59325
    },
    {
      "entropy": 5.984750461578369,
      "epoch": 4.616066913051935,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030501492910551094,
      "loss": 4.9704,
      "mean_token_accuracy": 0.21568020433187485,
      "num_tokens": 102910531.0,
      "step": 59330
    },
    {
      "entropy": 5.885979461669922,
      "epoch": 4.616455942423653,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030498746100325656,
      "loss": 4.8733,
      "mean_token_accuracy": 0.23082284182310103,
      "num_tokens": 102919884.0,
      "step": 59335
    },
    {
      "entropy": 5.987094068527222,
      "epoch": 4.616844971795371,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003049599924459903,
      "loss": 4.9749,
      "mean_token_accuracy": 0.21598953306674956,
      "num_tokens": 102928003.0,
      "step": 59340
    },
    {
      "entropy": 5.972020435333252,
      "epoch": 4.6172340011670885,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003049325234341287,
      "loss": 4.979,
      "mean_token_accuracy": 0.21572611033916472,
      "num_tokens": 102936624.0,
      "step": 59345
    },
    {
      "entropy": 5.936128759384156,
      "epoch": 4.617623030538805,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030490505396808874,
      "loss": 5.0233,
      "mean_token_accuracy": 0.21933944821357726,
      "num_tokens": 102945696.0,
      "step": 59350
    },
    {
      "entropy": 5.960077905654908,
      "epoch": 4.618012059910523,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003048775840482873,
      "loss": 4.9586,
      "mean_token_accuracy": 0.21459437012672425,
      "num_tokens": 102953472.0,
      "step": 59355
    },
    {
      "entropy": 5.999762105941772,
      "epoch": 4.618401089282241,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000304850113675141,
      "loss": 5.0998,
      "mean_token_accuracy": 0.21144948154687881,
      "num_tokens": 102962445.0,
      "step": 59360
    },
    {
      "entropy": 5.943924903869629,
      "epoch": 4.618790118653958,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003048226428490666,
      "loss": 4.9255,
      "mean_token_accuracy": 0.21812967509031295,
      "num_tokens": 102971223.0,
      "step": 59365
    },
    {
      "entropy": 5.895195245742798,
      "epoch": 4.619179148025676,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030479517157048125,
      "loss": 4.8599,
      "mean_token_accuracy": 0.22359820157289506,
      "num_tokens": 102979794.0,
      "step": 59370
    },
    {
      "entropy": 5.9199432849884035,
      "epoch": 4.619568177397394,
      "grad_norm": 1.5,
      "learning_rate": 0.0003047676998398015,
      "loss": 4.9437,
      "mean_token_accuracy": 0.21787693798542024,
      "num_tokens": 102989013.0,
      "step": 59375
    },
    {
      "entropy": 5.978321552276611,
      "epoch": 4.6199572067691115,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00030474022765744434,
      "loss": 5.0038,
      "mean_token_accuracy": 0.21596390008926392,
      "num_tokens": 102998050.0,
      "step": 59380
    },
    {
      "entropy": 5.937510919570923,
      "epoch": 4.620346236140828,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030471275502382656,
      "loss": 4.8803,
      "mean_token_accuracy": 0.2216783732175827,
      "num_tokens": 103005999.0,
      "step": 59385
    },
    {
      "entropy": 5.933714771270752,
      "epoch": 4.620735265512546,
      "grad_norm": 1.390625,
      "learning_rate": 0.000304685281939365,
      "loss": 4.8874,
      "mean_token_accuracy": 0.21822693049907685,
      "num_tokens": 103014217.0,
      "step": 59390
    },
    {
      "entropy": 5.9223555564880375,
      "epoch": 4.621124294884264,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003046578084044765,
      "loss": 5.0237,
      "mean_token_accuracy": 0.21587464958429337,
      "num_tokens": 103023649.0,
      "step": 59395
    },
    {
      "entropy": 5.914524030685425,
      "epoch": 4.621513324255981,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000304630334419578,
      "loss": 4.9528,
      "mean_token_accuracy": 0.21503783166408538,
      "num_tokens": 103032743.0,
      "step": 59400
    },
    {
      "entropy": 5.942310762405396,
      "epoch": 4.621902353627699,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003046028599850863,
      "loss": 4.974,
      "mean_token_accuracy": 0.21579717248678207,
      "num_tokens": 103041185.0,
      "step": 59405
    },
    {
      "entropy": 5.938474941253662,
      "epoch": 4.622291382999417,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003045753851014184,
      "loss": 4.984,
      "mean_token_accuracy": 0.21446997970342635,
      "num_tokens": 103049893.0,
      "step": 59410
    },
    {
      "entropy": 5.961822271347046,
      "epoch": 4.622680412371134,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030454790976899105,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21533093303442002,
      "num_tokens": 103058598.0,
      "step": 59415
    },
    {
      "entropy": 5.899148893356323,
      "epoch": 4.623069441742851,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003045204339882212,
      "loss": 4.843,
      "mean_token_accuracy": 0.22821187525987624,
      "num_tokens": 103067089.0,
      "step": 59420
    },
    {
      "entropy": 5.9310156345367435,
      "epoch": 4.623458471114569,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003044929577595257,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21629907935857773,
      "num_tokens": 103075304.0,
      "step": 59425
    },
    {
      "entropy": 5.971226167678833,
      "epoch": 4.623847500486287,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030446548108332164,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20771866738796235,
      "num_tokens": 103083303.0,
      "step": 59430
    },
    {
      "entropy": 5.906799983978272,
      "epoch": 4.624236529858004,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030443800396002575,
      "loss": 4.8451,
      "mean_token_accuracy": 0.2362408608198166,
      "num_tokens": 103092285.0,
      "step": 59435
    },
    {
      "entropy": 5.985685968399048,
      "epoch": 4.624625559229722,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003044105263900549,
      "loss": 5.004,
      "mean_token_accuracy": 0.21162452697753906,
      "num_tokens": 103101807.0,
      "step": 59440
    },
    {
      "entropy": 5.922775745391846,
      "epoch": 4.62501458860144,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003043830483738262,
      "loss": 4.9529,
      "mean_token_accuracy": 0.21218038052320481,
      "num_tokens": 103110317.0,
      "step": 59445
    },
    {
      "entropy": 5.83160924911499,
      "epoch": 4.6254036179731575,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003043555699117565,
      "loss": 4.8175,
      "mean_token_accuracy": 0.22202649861574172,
      "num_tokens": 103118861.0,
      "step": 59450
    },
    {
      "entropy": 5.901962471008301,
      "epoch": 4.625792647344874,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030432809100426286,
      "loss": 4.9358,
      "mean_token_accuracy": 0.21301253288984298,
      "num_tokens": 103126817.0,
      "step": 59455
    },
    {
      "entropy": 5.96149845123291,
      "epoch": 4.626181676716592,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003043006116517619,
      "loss": 5.0349,
      "mean_token_accuracy": 0.2121929258108139,
      "num_tokens": 103135619.0,
      "step": 59460
    },
    {
      "entropy": 5.869906568527222,
      "epoch": 4.62657070608831,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030427313185467104,
      "loss": 4.8541,
      "mean_token_accuracy": 0.22311229556798934,
      "num_tokens": 103143877.0,
      "step": 59465
    },
    {
      "entropy": 5.902308416366577,
      "epoch": 4.626959735460027,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003042456516134068,
      "loss": 4.829,
      "mean_token_accuracy": 0.22154193818569184,
      "num_tokens": 103152933.0,
      "step": 59470
    },
    {
      "entropy": 5.90068211555481,
      "epoch": 4.627348764831745,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00030421817092838644,
      "loss": 4.9103,
      "mean_token_accuracy": 0.21987145096063615,
      "num_tokens": 103161755.0,
      "step": 59475
    },
    {
      "entropy": 5.968494844436646,
      "epoch": 4.627737794203462,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003041906898000268,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2122408464550972,
      "num_tokens": 103170613.0,
      "step": 59480
    },
    {
      "entropy": 5.940574359893799,
      "epoch": 4.62812682357518,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003041632082287449,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2222209542989731,
      "num_tokens": 103178571.0,
      "step": 59485
    },
    {
      "entropy": 5.875456762313843,
      "epoch": 4.628515852946897,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003041357262149577,
      "loss": 4.9072,
      "mean_token_accuracy": 0.22155694514513016,
      "num_tokens": 103186879.0,
      "step": 59490
    },
    {
      "entropy": 5.945737504959107,
      "epoch": 4.628904882318615,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003041082437590824,
      "loss": 4.9547,
      "mean_token_accuracy": 0.21766619831323625,
      "num_tokens": 103195090.0,
      "step": 59495
    },
    {
      "entropy": 5.932640647888183,
      "epoch": 4.629293911690333,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030408076086153575,
      "loss": 4.93,
      "mean_token_accuracy": 0.21464783996343612,
      "num_tokens": 103204349.0,
      "step": 59500
    },
    {
      "entropy": 5.969381618499756,
      "epoch": 4.62968294106205,
      "grad_norm": 1.5,
      "learning_rate": 0.0003040532775227347,
      "loss": 5.0025,
      "mean_token_accuracy": 0.2150615081191063,
      "num_tokens": 103212561.0,
      "step": 59505
    },
    {
      "entropy": 5.927111339569092,
      "epoch": 4.630071970433768,
      "grad_norm": 1.625,
      "learning_rate": 0.00030402579374309656,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2138621225953102,
      "num_tokens": 103220876.0,
      "step": 59510
    },
    {
      "entropy": 5.9726580619812015,
      "epoch": 4.630460999805486,
      "grad_norm": 1.53125,
      "learning_rate": 0.00030399830952303815,
      "loss": 4.9958,
      "mean_token_accuracy": 0.2135028839111328,
      "num_tokens": 103229614.0,
      "step": 59515
    },
    {
      "entropy": 5.955676889419555,
      "epoch": 4.630850029177203,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030397082486297657,
      "loss": 4.9329,
      "mean_token_accuracy": 0.2159489393234253,
      "num_tokens": 103238092.0,
      "step": 59520
    },
    {
      "entropy": 5.966749954223633,
      "epoch": 4.63123905854892,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003039433397633288,
      "loss": 5.0117,
      "mean_token_accuracy": 0.21646805703639985,
      "num_tokens": 103247146.0,
      "step": 59525
    },
    {
      "entropy": 5.902210998535156,
      "epoch": 4.631628087920638,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030391585422451196,
      "loss": 4.8324,
      "mean_token_accuracy": 0.22386206835508346,
      "num_tokens": 103255718.0,
      "step": 59530
    },
    {
      "entropy": 5.959663677215576,
      "epoch": 4.632017117292356,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00030388836824694307,
      "loss": 5.014,
      "mean_token_accuracy": 0.212480066716671,
      "num_tokens": 103264120.0,
      "step": 59535
    },
    {
      "entropy": 5.92330322265625,
      "epoch": 4.632406146664073,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003038608818310391,
      "loss": 4.9555,
      "mean_token_accuracy": 0.20944853574037553,
      "num_tokens": 103273172.0,
      "step": 59540
    },
    {
      "entropy": 5.961110925674438,
      "epoch": 4.632795176035791,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030383339497721726,
      "loss": 5.0417,
      "mean_token_accuracy": 0.20883206725120546,
      "num_tokens": 103282345.0,
      "step": 59545
    },
    {
      "entropy": 5.993428564071655,
      "epoch": 4.633184205407508,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003038059076858944,
      "loss": 5.038,
      "mean_token_accuracy": 0.20808208882808685,
      "num_tokens": 103290845.0,
      "step": 59550
    },
    {
      "entropy": 5.955304670333862,
      "epoch": 4.633573234779226,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003037784199574879,
      "loss": 4.9382,
      "mean_token_accuracy": 0.2144925132393837,
      "num_tokens": 103299729.0,
      "step": 59555
    },
    {
      "entropy": 5.967109441757202,
      "epoch": 4.633962264150943,
      "grad_norm": 1.515625,
      "learning_rate": 0.00030375093179241464,
      "loss": 5.0541,
      "mean_token_accuracy": 0.21334277391433715,
      "num_tokens": 103308293.0,
      "step": 59560
    },
    {
      "entropy": 6.01031904220581,
      "epoch": 4.634351293522661,
      "grad_norm": 1.515625,
      "learning_rate": 0.00030372344319109173,
      "loss": 5.0527,
      "mean_token_accuracy": 0.21260146647691727,
      "num_tokens": 103317723.0,
      "step": 59565
    },
    {
      "entropy": 6.024979209899902,
      "epoch": 4.634740322894379,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003036959541539363,
      "loss": 5.0091,
      "mean_token_accuracy": 0.2067669451236725,
      "num_tokens": 103326252.0,
      "step": 59570
    },
    {
      "entropy": 5.924671602249146,
      "epoch": 4.635129352266096,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030366846468136543,
      "loss": 4.8452,
      "mean_token_accuracy": 0.22905461937189103,
      "num_tokens": 103334972.0,
      "step": 59575
    },
    {
      "entropy": 5.988159227371216,
      "epoch": 4.635518381637814,
      "grad_norm": 1.6875,
      "learning_rate": 0.0003036409747737963,
      "loss": 4.957,
      "mean_token_accuracy": 0.21538517922163009,
      "num_tokens": 103344116.0,
      "step": 59580
    },
    {
      "entropy": 5.963292407989502,
      "epoch": 4.635907411009531,
      "grad_norm": 1.546875,
      "learning_rate": 0.00030361348443164583,
      "loss": 5.0313,
      "mean_token_accuracy": 0.2101670891046524,
      "num_tokens": 103352500.0,
      "step": 59585
    },
    {
      "entropy": 5.940342378616333,
      "epoch": 4.6362964403812486,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003035859936553314,
      "loss": 4.9844,
      "mean_token_accuracy": 0.2185792863368988,
      "num_tokens": 103360233.0,
      "step": 59590
    },
    {
      "entropy": 5.916723823547363,
      "epoch": 4.636685469752966,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030355850244527,
      "loss": 4.9204,
      "mean_token_accuracy": 0.22125461101531982,
      "num_tokens": 103369574.0,
      "step": 59595
    },
    {
      "entropy": 5.895914077758789,
      "epoch": 4.637074499124684,
      "grad_norm": 1.375,
      "learning_rate": 0.0003035310108018787,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2222664773464203,
      "num_tokens": 103378455.0,
      "step": 59600
    },
    {
      "entropy": 5.964471340179443,
      "epoch": 4.637463528496402,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003035035187255748,
      "loss": 5.0176,
      "mean_token_accuracy": 0.21246451437473296,
      "num_tokens": 103386790.0,
      "step": 59605
    },
    {
      "entropy": 5.938351106643677,
      "epoch": 4.637852557868119,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00030347602621677534,
      "loss": 4.8925,
      "mean_token_accuracy": 0.22738542705774306,
      "num_tokens": 103395566.0,
      "step": 59610
    },
    {
      "entropy": 5.990320873260498,
      "epoch": 4.638241587239836,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003034485332758975,
      "loss": 5.0105,
      "mean_token_accuracy": 0.20197684317827225,
      "num_tokens": 103404642.0,
      "step": 59615
    },
    {
      "entropy": 6.072265481948852,
      "epoch": 4.638630616611554,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030342103990335846,
      "loss": 5.141,
      "mean_token_accuracy": 0.2016729772090912,
      "num_tokens": 103414149.0,
      "step": 59620
    },
    {
      "entropy": 5.971632289886474,
      "epoch": 4.6390196459832715,
      "grad_norm": 1.546875,
      "learning_rate": 0.00030339354609957545,
      "loss": 4.8677,
      "mean_token_accuracy": 0.22891562283039094,
      "num_tokens": 103422554.0,
      "step": 59625
    },
    {
      "entropy": 6.036442995071411,
      "epoch": 4.639408675354989,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00030336605186496546,
      "loss": 5.0065,
      "mean_token_accuracy": 0.211683189868927,
      "num_tokens": 103430795.0,
      "step": 59630
    },
    {
      "entropy": 5.9158721446990965,
      "epoch": 4.639797704726707,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030333855719994585,
      "loss": 4.9206,
      "mean_token_accuracy": 0.220899498462677,
      "num_tokens": 103439862.0,
      "step": 59635
    },
    {
      "entropy": 5.924229145050049,
      "epoch": 4.640186734098425,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003033110621049336,
      "loss": 4.915,
      "mean_token_accuracy": 0.21842632591724395,
      "num_tokens": 103448372.0,
      "step": 59640
    },
    {
      "entropy": 5.944809818267823,
      "epoch": 4.640575763470142,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030328356658034623,
      "loss": 4.9428,
      "mean_token_accuracy": 0.21024688631296157,
      "num_tokens": 103457004.0,
      "step": 59645
    },
    {
      "entropy": 5.942098569869995,
      "epoch": 4.64096479284186,
      "grad_norm": 1.578125,
      "learning_rate": 0.00030325607062660066,
      "loss": 4.9147,
      "mean_token_accuracy": 0.2236834317445755,
      "num_tokens": 103465655.0,
      "step": 59650
    },
    {
      "entropy": 5.920984411239624,
      "epoch": 4.641353822213577,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003032285742441143,
      "loss": 4.9164,
      "mean_token_accuracy": 0.21808114647865295,
      "num_tokens": 103474113.0,
      "step": 59655
    },
    {
      "entropy": 5.934466361999512,
      "epoch": 4.6417428515852945,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003032010774333041,
      "loss": 4.9902,
      "mean_token_accuracy": 0.20940814167261124,
      "num_tokens": 103483200.0,
      "step": 59660
    },
    {
      "entropy": 6.016442489624024,
      "epoch": 4.642131880957012,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030317358019458753,
      "loss": 5.0838,
      "mean_token_accuracy": 0.21545441895723344,
      "num_tokens": 103491899.0,
      "step": 59665
    },
    {
      "entropy": 5.960640144348145,
      "epoch": 4.64252091032873,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003031460825283817,
      "loss": 4.9539,
      "mean_token_accuracy": 0.22066253870725633,
      "num_tokens": 103501063.0,
      "step": 59670
    },
    {
      "entropy": 5.9230186462402346,
      "epoch": 4.642909939700448,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030311858443510383,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20904681235551834,
      "num_tokens": 103510202.0,
      "step": 59675
    },
    {
      "entropy": 5.914337062835694,
      "epoch": 4.643298969072165,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003030910859151712,
      "loss": 4.9042,
      "mean_token_accuracy": 0.21939670741558076,
      "num_tokens": 103518389.0,
      "step": 59680
    },
    {
      "entropy": 5.965895748138427,
      "epoch": 4.643687998443882,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00030306358696900114,
      "loss": 4.9931,
      "mean_token_accuracy": 0.21401809453964232,
      "num_tokens": 103527037.0,
      "step": 59685
    },
    {
      "entropy": 5.995781803131104,
      "epoch": 4.6440770278156,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00030303608759701077,
      "loss": 4.9338,
      "mean_token_accuracy": 0.22205109596252443,
      "num_tokens": 103535436.0,
      "step": 59690
    },
    {
      "entropy": 5.998572874069214,
      "epoch": 4.6444660571873175,
      "grad_norm": 1.5,
      "learning_rate": 0.0003030085877996174,
      "loss": 5.0117,
      "mean_token_accuracy": 0.21217791587114335,
      "num_tokens": 103543396.0,
      "step": 59695
    },
    {
      "entropy": 5.925953102111817,
      "epoch": 4.644855086559035,
      "grad_norm": 1.375,
      "learning_rate": 0.00030298108757723823,
      "loss": 4.9933,
      "mean_token_accuracy": 0.2115253046154976,
      "num_tokens": 103552341.0,
      "step": 59700
    },
    {
      "entropy": 5.941769027709961,
      "epoch": 4.645244115930753,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003029535869302906,
      "loss": 4.997,
      "mean_token_accuracy": 0.2107442647218704,
      "num_tokens": 103561209.0,
      "step": 59705
    },
    {
      "entropy": 6.015166187286377,
      "epoch": 4.645633145302471,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030292608585919185,
      "loss": 5.0047,
      "mean_token_accuracy": 0.21223630905151367,
      "num_tokens": 103568807.0,
      "step": 59710
    },
    {
      "entropy": 5.999033689498901,
      "epoch": 4.646022174674188,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030289858436435916,
      "loss": 5.0496,
      "mean_token_accuracy": 0.2177787736058235,
      "num_tokens": 103577770.0,
      "step": 59715
    },
    {
      "entropy": 5.899955701828003,
      "epoch": 4.646411204045905,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030287108244620993,
      "loss": 4.8513,
      "mean_token_accuracy": 0.22263091057538986,
      "num_tokens": 103586765.0,
      "step": 59720
    },
    {
      "entropy": 6.011713647842408,
      "epoch": 4.646800233417623,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00030284358010516135,
      "loss": 4.9921,
      "mean_token_accuracy": 0.2121526375412941,
      "num_tokens": 103595395.0,
      "step": 59725
    },
    {
      "entropy": 5.995758485794068,
      "epoch": 4.6471892627893405,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003028160773416307,
      "loss": 4.9843,
      "mean_token_accuracy": 0.20997796356678008,
      "num_tokens": 103604154.0,
      "step": 59730
    },
    {
      "entropy": 5.977533340454102,
      "epoch": 4.647578292161058,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00030278857415603534,
      "loss": 5.0008,
      "mean_token_accuracy": 0.212240768969059,
      "num_tokens": 103612682.0,
      "step": 59735
    },
    {
      "entropy": 5.837417793273926,
      "epoch": 4.647967321532776,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030276107054879263,
      "loss": 4.8516,
      "mean_token_accuracy": 0.2196103513240814,
      "num_tokens": 103620741.0,
      "step": 59740
    },
    {
      "entropy": 5.959500980377197,
      "epoch": 4.648356350904494,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030273356652031993,
      "loss": 5.0821,
      "mean_token_accuracy": 0.2080782860517502,
      "num_tokens": 103629371.0,
      "step": 59745
    },
    {
      "entropy": 6.036035060882568,
      "epoch": 4.64874538027621,
      "grad_norm": 1.5,
      "learning_rate": 0.0003027060620710344,
      "loss": 5.0861,
      "mean_token_accuracy": 0.20213526040315627,
      "num_tokens": 103638345.0,
      "step": 59750
    },
    {
      "entropy": 5.999483585357666,
      "epoch": 4.649134409647928,
      "grad_norm": 1.5625,
      "learning_rate": 0.00030267855720135343,
      "loss": 4.969,
      "mean_token_accuracy": 0.21915798783302307,
      "num_tokens": 103647078.0,
      "step": 59755
    },
    {
      "entropy": 5.924279975891113,
      "epoch": 4.649523439019646,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003026510519116946,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22552172541618348,
      "num_tokens": 103655260.0,
      "step": 59760
    },
    {
      "entropy": 5.889341211318969,
      "epoch": 4.6499124683913635,
      "grad_norm": 1.375,
      "learning_rate": 0.000302623546202475,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22440139204263687,
      "num_tokens": 103664933.0,
      "step": 59765
    },
    {
      "entropy": 5.933447408676147,
      "epoch": 4.650301497763081,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003025960400741119,
      "loss": 4.9037,
      "mean_token_accuracy": 0.2199086368083954,
      "num_tokens": 103673478.0,
      "step": 59770
    },
    {
      "entropy": 5.940554571151734,
      "epoch": 4.650690527134799,
      "grad_norm": 1.46875,
      "learning_rate": 0.000302568533527023,
      "loss": 5.0033,
      "mean_token_accuracy": 0.21493564397096634,
      "num_tokens": 103682534.0,
      "step": 59775
    },
    {
      "entropy": 5.998607397079468,
      "epoch": 4.651079556506517,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003025410265616254,
      "loss": 5.0499,
      "mean_token_accuracy": 0.2098961815237999,
      "num_tokens": 103691626.0,
      "step": 59780
    },
    {
      "entropy": 5.870136165618897,
      "epoch": 4.651468585878233,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003025135191783366,
      "loss": 4.9521,
      "mean_token_accuracy": 0.21231155544519426,
      "num_tokens": 103700707.0,
      "step": 59785
    },
    {
      "entropy": 6.059704685211182,
      "epoch": 4.651857615249951,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030248601137757394,
      "loss": 5.1006,
      "mean_token_accuracy": 0.20184117555618286,
      "num_tokens": 103709521.0,
      "step": 59790
    },
    {
      "entropy": 6.011986112594604,
      "epoch": 4.652246644621669,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003024585031597547,
      "loss": 5.0011,
      "mean_token_accuracy": 0.21430256515741347,
      "num_tokens": 103718046.0,
      "step": 59795
    },
    {
      "entropy": 5.997113132476807,
      "epoch": 4.6526356739933865,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003024309945252965,
      "loss": 5.0116,
      "mean_token_accuracy": 0.20558273047208786,
      "num_tokens": 103727107.0,
      "step": 59800
    },
    {
      "entropy": 5.895060729980469,
      "epoch": 4.653024703365104,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030240348547461653,
      "loss": 4.8862,
      "mean_token_accuracy": 0.21892718672752381,
      "num_tokens": 103735133.0,
      "step": 59805
    },
    {
      "entropy": 5.945917463302612,
      "epoch": 4.653413732736822,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00030237597600813235,
      "loss": 5.0323,
      "mean_token_accuracy": 0.1992183655500412,
      "num_tokens": 103743945.0,
      "step": 59810
    },
    {
      "entropy": 5.979691886901856,
      "epoch": 4.653802762108539,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003023484661262613,
      "loss": 4.9375,
      "mean_token_accuracy": 0.2115895688533783,
      "num_tokens": 103752316.0,
      "step": 59815
    },
    {
      "entropy": 5.95389347076416,
      "epoch": 4.654191791480256,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003023209558294208,
      "loss": 4.9303,
      "mean_token_accuracy": 0.2208834707736969,
      "num_tokens": 103760660.0,
      "step": 59820
    },
    {
      "entropy": 5.908425807952881,
      "epoch": 4.654580820851974,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030229344511802824,
      "loss": 4.8569,
      "mean_token_accuracy": 0.2259666845202446,
      "num_tokens": 103769078.0,
      "step": 59825
    },
    {
      "entropy": 6.003791666030883,
      "epoch": 4.654969850223692,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030226593399250114,
      "loss": 5.1268,
      "mean_token_accuracy": 0.20201010257005692,
      "num_tokens": 103778539.0,
      "step": 59830
    },
    {
      "entropy": 5.897380685806274,
      "epoch": 4.6553588795954095,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030223842245325687,
      "loss": 4.9248,
      "mean_token_accuracy": 0.22767056077718734,
      "num_tokens": 103787614.0,
      "step": 59835
    },
    {
      "entropy": 6.002761697769165,
      "epoch": 4.655747908967127,
      "grad_norm": 1.484375,
      "learning_rate": 0.00030221091050071293,
      "loss": 5.0143,
      "mean_token_accuracy": 0.21097975522279738,
      "num_tokens": 103796577.0,
      "step": 59840
    },
    {
      "entropy": 5.972037506103516,
      "epoch": 4.656136938338845,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00030218339813528665,
      "loss": 4.9465,
      "mean_token_accuracy": 0.21965831369161606,
      "num_tokens": 103804827.0,
      "step": 59845
    },
    {
      "entropy": 5.941406536102295,
      "epoch": 4.656525967710563,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003021558853573957,
      "loss": 4.9314,
      "mean_token_accuracy": 0.22272339165210725,
      "num_tokens": 103813586.0,
      "step": 59850
    },
    {
      "entropy": 5.9654182434082035,
      "epoch": 4.656914997082279,
      "grad_norm": 1.375,
      "learning_rate": 0.0003021283721674573,
      "loss": 5.039,
      "mean_token_accuracy": 0.21040942519903183,
      "num_tokens": 103822461.0,
      "step": 59855
    },
    {
      "entropy": 5.998545694351196,
      "epoch": 4.657304026453997,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003021008585658891,
      "loss": 4.9587,
      "mean_token_accuracy": 0.2132973313331604,
      "num_tokens": 103831100.0,
      "step": 59860
    },
    {
      "entropy": 5.9744713306427,
      "epoch": 4.657693055825715,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030207334455310846,
      "loss": 4.9569,
      "mean_token_accuracy": 0.21521892100572587,
      "num_tokens": 103839470.0,
      "step": 59865
    },
    {
      "entropy": 5.9622889995574955,
      "epoch": 4.6580820851974325,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003020458301295329,
      "loss": 5.0166,
      "mean_token_accuracy": 0.21285756528377534,
      "num_tokens": 103847930.0,
      "step": 59870
    },
    {
      "entropy": 5.940825080871582,
      "epoch": 4.65847111456915,
      "grad_norm": 1.5625,
      "learning_rate": 0.00030201831529557994,
      "loss": 4.8892,
      "mean_token_accuracy": 0.21458386033773422,
      "num_tokens": 103856296.0,
      "step": 59875
    },
    {
      "entropy": 5.933492231369018,
      "epoch": 4.658860143940868,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003019908000516671,
      "loss": 4.9029,
      "mean_token_accuracy": 0.22115705013275147,
      "num_tokens": 103865634.0,
      "step": 59880
    },
    {
      "entropy": 5.991169500350952,
      "epoch": 4.659249173312585,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003019632843982117,
      "loss": 5.0601,
      "mean_token_accuracy": 0.206548011302948,
      "num_tokens": 103874231.0,
      "step": 59885
    },
    {
      "entropy": 5.931983947753906,
      "epoch": 4.659638202684302,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030193576833563153,
      "loss": 4.8781,
      "mean_token_accuracy": 0.21873256415128708,
      "num_tokens": 103882944.0,
      "step": 59890
    },
    {
      "entropy": 5.957025051116943,
      "epoch": 4.66002723205602,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003019082518643438,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21811700463294983,
      "num_tokens": 103891963.0,
      "step": 59895
    },
    {
      "entropy": 5.9723228931427,
      "epoch": 4.660416261427738,
      "grad_norm": 1.53125,
      "learning_rate": 0.00030188073498476633,
      "loss": 4.947,
      "mean_token_accuracy": 0.21437387764453888,
      "num_tokens": 103900329.0,
      "step": 59900
    },
    {
      "entropy": 5.974957895278931,
      "epoch": 4.6608052907994555,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003018532176973163,
      "loss": 4.99,
      "mean_token_accuracy": 0.2133247122168541,
      "num_tokens": 103909086.0,
      "step": 59905
    },
    {
      "entropy": 5.984570026397705,
      "epoch": 4.661194320171173,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00030182570000241154,
      "loss": 4.9913,
      "mean_token_accuracy": 0.2262101098895073,
      "num_tokens": 103917446.0,
      "step": 59910
    },
    {
      "entropy": 5.941393232345581,
      "epoch": 4.661583349542891,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003017981819004696,
      "loss": 4.946,
      "mean_token_accuracy": 0.22162995636463165,
      "num_tokens": 103925451.0,
      "step": 59915
    },
    {
      "entropy": 5.895519971847534,
      "epoch": 4.661972378914608,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003017706633919078,
      "loss": 4.9499,
      "mean_token_accuracy": 0.2190278798341751,
      "num_tokens": 103934368.0,
      "step": 59920
    },
    {
      "entropy": 5.925657081604004,
      "epoch": 4.662361408286325,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003017431444771437,
      "loss": 4.9361,
      "mean_token_accuracy": 0.215387824177742,
      "num_tokens": 103943368.0,
      "step": 59925
    },
    {
      "entropy": 5.985759258270264,
      "epoch": 4.662750437658043,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030171562515659504,
      "loss": 4.9352,
      "mean_token_accuracy": 0.21602018028497696,
      "num_tokens": 103952387.0,
      "step": 59930
    },
    {
      "entropy": 6.040463209152222,
      "epoch": 4.663139467029761,
      "grad_norm": 1.5,
      "learning_rate": 0.0003016881054306793,
      "loss": 5.0951,
      "mean_token_accuracy": 0.2017020970582962,
      "num_tokens": 103961067.0,
      "step": 59935
    },
    {
      "entropy": 5.977185821533203,
      "epoch": 4.6635284964014785,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003016605852998139,
      "loss": 4.9502,
      "mean_token_accuracy": 0.21953993886709214,
      "num_tokens": 103970275.0,
      "step": 59940
    },
    {
      "entropy": 5.921248722076416,
      "epoch": 4.663917525773196,
      "grad_norm": 1.5625,
      "learning_rate": 0.00030163306476441675,
      "loss": 4.9783,
      "mean_token_accuracy": 0.21262012273073197,
      "num_tokens": 103978379.0,
      "step": 59945
    },
    {
      "entropy": 5.872812271118164,
      "epoch": 4.664306555144913,
      "grad_norm": 1.375,
      "learning_rate": 0.0003016055438249051,
      "loss": 4.8895,
      "mean_token_accuracy": 0.21890972107648848,
      "num_tokens": 103986726.0,
      "step": 59950
    },
    {
      "entropy": 5.889617109298706,
      "epoch": 4.664695584516631,
      "grad_norm": 1.375,
      "learning_rate": 0.0003015780224816968,
      "loss": 4.886,
      "mean_token_accuracy": 0.21857771426439285,
      "num_tokens": 103995073.0,
      "step": 59955
    },
    {
      "entropy": 5.944302129745483,
      "epoch": 4.665084613888348,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003015505007352093,
      "loss": 4.9665,
      "mean_token_accuracy": 0.2151861757040024,
      "num_tokens": 104004285.0,
      "step": 59960
    },
    {
      "entropy": 5.9863207817077635,
      "epoch": 4.665473643260066,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030152297858586016,
      "loss": 4.9095,
      "mean_token_accuracy": 0.21446677744388581,
      "num_tokens": 104012500.0,
      "step": 59965
    },
    {
      "entropy": 6.010613822937012,
      "epoch": 4.665862672631784,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00030149545603406694,
      "loss": 5.0493,
      "mean_token_accuracy": 0.2088080585002899,
      "num_tokens": 104021246.0,
      "step": 59970
    },
    {
      "entropy": 5.975566101074219,
      "epoch": 4.6662517020035015,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003014679330802474,
      "loss": 4.9628,
      "mean_token_accuracy": 0.21519633382558823,
      "num_tokens": 104029506.0,
      "step": 59975
    },
    {
      "entropy": 6.022634792327881,
      "epoch": 4.666640731375219,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030144040972481916,
      "loss": 5.0764,
      "mean_token_accuracy": 0.21139664947986603,
      "num_tokens": 104038669.0,
      "step": 59980
    },
    {
      "entropy": 5.903010034561158,
      "epoch": 4.667029760746937,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00030141288596819977,
      "loss": 4.9276,
      "mean_token_accuracy": 0.21862666308879852,
      "num_tokens": 104046859.0,
      "step": 59985
    },
    {
      "entropy": 5.979983472824097,
      "epoch": 4.667418790118654,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030138536181080683,
      "loss": 5.0316,
      "mean_token_accuracy": 0.21310158520936967,
      "num_tokens": 104055270.0,
      "step": 59990
    },
    {
      "entropy": 5.96666955947876,
      "epoch": 4.667807819490371,
      "grad_norm": 1.375,
      "learning_rate": 0.00030135783725305815,
      "loss": 4.9444,
      "mean_token_accuracy": 0.22415553629398347,
      "num_tokens": 104063819.0,
      "step": 59995
    },
    {
      "entropy": 5.941407823562622,
      "epoch": 4.668196848862089,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003013303122953712,
      "loss": 4.8909,
      "mean_token_accuracy": 0.22906938940286636,
      "num_tokens": 104072392.0,
      "step": 60000
    },
    {
      "epoch": 4.668196848862089,
      "eval_entropy": 5.687876052507652,
      "eval_loss": 5.061351299285889,
      "eval_mean_token_accuracy": 0.21988547209137557,
      "eval_num_tokens": 104072392.0,
      "eval_runtime": 21.6349,
      "eval_samples_per_second": 1920.881,
      "eval_steps_per_second": 240.122,
      "step": 60000
    },
    {
      "entropy": 5.9749502658844,
      "epoch": 4.668585878233807,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003013027869381635,
      "loss": 5.0189,
      "mean_token_accuracy": 0.2165118411183357,
      "num_tokens": 104080918.0,
      "step": 60005
    },
    {
      "entropy": 5.950287914276123,
      "epoch": 4.6689749076055245,
      "grad_norm": 1.46875,
      "learning_rate": 0.000301275261181853,
      "loss": 5.0366,
      "mean_token_accuracy": 0.20016543567180634,
      "num_tokens": 104089221.0,
      "step": 60010
    },
    {
      "entropy": 6.0116273880004885,
      "epoch": 4.669363936977242,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003012477350268572,
      "loss": 4.983,
      "mean_token_accuracy": 0.22090367078781128,
      "num_tokens": 104097863.0,
      "step": 60015
    },
    {
      "entropy": 5.946863126754761,
      "epoch": 4.669752966348959,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00030122020847359385,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21443540453910828,
      "num_tokens": 104106095.0,
      "step": 60020
    },
    {
      "entropy": 5.941083860397339,
      "epoch": 4.670141995720677,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003011926815224805,
      "loss": 4.9165,
      "mean_token_accuracy": 0.22072327584028245,
      "num_tokens": 104114225.0,
      "step": 60025
    },
    {
      "entropy": 5.919875383377075,
      "epoch": 4.670531025092394,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030116515417393495,
      "loss": 4.9414,
      "mean_token_accuracy": 0.2197279542684555,
      "num_tokens": 104123231.0,
      "step": 60030
    },
    {
      "entropy": 5.972499895095825,
      "epoch": 4.670920054464112,
      "grad_norm": 1.484375,
      "learning_rate": 0.00030113762642837487,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2179143935441971,
      "num_tokens": 104131415.0,
      "step": 60035
    },
    {
      "entropy": 5.989145088195801,
      "epoch": 4.67130908383583,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030111009828621786,
      "loss": 4.999,
      "mean_token_accuracy": 0.21043936163187027,
      "num_tokens": 104139553.0,
      "step": 60040
    },
    {
      "entropy": 5.941040992736816,
      "epoch": 4.6716981132075475,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003010825697478817,
      "loss": 4.9708,
      "mean_token_accuracy": 0.21498003453016282,
      "num_tokens": 104148746.0,
      "step": 60045
    },
    {
      "entropy": 5.9946372509002686,
      "epoch": 4.672087142579265,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000301055040813784,
      "loss": 5.0704,
      "mean_token_accuracy": 0.20169862359762192,
      "num_tokens": 104157581.0,
      "step": 60050
    },
    {
      "entropy": 5.974107646942139,
      "epoch": 4.672476171950982,
      "grad_norm": 1.484375,
      "learning_rate": 0.00030102751148434255,
      "loss": 4.9497,
      "mean_token_accuracy": 0.21227034628391267,
      "num_tokens": 104166367.0,
      "step": 60055
    },
    {
      "entropy": 5.885491514205933,
      "epoch": 4.6728652013227,
      "grad_norm": 1.484375,
      "learning_rate": 0.000300999981759975,
      "loss": 4.9004,
      "mean_token_accuracy": 0.2222553849220276,
      "num_tokens": 104174412.0,
      "step": 60060
    },
    {
      "entropy": 5.9316713333129885,
      "epoch": 4.673254230694417,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003009724516410992,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2245864138007164,
      "num_tokens": 104182344.0,
      "step": 60065
    },
    {
      "entropy": 5.90769476890564,
      "epoch": 4.673643260066135,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003009449211281328,
      "loss": 4.9549,
      "mean_token_accuracy": 0.21476221084594727,
      "num_tokens": 104190612.0,
      "step": 60070
    },
    {
      "entropy": 5.963969612121582,
      "epoch": 4.674032289437853,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003009173902214934,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21231108903884888,
      "num_tokens": 104198936.0,
      "step": 60075
    },
    {
      "entropy": 5.960894060134888,
      "epoch": 4.6744213188095705,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0003008898589215989,
      "loss": 5.016,
      "mean_token_accuracy": 0.21611536145210267,
      "num_tokens": 104207707.0,
      "step": 60080
    },
    {
      "entropy": 5.9502147197723385,
      "epoch": 4.674810348181287,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030086232722886704,
      "loss": 4.9334,
      "mean_token_accuracy": 0.22773289531469346,
      "num_tokens": 104216025.0,
      "step": 60085
    },
    {
      "entropy": 5.892511606216431,
      "epoch": 4.675199377553005,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030083479514371545,
      "loss": 4.8314,
      "mean_token_accuracy": 0.22890988737344742,
      "num_tokens": 104225283.0,
      "step": 60090
    },
    {
      "entropy": 5.931044578552246,
      "epoch": 4.675588406924723,
      "grad_norm": 1.53125,
      "learning_rate": 0.00030080726266656205,
      "loss": 4.9608,
      "mean_token_accuracy": 0.2216862052679062,
      "num_tokens": 104233252.0,
      "step": 60095
    },
    {
      "entropy": 5.899866008758545,
      "epoch": 4.67597743629644,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003007797297978245,
      "loss": 4.8726,
      "mean_token_accuracy": 0.2144427165389061,
      "num_tokens": 104241234.0,
      "step": 60100
    },
    {
      "entropy": 5.909007024765015,
      "epoch": 4.676366465668158,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003007521965379206,
      "loss": 4.9109,
      "mean_token_accuracy": 0.21852957606315612,
      "num_tokens": 104250091.0,
      "step": 60105
    },
    {
      "entropy": 5.904195356369018,
      "epoch": 4.676755495039876,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000300724662887268,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20896863043308259,
      "num_tokens": 104258668.0,
      "step": 60110
    },
    {
      "entropy": 5.939421796798706,
      "epoch": 4.6771445244115935,
      "grad_norm": 1.5,
      "learning_rate": 0.00030069712884628477,
      "loss": 4.8757,
      "mean_token_accuracy": 0.22455400675535203,
      "num_tokens": 104266799.0,
      "step": 60115
    },
    {
      "entropy": 6.000022459030151,
      "epoch": 4.67753355378331,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003006695944153883,
      "loss": 5.0411,
      "mean_token_accuracy": 0.20453110188245774,
      "num_tokens": 104275005.0,
      "step": 60120
    },
    {
      "entropy": 5.919785118103027,
      "epoch": 4.677922583155028,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003006420595949967,
      "loss": 4.9422,
      "mean_token_accuracy": 0.21564596891403198,
      "num_tokens": 104283997.0,
      "step": 60125
    },
    {
      "entropy": 6.0065771579742435,
      "epoch": 4.678311612526746,
      "grad_norm": 1.59375,
      "learning_rate": 0.00030061452438552767,
      "loss": 5.0419,
      "mean_token_accuracy": 0.2028913140296936,
      "num_tokens": 104291907.0,
      "step": 60130
    },
    {
      "entropy": 5.875115394592285,
      "epoch": 4.678700641898463,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000300586988787399,
      "loss": 4.9264,
      "mean_token_accuracy": 0.22264128029346467,
      "num_tokens": 104300917.0,
      "step": 60135
    },
    {
      "entropy": 5.918645572662354,
      "epoch": 4.679089671270181,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030055945280102844,
      "loss": 4.999,
      "mean_token_accuracy": 0.21630427688360215,
      "num_tokens": 104309814.0,
      "step": 60140
    },
    {
      "entropy": 5.965992498397827,
      "epoch": 4.679478700641899,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030053191642683393,
      "loss": 5.0179,
      "mean_token_accuracy": 0.20286183953285217,
      "num_tokens": 104318918.0,
      "step": 60145
    },
    {
      "entropy": 5.955486726760864,
      "epoch": 4.679867730013616,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003005043796652332,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21741945445537567,
      "num_tokens": 104327885.0,
      "step": 60150
    },
    {
      "entropy": 5.99527153968811,
      "epoch": 4.680256759385333,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030047684251664417,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20868892967700958,
      "num_tokens": 104337884.0,
      "step": 60155
    },
    {
      "entropy": 5.973678731918335,
      "epoch": 4.680645788757051,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003004493049814845,
      "loss": 4.9613,
      "mean_token_accuracy": 0.21858785748481752,
      "num_tokens": 104346572.0,
      "step": 60160
    },
    {
      "entropy": 5.920450592041016,
      "epoch": 4.681034818128769,
      "grad_norm": 1.578125,
      "learning_rate": 0.00030042176706017215,
      "loss": 4.9382,
      "mean_token_accuracy": 0.21860906928777696,
      "num_tokens": 104355243.0,
      "step": 60165
    },
    {
      "entropy": 5.94726185798645,
      "epoch": 4.681423847500486,
      "grad_norm": 1.453125,
      "learning_rate": 0.00030039422875312495,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21914635747671127,
      "num_tokens": 104363517.0,
      "step": 60170
    },
    {
      "entropy": 5.953428077697754,
      "epoch": 4.681812876872204,
      "grad_norm": 1.578125,
      "learning_rate": 0.00030036669006076075,
      "loss": 5.0212,
      "mean_token_accuracy": 0.2099926143884659,
      "num_tokens": 104372110.0,
      "step": 60175
    },
    {
      "entropy": 5.8969542503356935,
      "epoch": 4.682201906243922,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030033915098349746,
      "loss": 4.9692,
      "mean_token_accuracy": 0.21893234252929689,
      "num_tokens": 104380641.0,
      "step": 60180
    },
    {
      "entropy": 5.924350452423096,
      "epoch": 4.682590935615639,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003003116115217529,
      "loss": 4.869,
      "mean_token_accuracy": 0.2183280348777771,
      "num_tokens": 104389196.0,
      "step": 60185
    },
    {
      "entropy": 5.924060392379761,
      "epoch": 4.682979964987356,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00030028407167594486,
      "loss": 4.9413,
      "mean_token_accuracy": 0.22119003385305405,
      "num_tokens": 104397367.0,
      "step": 60190
    },
    {
      "entropy": 5.973294115066528,
      "epoch": 4.683368994359074,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003002565314464913,
      "loss": 5.0153,
      "mean_token_accuracy": 0.22505468875169754,
      "num_tokens": 104405750.0,
      "step": 60195
    },
    {
      "entropy": 5.978758668899536,
      "epoch": 4.683758023730792,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030022899083381006,
      "loss": 4.9351,
      "mean_token_accuracy": 0.21707394421100618,
      "num_tokens": 104415320.0,
      "step": 60200
    },
    {
      "entropy": 5.96633014678955,
      "epoch": 4.684147053102509,
      "grad_norm": 1.4375,
      "learning_rate": 0.000300201449838319,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20804767310619354,
      "num_tokens": 104424181.0,
      "step": 60205
    },
    {
      "entropy": 5.992878246307373,
      "epoch": 4.684536082474227,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003001739084604361,
      "loss": 5.0349,
      "mean_token_accuracy": 0.21035778373479844,
      "num_tokens": 104432858.0,
      "step": 60210
    },
    {
      "entropy": 5.946849298477173,
      "epoch": 4.684925111845945,
      "grad_norm": 2.71875,
      "learning_rate": 0.00030014636670057913,
      "loss": 4.8817,
      "mean_token_accuracy": 0.22842093259096147,
      "num_tokens": 104441662.0,
      "step": 60215
    },
    {
      "entropy": 5.928257417678833,
      "epoch": 4.6853141412176615,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003001188245591662,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21583738625049592,
      "num_tokens": 104450512.0,
      "step": 60220
    },
    {
      "entropy": 5.984553098678589,
      "epoch": 4.685703170589379,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003000912820366151,
      "loss": 4.9683,
      "mean_token_accuracy": 0.2116843655705452,
      "num_tokens": 104459056.0,
      "step": 60225
    },
    {
      "entropy": 5.930968999862671,
      "epoch": 4.686092199961097,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00030006373913334366,
      "loss": 4.9993,
      "mean_token_accuracy": 0.214417727291584,
      "num_tokens": 104469258.0,
      "step": 60230
    },
    {
      "entropy": 5.96537504196167,
      "epoch": 4.686481229332815,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003000361958497699,
      "loss": 4.9841,
      "mean_token_accuracy": 0.2127759799361229,
      "num_tokens": 104477177.0,
      "step": 60235
    },
    {
      "entropy": 5.966023969650268,
      "epoch": 4.686870258704532,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003000086521863116,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21279906034469603,
      "num_tokens": 104486235.0,
      "step": 60240
    },
    {
      "entropy": 5.916788101196289,
      "epoch": 4.68725928807625,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000299981108143387,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21391281336545945,
      "num_tokens": 104494736.0,
      "step": 60245
    },
    {
      "entropy": 5.973854064941406,
      "epoch": 4.687648317447968,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029995356372141367,
      "loss": 5.0347,
      "mean_token_accuracy": 0.20278994292020797,
      "num_tokens": 104503293.0,
      "step": 60250
    },
    {
      "entropy": 5.878686571121216,
      "epoch": 4.6880373468196845,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002999260189208098,
      "loss": 4.9181,
      "mean_token_accuracy": 0.215238581597805,
      "num_tokens": 104511350.0,
      "step": 60255
    },
    {
      "entropy": 5.914842128753662,
      "epoch": 4.688426376191402,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00029989847374199333,
      "loss": 4.8767,
      "mean_token_accuracy": 0.22610887587070466,
      "num_tokens": 104519902.0,
      "step": 60260
    },
    {
      "entropy": 5.963406085968018,
      "epoch": 4.68881540556312,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029987092818538217,
      "loss": 5.0252,
      "mean_token_accuracy": 0.2113843873143196,
      "num_tokens": 104528874.0,
      "step": 60265
    },
    {
      "entropy": 5.942490530014038,
      "epoch": 4.689204434934838,
      "grad_norm": 1.578125,
      "learning_rate": 0.00029984338225139405,
      "loss": 4.9052,
      "mean_token_accuracy": 0.22710263431072236,
      "num_tokens": 104536778.0,
      "step": 60270
    },
    {
      "entropy": 5.992015933990478,
      "epoch": 4.689593464306555,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002998158359404473,
      "loss": 5.0387,
      "mean_token_accuracy": 0.21157924383878707,
      "num_tokens": 104545207.0,
      "step": 60275
    },
    {
      "entropy": 5.919056606292725,
      "epoch": 4.689982493678273,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002997882892529597,
      "loss": 4.9302,
      "mean_token_accuracy": 0.22299935519695283,
      "num_tokens": 104553497.0,
      "step": 60280
    },
    {
      "entropy": 5.94305100440979,
      "epoch": 4.69037152304999,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002997607421893493,
      "loss": 5.0091,
      "mean_token_accuracy": 0.21505619287490846,
      "num_tokens": 104562837.0,
      "step": 60285
    },
    {
      "entropy": 5.937341356277466,
      "epoch": 4.6907605524217075,
      "grad_norm": 1.546875,
      "learning_rate": 0.000299733194750034,
      "loss": 4.9071,
      "mean_token_accuracy": 0.21794671267271043,
      "num_tokens": 104570628.0,
      "step": 60290
    },
    {
      "entropy": 5.934597778320312,
      "epoch": 4.691149581793425,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002997056469354319,
      "loss": 5.0411,
      "mean_token_accuracy": 0.20867305099964142,
      "num_tokens": 104579910.0,
      "step": 60295
    },
    {
      "entropy": 5.950962829589844,
      "epoch": 4.691538611165143,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002996780987459609,
      "loss": 4.9164,
      "mean_token_accuracy": 0.21882133781909943,
      "num_tokens": 104587936.0,
      "step": 60300
    },
    {
      "entropy": 6.031591749191284,
      "epoch": 4.691927640536861,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002996505501820391,
      "loss": 5.0586,
      "mean_token_accuracy": 0.20456299036741257,
      "num_tokens": 104597072.0,
      "step": 60305
    },
    {
      "entropy": 5.947728967666626,
      "epoch": 4.692316669908578,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002996230012440843,
      "loss": 4.9487,
      "mean_token_accuracy": 0.22061653584241867,
      "num_tokens": 104605573.0,
      "step": 60310
    },
    {
      "entropy": 5.910471677780151,
      "epoch": 4.692705699280296,
      "grad_norm": 1.515625,
      "learning_rate": 0.00029959545193251473,
      "loss": 5.0098,
      "mean_token_accuracy": 0.2157890409231186,
      "num_tokens": 104613954.0,
      "step": 60315
    },
    {
      "entropy": 5.946838617324829,
      "epoch": 4.693094728652014,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00029956790224774834,
      "loss": 5.0065,
      "mean_token_accuracy": 0.21560630798339844,
      "num_tokens": 104622410.0,
      "step": 60320
    },
    {
      "entropy": 5.992386054992676,
      "epoch": 4.6934837580237305,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029954035219020305,
      "loss": 4.9539,
      "mean_token_accuracy": 0.2133355438709259,
      "num_tokens": 104631953.0,
      "step": 60325
    },
    {
      "entropy": 5.920720529556275,
      "epoch": 4.693872787395448,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029951280176029703,
      "loss": 4.9099,
      "mean_token_accuracy": 0.2224492311477661,
      "num_tokens": 104640714.0,
      "step": 60330
    },
    {
      "entropy": 5.974487781524658,
      "epoch": 4.694261816767166,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029948525095844825,
      "loss": 5.0322,
      "mean_token_accuracy": 0.2098902329802513,
      "num_tokens": 104649773.0,
      "step": 60335
    },
    {
      "entropy": 5.972251462936401,
      "epoch": 4.694650846138884,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002994576997850748,
      "loss": 5.0039,
      "mean_token_accuracy": 0.2177252322435379,
      "num_tokens": 104658581.0,
      "step": 60340
    },
    {
      "entropy": 5.9842249870300295,
      "epoch": 4.695039875510601,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002994301482405946,
      "loss": 5.0337,
      "mean_token_accuracy": 0.211082324385643,
      "num_tokens": 104666735.0,
      "step": 60345
    },
    {
      "entropy": 5.869221639633179,
      "epoch": 4.695428904882318,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002994025963254259,
      "loss": 4.9318,
      "mean_token_accuracy": 0.2180479198694229,
      "num_tokens": 104675301.0,
      "step": 60350
    },
    {
      "entropy": 6.003762531280517,
      "epoch": 4.695817934254036,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002993750440399866,
      "loss": 4.9924,
      "mean_token_accuracy": 0.21911318898200988,
      "num_tokens": 104684468.0,
      "step": 60355
    },
    {
      "entropy": 5.975041484832763,
      "epoch": 4.6962069636257535,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002993474913846948,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2138512298464775,
      "num_tokens": 104693090.0,
      "step": 60360
    },
    {
      "entropy": 5.92622332572937,
      "epoch": 4.696595992997471,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002993199383599685,
      "loss": 4.969,
      "mean_token_accuracy": 0.2125132903456688,
      "num_tokens": 104701597.0,
      "step": 60365
    },
    {
      "entropy": 5.95840048789978,
      "epoch": 4.696985022369189,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002992923849662259,
      "loss": 5.0486,
      "mean_token_accuracy": 0.20684927850961685,
      "num_tokens": 104711000.0,
      "step": 60370
    },
    {
      "entropy": 5.974198675155639,
      "epoch": 4.697374051740907,
      "grad_norm": 1.375,
      "learning_rate": 0.00029926483120388506,
      "loss": 5.0088,
      "mean_token_accuracy": 0.20799960047006608,
      "num_tokens": 104719581.0,
      "step": 60375
    },
    {
      "entropy": 5.961877679824829,
      "epoch": 4.697763081112624,
      "grad_norm": 1.59375,
      "learning_rate": 0.000299237277073364,
      "loss": 5.0067,
      "mean_token_accuracy": 0.20907308161258698,
      "num_tokens": 104727060.0,
      "step": 60380
    },
    {
      "entropy": 5.998099899291992,
      "epoch": 4.698152110484342,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002992097225750808,
      "loss": 4.9813,
      "mean_token_accuracy": 0.2156379461288452,
      "num_tokens": 104734974.0,
      "step": 60385
    },
    {
      "entropy": 5.9509923458099365,
      "epoch": 4.698541139856059,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00029918216770945363,
      "loss": 4.9666,
      "mean_token_accuracy": 0.2127162903547287,
      "num_tokens": 104744352.0,
      "step": 60390
    },
    {
      "entropy": 5.966688632965088,
      "epoch": 4.6989301692277765,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002991546124769005,
      "loss": 4.9536,
      "mean_token_accuracy": 0.2166674554347992,
      "num_tokens": 104752485.0,
      "step": 60395
    },
    {
      "entropy": 5.945555591583252,
      "epoch": 4.699319198599494,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002991270568778396,
      "loss": 4.972,
      "mean_token_accuracy": 0.21471191197633743,
      "num_tokens": 104760469.0,
      "step": 60400
    },
    {
      "entropy": 5.914256143569946,
      "epoch": 4.699708227971212,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002990995009126891,
      "loss": 5.0085,
      "mean_token_accuracy": 0.21454692333936692,
      "num_tokens": 104769381.0,
      "step": 60405
    },
    {
      "entropy": 5.985376310348511,
      "epoch": 4.70009725734293,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002990719445818669,
      "loss": 5.0279,
      "mean_token_accuracy": 0.20911774933338165,
      "num_tokens": 104778222.0,
      "step": 60410
    },
    {
      "entropy": 5.969920063018799,
      "epoch": 4.700486286714647,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002990443878857914,
      "loss": 4.9418,
      "mean_token_accuracy": 0.2146775722503662,
      "num_tokens": 104786784.0,
      "step": 60415
    },
    {
      "entropy": 5.966796922683716,
      "epoch": 4.700875316086364,
      "grad_norm": 1.484375,
      "learning_rate": 0.00029901683082488045,
      "loss": 4.9516,
      "mean_token_accuracy": 0.21216151267290115,
      "num_tokens": 104795260.0,
      "step": 60420
    },
    {
      "entropy": 5.9179071426391605,
      "epoch": 4.701264345458082,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002989892733995524,
      "loss": 4.9208,
      "mean_token_accuracy": 0.21633727550506593,
      "num_tokens": 104804330.0,
      "step": 60425
    },
    {
      "entropy": 5.898036956787109,
      "epoch": 4.7016533748297995,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002989617156102253,
      "loss": 4.9722,
      "mean_token_accuracy": 0.21213054656982422,
      "num_tokens": 104812605.0,
      "step": 60430
    },
    {
      "entropy": 5.929125070571899,
      "epoch": 4.702042404201517,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002989341574573172,
      "loss": 5.0265,
      "mean_token_accuracy": 0.21408019512891768,
      "num_tokens": 104821207.0,
      "step": 60435
    },
    {
      "entropy": 5.949772644042969,
      "epoch": 4.702431433573235,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029890659894124647,
      "loss": 5.01,
      "mean_token_accuracy": 0.2191919520497322,
      "num_tokens": 104829628.0,
      "step": 60440
    },
    {
      "entropy": 5.9774659156799315,
      "epoch": 4.702820462944953,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002988790400624312,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21224894225597382,
      "num_tokens": 104838003.0,
      "step": 60445
    },
    {
      "entropy": 5.907231378555298,
      "epoch": 4.70320949231667,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002988514808212894,
      "loss": 4.9053,
      "mean_token_accuracy": 0.21582765728235245,
      "num_tokens": 104846008.0,
      "step": 60450
    },
    {
      "entropy": 5.927509403228759,
      "epoch": 4.703598521688387,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002988239212182394,
      "loss": 4.9161,
      "mean_token_accuracy": 0.2345385178923607,
      "num_tokens": 104855337.0,
      "step": 60455
    },
    {
      "entropy": 5.873818635940552,
      "epoch": 4.703987551060105,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002987963612536993,
      "loss": 4.8197,
      "mean_token_accuracy": 0.2339138299226761,
      "num_tokens": 104863780.0,
      "step": 60460
    },
    {
      "entropy": 5.842816495895386,
      "epoch": 4.7043765804318225,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00029876880092808724,
      "loss": 4.8074,
      "mean_token_accuracy": 0.22662388235330583,
      "num_tokens": 104871956.0,
      "step": 60465
    },
    {
      "entropy": 5.925304985046386,
      "epoch": 4.70476560980354,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002987412402418215,
      "loss": 4.9706,
      "mean_token_accuracy": 0.21215767413377762,
      "num_tokens": 104880454.0,
      "step": 60470
    },
    {
      "entropy": 5.849478006362915,
      "epoch": 4.705154639175258,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029871367919532014,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22777743637561798,
      "num_tokens": 104888224.0,
      "step": 60475
    },
    {
      "entropy": 5.988302326202392,
      "epoch": 4.705543668546976,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002986861177890016,
      "loss": 5.0005,
      "mean_token_accuracy": 0.21135144978761672,
      "num_tokens": 104897496.0,
      "step": 60480
    },
    {
      "entropy": 5.996218967437744,
      "epoch": 4.705932697918692,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002986585560232838,
      "loss": 5.0191,
      "mean_token_accuracy": 0.2080491617321968,
      "num_tokens": 104905911.0,
      "step": 60485
    },
    {
      "entropy": 6.001829481124878,
      "epoch": 4.70632172729041,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029863099389858516,
      "loss": 4.9488,
      "mean_token_accuracy": 0.2178354650735855,
      "num_tokens": 104914174.0,
      "step": 60490
    },
    {
      "entropy": 5.897401332855225,
      "epoch": 4.706710756662128,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002986034314153238,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22708578109741212,
      "num_tokens": 104922712.0,
      "step": 60495
    },
    {
      "entropy": 5.927610778808594,
      "epoch": 4.7070997860338455,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029857586857391796,
      "loss": 4.9302,
      "mean_token_accuracy": 0.2189048007130623,
      "num_tokens": 104930821.0,
      "step": 60500
    },
    {
      "entropy": 5.866061496734619,
      "epoch": 4.707488815405563,
      "grad_norm": 1.484375,
      "learning_rate": 0.00029854830537478567,
      "loss": 5.0051,
      "mean_token_accuracy": 0.21676801890134811,
      "num_tokens": 104938843.0,
      "step": 60505
    },
    {
      "entropy": 5.812976455688476,
      "epoch": 4.707877844777281,
      "grad_norm": 1.5625,
      "learning_rate": 0.00029852074181834545,
      "loss": 4.8362,
      "mean_token_accuracy": 0.22138829231262208,
      "num_tokens": 104947162.0,
      "step": 60510
    },
    {
      "entropy": 5.958100461959839,
      "epoch": 4.708266874148999,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029849317790501536,
      "loss": 4.9843,
      "mean_token_accuracy": 0.21008038967847825,
      "num_tokens": 104956478.0,
      "step": 60515
    },
    {
      "entropy": 6.029663944244385,
      "epoch": 4.708655903520716,
      "grad_norm": 1.375,
      "learning_rate": 0.00029846561363521375,
      "loss": 5.0189,
      "mean_token_accuracy": 0.20409189015626908,
      "num_tokens": 104964968.0,
      "step": 60520
    },
    {
      "entropy": 5.918820142745972,
      "epoch": 4.709044932892433,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002984380490093588,
      "loss": 4.9322,
      "mean_token_accuracy": 0.2139691084623337,
      "num_tokens": 104973965.0,
      "step": 60525
    },
    {
      "entropy": 5.973032379150391,
      "epoch": 4.709433962264151,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002984104840278687,
      "loss": 5.0305,
      "mean_token_accuracy": 0.21424041241407393,
      "num_tokens": 104982928.0,
      "step": 60530
    },
    {
      "entropy": 5.989659357070923,
      "epoch": 4.7098229916358685,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002983829186911618,
      "loss": 4.9605,
      "mean_token_accuracy": 0.21364318281412126,
      "num_tokens": 104991189.0,
      "step": 60535
    },
    {
      "entropy": 5.947399234771728,
      "epoch": 4.710212021007586,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029835535299965635,
      "loss": 4.9515,
      "mean_token_accuracy": 0.21920880824327468,
      "num_tokens": 104999556.0,
      "step": 60540
    },
    {
      "entropy": 5.999871397018433,
      "epoch": 4.710601050379304,
      "grad_norm": 1.5,
      "learning_rate": 0.00029832778695377057,
      "loss": 5.0279,
      "mean_token_accuracy": 0.21180133074522017,
      "num_tokens": 105008726.0,
      "step": 60545
    },
    {
      "entropy": 5.914922904968262,
      "epoch": 4.710990079751022,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029830022055392277,
      "loss": 4.917,
      "mean_token_accuracy": 0.2216814339160919,
      "num_tokens": 105017591.0,
      "step": 60550
    },
    {
      "entropy": 5.979308176040649,
      "epoch": 4.711379109122738,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029827265380053116,
      "loss": 5.0411,
      "mean_token_accuracy": 0.20886907875537872,
      "num_tokens": 105026939.0,
      "step": 60555
    },
    {
      "entropy": 5.996488189697265,
      "epoch": 4.711768138494456,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002982450866940141,
      "loss": 5.0063,
      "mean_token_accuracy": 0.21026464849710463,
      "num_tokens": 105035889.0,
      "step": 60560
    },
    {
      "entropy": 5.998342800140381,
      "epoch": 4.712157167866174,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002982175192347899,
      "loss": 5.0545,
      "mean_token_accuracy": 0.2110433652997017,
      "num_tokens": 105044118.0,
      "step": 60565
    },
    {
      "entropy": 5.9598921775817875,
      "epoch": 4.7125461972378915,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002981899514232767,
      "loss": 4.9858,
      "mean_token_accuracy": 0.21285568326711654,
      "num_tokens": 105051923.0,
      "step": 60570
    },
    {
      "entropy": 6.027708673477173,
      "epoch": 4.712935226609609,
      "grad_norm": 1.40625,
      "learning_rate": 0.000298162383259893,
      "loss": 5.1374,
      "mean_token_accuracy": 0.20247529745101928,
      "num_tokens": 105060892.0,
      "step": 60575
    },
    {
      "entropy": 6.031220436096191,
      "epoch": 4.713324255981327,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029813481474505693,
      "loss": 5.0305,
      "mean_token_accuracy": 0.21414585709571837,
      "num_tokens": 105069147.0,
      "step": 60580
    },
    {
      "entropy": 5.974422407150269,
      "epoch": 4.713713285353045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029810724587918684,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22520148605108262,
      "num_tokens": 105077751.0,
      "step": 60585
    },
    {
      "entropy": 5.931685590744019,
      "epoch": 4.714102314724761,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002980796766627012,
      "loss": 4.8774,
      "mean_token_accuracy": 0.2243342563509941,
      "num_tokens": 105085980.0,
      "step": 60590
    },
    {
      "entropy": 5.891200351715088,
      "epoch": 4.714491344096479,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002980521070960181,
      "loss": 4.8812,
      "mean_token_accuracy": 0.22216591089963914,
      "num_tokens": 105094147.0,
      "step": 60595
    },
    {
      "entropy": 5.912204933166504,
      "epoch": 4.714880373468197,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029802453717955597,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22138848751783372,
      "num_tokens": 105102304.0,
      "step": 60600
    },
    {
      "entropy": 5.952026987075806,
      "epoch": 4.7152694028399145,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002979969669137332,
      "loss": 4.9568,
      "mean_token_accuracy": 0.21896468698978425,
      "num_tokens": 105110702.0,
      "step": 60605
    },
    {
      "entropy": 5.990160894393921,
      "epoch": 4.715658432211632,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000297969396298968,
      "loss": 4.9031,
      "mean_token_accuracy": 0.2238642618060112,
      "num_tokens": 105119408.0,
      "step": 60610
    },
    {
      "entropy": 5.94277024269104,
      "epoch": 4.71604746158335,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002979418253356788,
      "loss": 4.9559,
      "mean_token_accuracy": 0.2168534904718399,
      "num_tokens": 105127921.0,
      "step": 60615
    },
    {
      "entropy": 5.916078853607178,
      "epoch": 4.716436490955067,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002979142540242839,
      "loss": 4.9374,
      "mean_token_accuracy": 0.21593323051929475,
      "num_tokens": 105136660.0,
      "step": 60620
    },
    {
      "entropy": 5.937439012527466,
      "epoch": 4.716825520326784,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029788668236520164,
      "loss": 5.01,
      "mean_token_accuracy": 0.21429550051689147,
      "num_tokens": 105145648.0,
      "step": 60625
    },
    {
      "entropy": 5.914355134963989,
      "epoch": 4.717214549698502,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029785911035885046,
      "loss": 4.9314,
      "mean_token_accuracy": 0.22205056697130204,
      "num_tokens": 105154730.0,
      "step": 60630
    },
    {
      "entropy": 5.946883296966552,
      "epoch": 4.71760357907022,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029783153800564863,
      "loss": 4.9674,
      "mean_token_accuracy": 0.21616688668727874,
      "num_tokens": 105163771.0,
      "step": 60635
    },
    {
      "entropy": 5.935117244720459,
      "epoch": 4.7179926084419375,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029780396530601456,
      "loss": 4.9761,
      "mean_token_accuracy": 0.21681013703346252,
      "num_tokens": 105172699.0,
      "step": 60640
    },
    {
      "entropy": 5.975795841217041,
      "epoch": 4.718381637813655,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029777639226036663,
      "loss": 5.0161,
      "mean_token_accuracy": 0.21433059275150299,
      "num_tokens": 105180777.0,
      "step": 60645
    },
    {
      "entropy": 5.887193584442139,
      "epoch": 4.718770667185373,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002977488188691232,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21883702427148818,
      "num_tokens": 105189858.0,
      "step": 60650
    },
    {
      "entropy": 6.016885375976562,
      "epoch": 4.7191596965570906,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002977212451327026,
      "loss": 5.0579,
      "mean_token_accuracy": 0.210029736161232,
      "num_tokens": 105198502.0,
      "step": 60655
    },
    {
      "entropy": 5.973077487945557,
      "epoch": 4.719548725928807,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002976936710515233,
      "loss": 5.0542,
      "mean_token_accuracy": 0.21264886409044265,
      "num_tokens": 105207916.0,
      "step": 60660
    },
    {
      "entropy": 6.012224435806274,
      "epoch": 4.719937755300525,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00029766609662600367,
      "loss": 4.9797,
      "mean_token_accuracy": 0.22141997367143632,
      "num_tokens": 105215355.0,
      "step": 60665
    },
    {
      "entropy": 6.002992153167725,
      "epoch": 4.720326784672243,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029763852185656206,
      "loss": 5.0291,
      "mean_token_accuracy": 0.20452914834022523,
      "num_tokens": 105224805.0,
      "step": 60670
    },
    {
      "entropy": 6.020327234268189,
      "epoch": 4.7207158140439605,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029761094674361695,
      "loss": 4.9567,
      "mean_token_accuracy": 0.22562595456838608,
      "num_tokens": 105233219.0,
      "step": 60675
    },
    {
      "entropy": 5.889328289031982,
      "epoch": 4.721104843415678,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029758337128758675,
      "loss": 4.933,
      "mean_token_accuracy": 0.2189657673239708,
      "num_tokens": 105241895.0,
      "step": 60680
    },
    {
      "entropy": 5.8939097881317135,
      "epoch": 4.721493872787395,
      "grad_norm": 1.5,
      "learning_rate": 0.0002975557954888898,
      "loss": 4.8773,
      "mean_token_accuracy": 0.22120548784732819,
      "num_tokens": 105250058.0,
      "step": 60685
    },
    {
      "entropy": 5.994195985794067,
      "epoch": 4.721882902159113,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029752821934794457,
      "loss": 4.9783,
      "mean_token_accuracy": 0.21168881356716157,
      "num_tokens": 105258946.0,
      "step": 60690
    },
    {
      "entropy": 5.9928162574768065,
      "epoch": 4.72227193153083,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029750064286516943,
      "loss": 5.0683,
      "mean_token_accuracy": 0.21164128482341765,
      "num_tokens": 105267598.0,
      "step": 60695
    },
    {
      "entropy": 6.010844087600708,
      "epoch": 4.722660960902548,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002974730660409828,
      "loss": 5.0126,
      "mean_token_accuracy": 0.21195673644542695,
      "num_tokens": 105275836.0,
      "step": 60700
    },
    {
      "entropy": 5.987272882461548,
      "epoch": 4.723049990274266,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029744548887580325,
      "loss": 4.9631,
      "mean_token_accuracy": 0.2111201211810112,
      "num_tokens": 105284378.0,
      "step": 60705
    },
    {
      "entropy": 5.876537704467774,
      "epoch": 4.7234390196459834,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029741791137004916,
      "loss": 4.8727,
      "mean_token_accuracy": 0.21991569101810454,
      "num_tokens": 105292575.0,
      "step": 60710
    },
    {
      "entropy": 5.943286991119384,
      "epoch": 4.723828049017701,
      "grad_norm": 1.375,
      "learning_rate": 0.00029739033352413883,
      "loss": 4.9955,
      "mean_token_accuracy": 0.21339594572782516,
      "num_tokens": 105301379.0,
      "step": 60715
    },
    {
      "entropy": 5.942796134948731,
      "epoch": 4.724217078389419,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00029736275533849094,
      "loss": 4.9796,
      "mean_token_accuracy": 0.21703360825777054,
      "num_tokens": 105309043.0,
      "step": 60720
    },
    {
      "entropy": 5.972658920288086,
      "epoch": 4.724606107761136,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002973351768135239,
      "loss": 5.0126,
      "mean_token_accuracy": 0.2171315997838974,
      "num_tokens": 105317137.0,
      "step": 60725
    },
    {
      "entropy": 5.939786958694458,
      "epoch": 4.724995137132853,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029730759794965595,
      "loss": 4.9143,
      "mean_token_accuracy": 0.2229238286614418,
      "num_tokens": 105325916.0,
      "step": 60730
    },
    {
      "entropy": 6.003970813751221,
      "epoch": 4.725384166504571,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002972800187473057,
      "loss": 5.0917,
      "mean_token_accuracy": 0.2137078672647476,
      "num_tokens": 105334694.0,
      "step": 60735
    },
    {
      "entropy": 5.905937147140503,
      "epoch": 4.725773195876289,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029725243920689163,
      "loss": 4.857,
      "mean_token_accuracy": 0.21843256950378417,
      "num_tokens": 105343604.0,
      "step": 60740
    },
    {
      "entropy": 5.9658522605896,
      "epoch": 4.726162225248006,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029722485932883233,
      "loss": 5.0316,
      "mean_token_accuracy": 0.20829534530639648,
      "num_tokens": 105351677.0,
      "step": 60745
    },
    {
      "entropy": 5.952255773544311,
      "epoch": 4.726551254619724,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029719727911354604,
      "loss": 4.9439,
      "mean_token_accuracy": 0.2193733811378479,
      "num_tokens": 105360348.0,
      "step": 60750
    },
    {
      "entropy": 5.979145908355713,
      "epoch": 4.726940283991441,
      "grad_norm": 1.609375,
      "learning_rate": 0.00029716969856145143,
      "loss": 5.0877,
      "mean_token_accuracy": 0.2053809344768524,
      "num_tokens": 105368816.0,
      "step": 60755
    },
    {
      "entropy": 5.935454511642456,
      "epoch": 4.727329313363159,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00029714211767296693,
      "loss": 4.9139,
      "mean_token_accuracy": 0.22055646479129792,
      "num_tokens": 105377524.0,
      "step": 60760
    },
    {
      "entropy": 6.0137535572052006,
      "epoch": 4.727718342734876,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002971145364485111,
      "loss": 4.9968,
      "mean_token_accuracy": 0.21270500868558884,
      "num_tokens": 105386349.0,
      "step": 60765
    },
    {
      "entropy": 5.996789264678955,
      "epoch": 4.728107372106594,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002970869548885023,
      "loss": 4.9733,
      "mean_token_accuracy": 0.2200220748782158,
      "num_tokens": 105394995.0,
      "step": 60770
    },
    {
      "entropy": 5.925554895401001,
      "epoch": 4.728496401478312,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002970593729933591,
      "loss": 4.9539,
      "mean_token_accuracy": 0.21670310348272323,
      "num_tokens": 105403820.0,
      "step": 60775
    },
    {
      "entropy": 6.018563604354858,
      "epoch": 4.728885430850029,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029703179076350005,
      "loss": 5.0773,
      "mean_token_accuracy": 0.21080236583948136,
      "num_tokens": 105413108.0,
      "step": 60780
    },
    {
      "entropy": 5.948452949523926,
      "epoch": 4.729274460221747,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029700420819934363,
      "loss": 4.9261,
      "mean_token_accuracy": 0.21941008269786835,
      "num_tokens": 105420405.0,
      "step": 60785
    },
    {
      "entropy": 5.890806341171265,
      "epoch": 4.729663489593464,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029697662530130836,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21683682054281234,
      "num_tokens": 105428737.0,
      "step": 60790
    },
    {
      "entropy": 5.87924280166626,
      "epoch": 4.730052518965182,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002969490420698129,
      "loss": 4.947,
      "mean_token_accuracy": 0.22036742866039277,
      "num_tokens": 105437680.0,
      "step": 60795
    },
    {
      "entropy": 5.967392492294311,
      "epoch": 4.730441548336899,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029692145850527553,
      "loss": 4.9147,
      "mean_token_accuracy": 0.21189166903495787,
      "num_tokens": 105446329.0,
      "step": 60800
    },
    {
      "entropy": 6.0005714893341064,
      "epoch": 4.730830577708617,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029689387460811494,
      "loss": 5.0242,
      "mean_token_accuracy": 0.20486094951629638,
      "num_tokens": 105455278.0,
      "step": 60805
    },
    {
      "entropy": 5.9375592231750485,
      "epoch": 4.731219607080335,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002968662903787497,
      "loss": 4.9872,
      "mean_token_accuracy": 0.21125925928354264,
      "num_tokens": 105464384.0,
      "step": 60810
    },
    {
      "entropy": 5.871532535552978,
      "epoch": 4.731608636452052,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002968387058175983,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22466182410717012,
      "num_tokens": 105473637.0,
      "step": 60815
    },
    {
      "entropy": 5.976289176940918,
      "epoch": 4.731997665823769,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002968111209250792,
      "loss": 4.9624,
      "mean_token_accuracy": 0.2125663548707962,
      "num_tokens": 105481975.0,
      "step": 60820
    },
    {
      "entropy": 5.9756158828735355,
      "epoch": 4.732386695195487,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029678353570161117,
      "loss": 5.0436,
      "mean_token_accuracy": 0.20751359760761262,
      "num_tokens": 105490464.0,
      "step": 60825
    },
    {
      "entropy": 5.970704555511475,
      "epoch": 4.732775724567205,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002967559501476126,
      "loss": 5.0516,
      "mean_token_accuracy": 0.20667077749967575,
      "num_tokens": 105498777.0,
      "step": 60830
    },
    {
      "entropy": 6.009219408035278,
      "epoch": 4.733164753938922,
      "grad_norm": 1.546875,
      "learning_rate": 0.00029672836426350213,
      "loss": 5.0919,
      "mean_token_accuracy": 0.20572975575923919,
      "num_tokens": 105507629.0,
      "step": 60835
    },
    {
      "entropy": 6.026579856872559,
      "epoch": 4.73355378331064,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029670077804969835,
      "loss": 5.0575,
      "mean_token_accuracy": 0.20777708739042283,
      "num_tokens": 105516193.0,
      "step": 60840
    },
    {
      "entropy": 5.957734537124634,
      "epoch": 4.733942812682358,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029667319150661976,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21069980412721634,
      "num_tokens": 105524912.0,
      "step": 60845
    },
    {
      "entropy": 5.912285900115966,
      "epoch": 4.734331842054075,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000296645604634685,
      "loss": 4.8911,
      "mean_token_accuracy": 0.2256319835782051,
      "num_tokens": 105533594.0,
      "step": 60850
    },
    {
      "entropy": 5.954019260406494,
      "epoch": 4.734720871425793,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002966180174343127,
      "loss": 4.9664,
      "mean_token_accuracy": 0.21778521090745925,
      "num_tokens": 105541896.0,
      "step": 60855
    },
    {
      "entropy": 5.923523473739624,
      "epoch": 4.73510990079751,
      "grad_norm": 1.375,
      "learning_rate": 0.00029659042990592125,
      "loss": 4.9674,
      "mean_token_accuracy": 0.21479176431894303,
      "num_tokens": 105551132.0,
      "step": 60860
    },
    {
      "entropy": 5.915620756149292,
      "epoch": 4.735498930169228,
      "grad_norm": 1.375,
      "learning_rate": 0.0002965628420499295,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22410782277584076,
      "num_tokens": 105560223.0,
      "step": 60865
    },
    {
      "entropy": 5.940981769561768,
      "epoch": 4.735887959540945,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002965352538667559,
      "loss": 4.923,
      "mean_token_accuracy": 0.22479704767465591,
      "num_tokens": 105569282.0,
      "step": 60870
    },
    {
      "entropy": 5.902200412750244,
      "epoch": 4.736276988912663,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029650766535681917,
      "loss": 4.87,
      "mean_token_accuracy": 0.22129233479499816,
      "num_tokens": 105577302.0,
      "step": 60875
    },
    {
      "entropy": 5.926599407196045,
      "epoch": 4.736666018284381,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00029648007652053774,
      "loss": 4.9775,
      "mean_token_accuracy": 0.2240156874060631,
      "num_tokens": 105586163.0,
      "step": 60880
    },
    {
      "entropy": 5.897101783752442,
      "epoch": 4.737055047656098,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029645248735833045,
      "loss": 4.9338,
      "mean_token_accuracy": 0.21011360585689545,
      "num_tokens": 105594739.0,
      "step": 60885
    },
    {
      "entropy": 5.911568355560303,
      "epoch": 4.737444077027815,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029642489787061574,
      "loss": 4.8915,
      "mean_token_accuracy": 0.21753791570663453,
      "num_tokens": 105604012.0,
      "step": 60890
    },
    {
      "entropy": 5.944912719726562,
      "epoch": 4.737833106399533,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002963973080578123,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21998822689056396,
      "num_tokens": 105612217.0,
      "step": 60895
    },
    {
      "entropy": 6.032860469818115,
      "epoch": 4.738222135771251,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029636971792033873,
      "loss": 5.1235,
      "mean_token_accuracy": 0.2030549317598343,
      "num_tokens": 105622062.0,
      "step": 60900
    },
    {
      "entropy": 5.939892435073853,
      "epoch": 4.738611165142968,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002963421274586138,
      "loss": 4.9551,
      "mean_token_accuracy": 0.22139153331518174,
      "num_tokens": 105630415.0,
      "step": 60905
    },
    {
      "entropy": 5.951379251480103,
      "epoch": 4.739000194514686,
      "grad_norm": 1.53125,
      "learning_rate": 0.00029631453667305603,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21685852557420732,
      "num_tokens": 105638631.0,
      "step": 60910
    },
    {
      "entropy": 5.997984123229981,
      "epoch": 4.739389223886404,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002962869455640841,
      "loss": 5.0164,
      "mean_token_accuracy": 0.21207151263952256,
      "num_tokens": 105647448.0,
      "step": 60915
    },
    {
      "entropy": 5.891494464874268,
      "epoch": 4.739778253258121,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002962593541321167,
      "loss": 4.8594,
      "mean_token_accuracy": 0.23076231628656388,
      "num_tokens": 105655600.0,
      "step": 60920
    },
    {
      "entropy": 5.936135482788086,
      "epoch": 4.740167282629838,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002962317623775723,
      "loss": 4.9083,
      "mean_token_accuracy": 0.22458963841199875,
      "num_tokens": 105663619.0,
      "step": 60925
    },
    {
      "entropy": 5.975107479095459,
      "epoch": 4.740556312001556,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029620417030086976,
      "loss": 4.9563,
      "mean_token_accuracy": 0.21841800659894944,
      "num_tokens": 105671835.0,
      "step": 60930
    },
    {
      "entropy": 5.978746509552002,
      "epoch": 4.740945341373274,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00029617657790242767,
      "loss": 4.9772,
      "mean_token_accuracy": 0.21409225314855576,
      "num_tokens": 105680334.0,
      "step": 60935
    },
    {
      "entropy": 5.94514536857605,
      "epoch": 4.741334370744991,
      "grad_norm": 1.484375,
      "learning_rate": 0.00029614898518266476,
      "loss": 5.0151,
      "mean_token_accuracy": 0.21008685529232024,
      "num_tokens": 105689374.0,
      "step": 60940
    },
    {
      "entropy": 5.929374265670776,
      "epoch": 4.741723400116709,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002961213921419997,
      "loss": 4.9669,
      "mean_token_accuracy": 0.2112218737602234,
      "num_tokens": 105698008.0,
      "step": 60945
    },
    {
      "entropy": 5.887220001220703,
      "epoch": 4.742112429488427,
      "grad_norm": 1.375,
      "learning_rate": 0.000296093798780851,
      "loss": 4.9481,
      "mean_token_accuracy": 0.2213636502623558,
      "num_tokens": 105706574.0,
      "step": 60950
    },
    {
      "entropy": 6.001836252212525,
      "epoch": 4.7425014588601435,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002960662050996377,
      "loss": 5.0464,
      "mean_token_accuracy": 0.21277009397745134,
      "num_tokens": 105715123.0,
      "step": 60955
    },
    {
      "entropy": 6.0183172702789305,
      "epoch": 4.742890488231861,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029603861109877807,
      "loss": 5.0486,
      "mean_token_accuracy": 0.2006484717130661,
      "num_tokens": 105723773.0,
      "step": 60960
    },
    {
      "entropy": 6.060317325592041,
      "epoch": 4.743279517603579,
      "grad_norm": 1.375,
      "learning_rate": 0.00029601101677869106,
      "loss": 5.084,
      "mean_token_accuracy": 0.20515455454587936,
      "num_tokens": 105732374.0,
      "step": 60965
    },
    {
      "entropy": 6.003231716156006,
      "epoch": 4.743668546975297,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002959834221397952,
      "loss": 5.0577,
      "mean_token_accuracy": 0.20738590657711028,
      "num_tokens": 105741527.0,
      "step": 60970
    },
    {
      "entropy": 5.91450252532959,
      "epoch": 4.744057576347014,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029595582718250954,
      "loss": 4.9147,
      "mean_token_accuracy": 0.22071204632520675,
      "num_tokens": 105750484.0,
      "step": 60975
    },
    {
      "entropy": 5.944791698455811,
      "epoch": 4.744446605718732,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002959282319072524,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21476669311523439,
      "num_tokens": 105759404.0,
      "step": 60980
    },
    {
      "entropy": 5.957839679718018,
      "epoch": 4.74483563509045,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002959006363144427,
      "loss": 4.9771,
      "mean_token_accuracy": 0.21243807226419448,
      "num_tokens": 105767188.0,
      "step": 60985
    },
    {
      "entropy": 5.95011682510376,
      "epoch": 4.7452246644621665,
      "grad_norm": 1.5,
      "learning_rate": 0.0002958730404044991,
      "loss": 5.0201,
      "mean_token_accuracy": 0.21458264142274858,
      "num_tokens": 105775862.0,
      "step": 60990
    },
    {
      "entropy": 5.887906789779663,
      "epoch": 4.745613693833884,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002958454441778403,
      "loss": 4.9093,
      "mean_token_accuracy": 0.2127907857298851,
      "num_tokens": 105784253.0,
      "step": 60995
    },
    {
      "entropy": 5.959939002990723,
      "epoch": 4.746002723205602,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029581784763488506,
      "loss": 5.0213,
      "mean_token_accuracy": 0.20706727504730224,
      "num_tokens": 105792943.0,
      "step": 61000
    },
    {
      "entropy": 5.986517333984375,
      "epoch": 4.74639175257732,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002957902507760522,
      "loss": 4.9538,
      "mean_token_accuracy": 0.22315257638692856,
      "num_tokens": 105801764.0,
      "step": 61005
    },
    {
      "entropy": 5.910501003265381,
      "epoch": 4.746780781949037,
      "grad_norm": 1.5,
      "learning_rate": 0.0002957626536017603,
      "loss": 4.9266,
      "mean_token_accuracy": 0.2222191348671913,
      "num_tokens": 105809842.0,
      "step": 61010
    },
    {
      "entropy": 5.953701782226562,
      "epoch": 4.747169811320755,
      "grad_norm": 1.5,
      "learning_rate": 0.0002957350561124282,
      "loss": 5.016,
      "mean_token_accuracy": 0.21612288057804108,
      "num_tokens": 105817535.0,
      "step": 61015
    },
    {
      "entropy": 5.916814661026001,
      "epoch": 4.747558840692472,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002957074583084747,
      "loss": 4.9272,
      "mean_token_accuracy": 0.221745041012764,
      "num_tokens": 105826838.0,
      "step": 61020
    },
    {
      "entropy": 6.031156158447265,
      "epoch": 4.7479478700641895,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002956798601903184,
      "loss": 5.0126,
      "mean_token_accuracy": 0.21212237924337388,
      "num_tokens": 105836603.0,
      "step": 61025
    },
    {
      "entropy": 6.025122261047363,
      "epoch": 4.748336899435907,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029565226175837814,
      "loss": 5.0365,
      "mean_token_accuracy": 0.20759295523166657,
      "num_tokens": 105845888.0,
      "step": 61030
    },
    {
      "entropy": 5.971699666976929,
      "epoch": 4.748725928807625,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002956246630130727,
      "loss": 4.9072,
      "mean_token_accuracy": 0.21607222706079482,
      "num_tokens": 105854736.0,
      "step": 61035
    },
    {
      "entropy": 5.959136724472046,
      "epoch": 4.749114958179343,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029559706395482075,
      "loss": 5.0049,
      "mean_token_accuracy": 0.2072744369506836,
      "num_tokens": 105863423.0,
      "step": 61040
    },
    {
      "entropy": 5.929248857498169,
      "epoch": 4.74950398755106,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029556946458404124,
      "loss": 4.931,
      "mean_token_accuracy": 0.21890706717967987,
      "num_tokens": 105871656.0,
      "step": 61045
    },
    {
      "entropy": 5.958563041687012,
      "epoch": 4.749893016922778,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029554186490115284,
      "loss": 5.0474,
      "mean_token_accuracy": 0.2163580521941185,
      "num_tokens": 105881119.0,
      "step": 61050
    },
    {
      "entropy": 5.964885282516479,
      "epoch": 4.750282046294496,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002955142649065743,
      "loss": 4.9827,
      "mean_token_accuracy": 0.21643858253955842,
      "num_tokens": 105889601.0,
      "step": 61055
    },
    {
      "entropy": 6.027698373794555,
      "epoch": 4.7506710756662125,
      "grad_norm": 1.375,
      "learning_rate": 0.0002954866646007244,
      "loss": 5.0,
      "mean_token_accuracy": 0.20681782215833663,
      "num_tokens": 105898106.0,
      "step": 61060
    },
    {
      "entropy": 5.921792125701904,
      "epoch": 4.75106010503793,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029545906398402205,
      "loss": 4.891,
      "mean_token_accuracy": 0.2256097063422203,
      "num_tokens": 105907403.0,
      "step": 61065
    },
    {
      "entropy": 5.930987119674683,
      "epoch": 4.751449134409648,
      "grad_norm": 1.5,
      "learning_rate": 0.00029543146305688594,
      "loss": 4.9998,
      "mean_token_accuracy": 0.2158896416425705,
      "num_tokens": 105916097.0,
      "step": 61070
    },
    {
      "entropy": 5.940780353546143,
      "epoch": 4.751838163781366,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002954038618197349,
      "loss": 4.8886,
      "mean_token_accuracy": 0.2247534692287445,
      "num_tokens": 105924511.0,
      "step": 61075
    },
    {
      "entropy": 6.03198938369751,
      "epoch": 4.752227193153083,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002953762602729877,
      "loss": 5.0705,
      "mean_token_accuracy": 0.20593287646770478,
      "num_tokens": 105933904.0,
      "step": 61080
    },
    {
      "entropy": 6.03869047164917,
      "epoch": 4.752616222524801,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029534865841706324,
      "loss": 4.997,
      "mean_token_accuracy": 0.21649632900953292,
      "num_tokens": 105942581.0,
      "step": 61085
    },
    {
      "entropy": 5.999623394012451,
      "epoch": 4.753005251896518,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002953210562523803,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21703209429979325,
      "num_tokens": 105950526.0,
      "step": 61090
    },
    {
      "entropy": 5.921844863891602,
      "epoch": 4.7533942812682355,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002952934537793576,
      "loss": 4.9216,
      "mean_token_accuracy": 0.21889466494321824,
      "num_tokens": 105959668.0,
      "step": 61095
    },
    {
      "entropy": 5.94790358543396,
      "epoch": 4.753783310639953,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029526585099841395,
      "loss": 5.0304,
      "mean_token_accuracy": 0.21065551340579985,
      "num_tokens": 105968556.0,
      "step": 61100
    },
    {
      "entropy": 5.953811836242676,
      "epoch": 4.754172340011671,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00029523824790996845,
      "loss": 4.9311,
      "mean_token_accuracy": 0.2210144206881523,
      "num_tokens": 105976709.0,
      "step": 61105
    },
    {
      "entropy": 5.933486604690552,
      "epoch": 4.754561369383389,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029521064451443966,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21330556869506836,
      "num_tokens": 105985490.0,
      "step": 61110
    },
    {
      "entropy": 6.007768011093139,
      "epoch": 4.754950398755106,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029518304081224656,
      "loss": 5.0859,
      "mean_token_accuracy": 0.20099729001522065,
      "num_tokens": 105994530.0,
      "step": 61115
    },
    {
      "entropy": 5.968001794815064,
      "epoch": 4.755339428126824,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029515543680380793,
      "loss": 4.9028,
      "mean_token_accuracy": 0.21956778913736344,
      "num_tokens": 106003960.0,
      "step": 61120
    },
    {
      "entropy": 5.915892457962036,
      "epoch": 4.755728457498541,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00029512783248954253,
      "loss": 4.932,
      "mean_token_accuracy": 0.2177500158548355,
      "num_tokens": 106013390.0,
      "step": 61125
    },
    {
      "entropy": 6.009138202667236,
      "epoch": 4.7561174868702585,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002951002278698695,
      "loss": 5.0537,
      "mean_token_accuracy": 0.21438927054405213,
      "num_tokens": 106021644.0,
      "step": 61130
    },
    {
      "entropy": 5.9981162548065186,
      "epoch": 4.756506516241976,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029507262294520726,
      "loss": 4.9431,
      "mean_token_accuracy": 0.22104644775390625,
      "num_tokens": 106030300.0,
      "step": 61135
    },
    {
      "entropy": 5.89173731803894,
      "epoch": 4.756895545613694,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002950450177159751,
      "loss": 4.8331,
      "mean_token_accuracy": 0.22320046573877333,
      "num_tokens": 106038965.0,
      "step": 61140
    },
    {
      "entropy": 5.999897146224976,
      "epoch": 4.757284574985412,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029501741218259167,
      "loss": 5.0022,
      "mean_token_accuracy": 0.22010022401809692,
      "num_tokens": 106047526.0,
      "step": 61145
    },
    {
      "entropy": 5.879130601882935,
      "epoch": 4.757673604357129,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029498980634547585,
      "loss": 4.8859,
      "mean_token_accuracy": 0.21886476874351501,
      "num_tokens": 106056429.0,
      "step": 61150
    },
    {
      "entropy": 5.920104837417602,
      "epoch": 4.758062633728846,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029496220020504656,
      "loss": 4.986,
      "mean_token_accuracy": 0.2122851312160492,
      "num_tokens": 106064596.0,
      "step": 61155
    },
    {
      "entropy": 5.968941020965576,
      "epoch": 4.758451663100564,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029493459376172263,
      "loss": 4.9876,
      "mean_token_accuracy": 0.2206583634018898,
      "num_tokens": 106072379.0,
      "step": 61160
    },
    {
      "entropy": 5.987525844573975,
      "epoch": 4.7588406924722815,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002949069870159229,
      "loss": 4.97,
      "mean_token_accuracy": 0.2173006147146225,
      "num_tokens": 106080910.0,
      "step": 61165
    },
    {
      "entropy": 5.9502486228942875,
      "epoch": 4.759229721843999,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029487937996806635,
      "loss": 4.9645,
      "mean_token_accuracy": 0.21928199529647827,
      "num_tokens": 106089814.0,
      "step": 61170
    },
    {
      "entropy": 5.96786470413208,
      "epoch": 4.759618751215717,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029485177261857193,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20693484246730803,
      "num_tokens": 106098206.0,
      "step": 61175
    },
    {
      "entropy": 5.885225534439087,
      "epoch": 4.760007780587435,
      "grad_norm": 1.5,
      "learning_rate": 0.0002948241649678584,
      "loss": 4.9176,
      "mean_token_accuracy": 0.22199296951293945,
      "num_tokens": 106107516.0,
      "step": 61180
    },
    {
      "entropy": 5.893211364746094,
      "epoch": 4.760396809959152,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002947965570163448,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2103322848677635,
      "num_tokens": 106116407.0,
      "step": 61185
    },
    {
      "entropy": 5.976173543930054,
      "epoch": 4.76078583933087,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002947689487644499,
      "loss": 5.0779,
      "mean_token_accuracy": 0.2085563987493515,
      "num_tokens": 106125761.0,
      "step": 61190
    },
    {
      "entropy": 6.0021806240081785,
      "epoch": 4.761174868702587,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002947413402125927,
      "loss": 4.9927,
      "mean_token_accuracy": 0.21025941967964173,
      "num_tokens": 106135220.0,
      "step": 61195
    },
    {
      "entropy": 6.054682445526123,
      "epoch": 4.7615638980743045,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002947137313611919,
      "loss": 5.0172,
      "mean_token_accuracy": 0.2104533925652504,
      "num_tokens": 106143851.0,
      "step": 61200
    },
    {
      "entropy": 5.972072982788086,
      "epoch": 4.761952927446022,
      "grad_norm": 1.546875,
      "learning_rate": 0.00029468612221066683,
      "loss": 4.9278,
      "mean_token_accuracy": 0.2175328776240349,
      "num_tokens": 106152681.0,
      "step": 61205
    },
    {
      "entropy": 5.921595668792724,
      "epoch": 4.76234195681774,
      "grad_norm": 1.5,
      "learning_rate": 0.0002946585127614361,
      "loss": 5.0083,
      "mean_token_accuracy": 0.21182254254817962,
      "num_tokens": 106161805.0,
      "step": 61210
    },
    {
      "entropy": 5.985337734222412,
      "epoch": 4.762730986189458,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002946309030139188,
      "loss": 5.0528,
      "mean_token_accuracy": 0.21301131099462509,
      "num_tokens": 106170308.0,
      "step": 61215
    },
    {
      "entropy": 6.025498294830323,
      "epoch": 4.763120015561174,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002946032929685337,
      "loss": 5.0587,
      "mean_token_accuracy": 0.2182350665330887,
      "num_tokens": 106179280.0,
      "step": 61220
    },
    {
      "entropy": 5.917948532104492,
      "epoch": 4.763509044932892,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002945756826256999,
      "loss": 4.9209,
      "mean_token_accuracy": 0.21364012211561204,
      "num_tokens": 106188467.0,
      "step": 61225
    },
    {
      "entropy": 5.927189683914184,
      "epoch": 4.76389807430461,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002945480719858363,
      "loss": 4.9309,
      "mean_token_accuracy": 0.21511257588863372,
      "num_tokens": 106197479.0,
      "step": 61230
    },
    {
      "entropy": 6.102901363372803,
      "epoch": 4.7642871036763275,
      "grad_norm": 1.5,
      "learning_rate": 0.0002945204610493617,
      "loss": 5.0121,
      "mean_token_accuracy": 0.21197570860385895,
      "num_tokens": 106205608.0,
      "step": 61235
    },
    {
      "entropy": 5.960202360153199,
      "epoch": 4.764676133048045,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029449284981669524,
      "loss": 4.96,
      "mean_token_accuracy": 0.21512148827314376,
      "num_tokens": 106214719.0,
      "step": 61240
    },
    {
      "entropy": 5.913327980041504,
      "epoch": 4.765065162419763,
      "grad_norm": 1.515625,
      "learning_rate": 0.00029446523828825577,
      "loss": 4.9424,
      "mean_token_accuracy": 0.21678148657083512,
      "num_tokens": 106223372.0,
      "step": 61245
    },
    {
      "entropy": 5.923130178451538,
      "epoch": 4.7654541917914806,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002944376264644624,
      "loss": 4.9586,
      "mean_token_accuracy": 0.21946075558662415,
      "num_tokens": 106231342.0,
      "step": 61250
    },
    {
      "entropy": 5.958522319793701,
      "epoch": 4.765843221163198,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002944100143457339,
      "loss": 5.0045,
      "mean_token_accuracy": 0.2105649530887604,
      "num_tokens": 106238702.0,
      "step": 61255
    },
    {
      "entropy": 5.984606313705444,
      "epoch": 4.766232250534915,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002943824019324893,
      "loss": 4.9597,
      "mean_token_accuracy": 0.2206313580274582,
      "num_tokens": 106246585.0,
      "step": 61260
    },
    {
      "entropy": 5.923702096939087,
      "epoch": 4.766621279906633,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002943547892251476,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21878782361745835,
      "num_tokens": 106254685.0,
      "step": 61265
    },
    {
      "entropy": 5.880874347686768,
      "epoch": 4.7670103092783505,
      "grad_norm": 1.578125,
      "learning_rate": 0.00029432717622412784,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21842967867851257,
      "num_tokens": 106263115.0,
      "step": 61270
    },
    {
      "entropy": 5.964881229400635,
      "epoch": 4.767399338650068,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029429956292984893,
      "loss": 5.0129,
      "mean_token_accuracy": 0.20675815939903258,
      "num_tokens": 106271621.0,
      "step": 61275
    },
    {
      "entropy": 5.899925422668457,
      "epoch": 4.767788368021786,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002942719493427299,
      "loss": 4.8709,
      "mean_token_accuracy": 0.2250313177704811,
      "num_tokens": 106280276.0,
      "step": 61280
    },
    {
      "entropy": 5.958998823165894,
      "epoch": 4.7681773973935035,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002942443354631897,
      "loss": 4.9166,
      "mean_token_accuracy": 0.2255090057849884,
      "num_tokens": 106288117.0,
      "step": 61285
    },
    {
      "entropy": 5.8781343460083,
      "epoch": 4.76856642676522,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029421672129164726,
      "loss": 4.9348,
      "mean_token_accuracy": 0.22266266644001007,
      "num_tokens": 106296794.0,
      "step": 61290
    },
    {
      "entropy": 5.848884773254395,
      "epoch": 4.768955456136938,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029418910682852176,
      "loss": 4.9227,
      "mean_token_accuracy": 0.22358662784099578,
      "num_tokens": 106305706.0,
      "step": 61295
    },
    {
      "entropy": 5.894838809967041,
      "epoch": 4.769344485508656,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029416149207423203,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21766692101955415,
      "num_tokens": 106314827.0,
      "step": 61300
    },
    {
      "entropy": 5.948349285125732,
      "epoch": 4.7697335148803734,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029413387702919723,
      "loss": 4.9466,
      "mean_token_accuracy": 0.2217831239104271,
      "num_tokens": 106323086.0,
      "step": 61305
    },
    {
      "entropy": 5.932986259460449,
      "epoch": 4.770122544252091,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029410626169383623,
      "loss": 4.9807,
      "mean_token_accuracy": 0.21633775234222413,
      "num_tokens": 106332152.0,
      "step": 61310
    },
    {
      "entropy": 5.946397924423218,
      "epoch": 4.770511573623809,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029407864606856814,
      "loss": 4.9208,
      "mean_token_accuracy": 0.2188173398375511,
      "num_tokens": 106341125.0,
      "step": 61315
    },
    {
      "entropy": 5.9035626411437985,
      "epoch": 4.7709006029955265,
      "grad_norm": 1.4375,
      "learning_rate": 0.000294051030153812,
      "loss": 4.8578,
      "mean_token_accuracy": 0.2178629919886589,
      "num_tokens": 106349825.0,
      "step": 61320
    },
    {
      "entropy": 5.94481840133667,
      "epoch": 4.771289632367243,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002940234139499868,
      "loss": 4.9365,
      "mean_token_accuracy": 0.21076703667640687,
      "num_tokens": 106358081.0,
      "step": 61325
    },
    {
      "entropy": 5.882958889007568,
      "epoch": 4.771678661738961,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029399579745751156,
      "loss": 4.9758,
      "mean_token_accuracy": 0.21090442091226577,
      "num_tokens": 106366624.0,
      "step": 61330
    },
    {
      "entropy": 5.9713812351226805,
      "epoch": 4.772067691110679,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00029396818067680525,
      "loss": 4.9817,
      "mean_token_accuracy": 0.2186049520969391,
      "num_tokens": 106374584.0,
      "step": 61335
    },
    {
      "entropy": 6.042605352401734,
      "epoch": 4.772456720482396,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029394056360828707,
      "loss": 4.9759,
      "mean_token_accuracy": 0.21666189283132553,
      "num_tokens": 106383553.0,
      "step": 61340
    },
    {
      "entropy": 5.983492183685303,
      "epoch": 4.772845749854114,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029391294625237595,
      "loss": 4.9441,
      "mean_token_accuracy": 0.21907827109098435,
      "num_tokens": 106392431.0,
      "step": 61345
    },
    {
      "entropy": 5.862145137786865,
      "epoch": 4.773234779225832,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002938853286094911,
      "loss": 4.8869,
      "mean_token_accuracy": 0.21722941100597382,
      "num_tokens": 106400766.0,
      "step": 61350
    },
    {
      "entropy": 5.952121877670288,
      "epoch": 4.773623808597549,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029385771068005123,
      "loss": 5.0138,
      "mean_token_accuracy": 0.21859457790851594,
      "num_tokens": 106409484.0,
      "step": 61355
    },
    {
      "entropy": 5.9057807445526125,
      "epoch": 4.774012837969266,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002938300924644757,
      "loss": 4.9436,
      "mean_token_accuracy": 0.21534850746393203,
      "num_tokens": 106418346.0,
      "step": 61360
    },
    {
      "entropy": 5.926250982284546,
      "epoch": 4.774401867340984,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029380247396318355,
      "loss": 4.9524,
      "mean_token_accuracy": 0.2181362509727478,
      "num_tokens": 106427036.0,
      "step": 61365
    },
    {
      "entropy": 5.884410667419433,
      "epoch": 4.774790896712702,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002937748551765937,
      "loss": 4.9217,
      "mean_token_accuracy": 0.21096862703561783,
      "num_tokens": 106436466.0,
      "step": 61370
    },
    {
      "entropy": 6.032669305801392,
      "epoch": 4.775179926084419,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002937472361051254,
      "loss": 5.0932,
      "mean_token_accuracy": 0.20683833956718445,
      "num_tokens": 106444715.0,
      "step": 61375
    },
    {
      "entropy": 5.998015022277832,
      "epoch": 4.775568955456137,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002937196167491976,
      "loss": 4.9229,
      "mean_token_accuracy": 0.21962317526340486,
      "num_tokens": 106452508.0,
      "step": 61380
    },
    {
      "entropy": 5.949276447296143,
      "epoch": 4.775957984827855,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029369199710922944,
      "loss": 4.9289,
      "mean_token_accuracy": 0.220414000749588,
      "num_tokens": 106461194.0,
      "step": 61385
    },
    {
      "entropy": 5.86795711517334,
      "epoch": 4.7763470141995725,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00029366437718563994,
      "loss": 5.035,
      "mean_token_accuracy": 0.20605459362268447,
      "num_tokens": 106470450.0,
      "step": 61390
    },
    {
      "entropy": 5.99954776763916,
      "epoch": 4.776736043571289,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002936367569788482,
      "loss": 5.0247,
      "mean_token_accuracy": 0.20747994184494017,
      "num_tokens": 106479338.0,
      "step": 61395
    },
    {
      "entropy": 6.005270290374756,
      "epoch": 4.777125072943007,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002936091364892733,
      "loss": 4.9403,
      "mean_token_accuracy": 0.21769641041755677,
      "num_tokens": 106488035.0,
      "step": 61400
    },
    {
      "entropy": 5.9399219989776615,
      "epoch": 4.777514102314725,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029358151571733446,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20748697966337204,
      "num_tokens": 106496190.0,
      "step": 61405
    },
    {
      "entropy": 5.931431245803833,
      "epoch": 4.777903131686442,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00029355389466345064,
      "loss": 5.0024,
      "mean_token_accuracy": 0.20970351994037628,
      "num_tokens": 106504894.0,
      "step": 61410
    },
    {
      "entropy": 5.992014169692993,
      "epoch": 4.77829216105816,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029352627332804104,
      "loss": 5.0482,
      "mean_token_accuracy": 0.20823558270931244,
      "num_tokens": 106513506.0,
      "step": 61415
    },
    {
      "entropy": 5.973284530639648,
      "epoch": 4.778681190429878,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029349865171152475,
      "loss": 4.8937,
      "mean_token_accuracy": 0.2211463361978531,
      "num_tokens": 106522354.0,
      "step": 61420
    },
    {
      "entropy": 6.0235950469970705,
      "epoch": 4.779070219801595,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029347102981432084,
      "loss": 5.1218,
      "mean_token_accuracy": 0.2072615757584572,
      "num_tokens": 106531516.0,
      "step": 61425
    },
    {
      "entropy": 5.907263469696045,
      "epoch": 4.779459249173312,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002934434076368484,
      "loss": 4.9271,
      "mean_token_accuracy": 0.2165688768029213,
      "num_tokens": 106539759.0,
      "step": 61430
    },
    {
      "entropy": 5.994666194915771,
      "epoch": 4.77984827854503,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002934157851795266,
      "loss": 5.0082,
      "mean_token_accuracy": 0.21297886669635774,
      "num_tokens": 106548421.0,
      "step": 61435
    },
    {
      "entropy": 6.004288053512573,
      "epoch": 4.780237307916748,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002933881624427746,
      "loss": 5.0622,
      "mean_token_accuracy": 0.21275173723697663,
      "num_tokens": 106556919.0,
      "step": 61440
    },
    {
      "entropy": 5.922523832321167,
      "epoch": 4.780626337288465,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002933605394270116,
      "loss": 4.8604,
      "mean_token_accuracy": 0.22581852972507477,
      "num_tokens": 106565394.0,
      "step": 61445
    },
    {
      "entropy": 5.962975931167603,
      "epoch": 4.781015366660183,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002933329161326565,
      "loss": 5.0025,
      "mean_token_accuracy": 0.2135481745004654,
      "num_tokens": 106574669.0,
      "step": 61450
    },
    {
      "entropy": 5.962053632736206,
      "epoch": 4.781404396031901,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029330529256012863,
      "loss": 4.9289,
      "mean_token_accuracy": 0.2167219340801239,
      "num_tokens": 106582505.0,
      "step": 61455
    },
    {
      "entropy": 5.966962671279907,
      "epoch": 4.781793425403618,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029327766870984714,
      "loss": 4.9295,
      "mean_token_accuracy": 0.22376034855842591,
      "num_tokens": 106591777.0,
      "step": 61460
    },
    {
      "entropy": 5.943751287460327,
      "epoch": 4.782182454775335,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029325004458223097,
      "loss": 4.9679,
      "mean_token_accuracy": 0.21213174164295195,
      "num_tokens": 106600951.0,
      "step": 61465
    },
    {
      "entropy": 5.86459584236145,
      "epoch": 4.782571484147053,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00029322242017769954,
      "loss": 4.8405,
      "mean_token_accuracy": 0.224060720205307,
      "num_tokens": 106608955.0,
      "step": 61470
    },
    {
      "entropy": 6.01353349685669,
      "epoch": 4.782960513518771,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029319479549667183,
      "loss": 5.0709,
      "mean_token_accuracy": 0.20552381724119187,
      "num_tokens": 106617778.0,
      "step": 61475
    },
    {
      "entropy": 5.987853956222534,
      "epoch": 4.783349542890488,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002931671705395671,
      "loss": 5.0045,
      "mean_token_accuracy": 0.2065620929002762,
      "num_tokens": 106626781.0,
      "step": 61480
    },
    {
      "entropy": 5.992259454727173,
      "epoch": 4.783738572262206,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00029313954530680447,
      "loss": 4.9501,
      "mean_token_accuracy": 0.21295444816350936,
      "num_tokens": 106635013.0,
      "step": 61485
    },
    {
      "entropy": 5.908585882186889,
      "epoch": 4.784127601633923,
      "grad_norm": 1.53125,
      "learning_rate": 0.00029311191979880313,
      "loss": 4.8671,
      "mean_token_accuracy": 0.2235936164855957,
      "num_tokens": 106643473.0,
      "step": 61490
    },
    {
      "entropy": 5.887950038909912,
      "epoch": 4.784516631005641,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002930842940159822,
      "loss": 4.9539,
      "mean_token_accuracy": 0.21317344009876252,
      "num_tokens": 106652681.0,
      "step": 61495
    },
    {
      "entropy": 5.949651575088501,
      "epoch": 4.784905660377358,
      "grad_norm": 1.375,
      "learning_rate": 0.00029305666795876093,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21482092887163162,
      "num_tokens": 106660988.0,
      "step": 61500
    },
    {
      "entropy": 5.927076768875122,
      "epoch": 4.785294689749076,
      "grad_norm": 1.375,
      "learning_rate": 0.00029302904162755843,
      "loss": 4.8964,
      "mean_token_accuracy": 0.22049204558134078,
      "num_tokens": 106669799.0,
      "step": 61505
    },
    {
      "entropy": 6.008852815628051,
      "epoch": 4.785683719120794,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029300141502279395,
      "loss": 5.0382,
      "mean_token_accuracy": 0.21154449880123138,
      "num_tokens": 106678730.0,
      "step": 61510
    },
    {
      "entropy": 5.897548675537109,
      "epoch": 4.786072748492511,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029297378814488657,
      "loss": 4.9092,
      "mean_token_accuracy": 0.2300153598189354,
      "num_tokens": 106687819.0,
      "step": 61515
    },
    {
      "entropy": 5.960308980941773,
      "epoch": 4.786461777864229,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029294616099425556,
      "loss": 4.9783,
      "mean_token_accuracy": 0.2101923868060112,
      "num_tokens": 106697033.0,
      "step": 61520
    },
    {
      "entropy": 5.969588327407837,
      "epoch": 4.786850807235947,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002929185335713202,
      "loss": 4.9865,
      "mean_token_accuracy": 0.20933570116758346,
      "num_tokens": 106705076.0,
      "step": 61525
    },
    {
      "entropy": 5.960909652709961,
      "epoch": 4.787239836607664,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029289090587649965,
      "loss": 4.9561,
      "mean_token_accuracy": 0.2112593248486519,
      "num_tokens": 106713946.0,
      "step": 61530
    },
    {
      "entropy": 5.881396770477295,
      "epoch": 4.787628865979381,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000292863277910213,
      "loss": 4.8542,
      "mean_token_accuracy": 0.22174736261367797,
      "num_tokens": 106722671.0,
      "step": 61535
    },
    {
      "entropy": 5.968640375137329,
      "epoch": 4.788017895351099,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002928356496728796,
      "loss": 4.9439,
      "mean_token_accuracy": 0.22249559313058853,
      "num_tokens": 106730711.0,
      "step": 61540
    },
    {
      "entropy": 6.02301139831543,
      "epoch": 4.788406924722817,
      "grad_norm": 1.546875,
      "learning_rate": 0.00029280802116491854,
      "loss": 5.0764,
      "mean_token_accuracy": 0.20820405781269075,
      "num_tokens": 106738918.0,
      "step": 61545
    },
    {
      "entropy": 5.997980833053589,
      "epoch": 4.788795954094534,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029278039238674915,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21173553615808488,
      "num_tokens": 106748327.0,
      "step": 61550
    },
    {
      "entropy": 6.080959987640381,
      "epoch": 4.789184983466251,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002927527633387906,
      "loss": 5.0619,
      "mean_token_accuracy": 0.21607758700847626,
      "num_tokens": 106757029.0,
      "step": 61555
    },
    {
      "entropy": 6.0694502830505375,
      "epoch": 4.789574012837969,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029272513402146203,
      "loss": 5.0915,
      "mean_token_accuracy": 0.2029522478580475,
      "num_tokens": 106766459.0,
      "step": 61560
    },
    {
      "entropy": 5.941664314270019,
      "epoch": 4.789963042209687,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002926975044351828,
      "loss": 4.9722,
      "mean_token_accuracy": 0.2114310473203659,
      "num_tokens": 106775841.0,
      "step": 61565
    },
    {
      "entropy": 5.944850826263428,
      "epoch": 4.790352071581404,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002926698745803723,
      "loss": 4.8935,
      "mean_token_accuracy": 0.21420353800058364,
      "num_tokens": 106784684.0,
      "step": 61570
    },
    {
      "entropy": 5.946134185791015,
      "epoch": 4.790741100953122,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029264224445744937,
      "loss": 4.8415,
      "mean_token_accuracy": 0.22889447957277298,
      "num_tokens": 106792932.0,
      "step": 61575
    },
    {
      "entropy": 5.900412225723267,
      "epoch": 4.79113013032484,
      "grad_norm": 1.5,
      "learning_rate": 0.00029261461406683353,
      "loss": 4.951,
      "mean_token_accuracy": 0.21444661915302277,
      "num_tokens": 106801753.0,
      "step": 61580
    },
    {
      "entropy": 5.922268342971802,
      "epoch": 4.791519159696557,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029258698340894397,
      "loss": 5.0045,
      "mean_token_accuracy": 0.213982093334198,
      "num_tokens": 106810710.0,
      "step": 61585
    },
    {
      "entropy": 5.988828468322754,
      "epoch": 4.791908189068275,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00029255935248419983,
      "loss": 5.0077,
      "mean_token_accuracy": 0.208380825817585,
      "num_tokens": 106818677.0,
      "step": 61590
    },
    {
      "entropy": 5.968517732620239,
      "epoch": 4.792297218439992,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002925317212930206,
      "loss": 4.9871,
      "mean_token_accuracy": 0.21657801419496536,
      "num_tokens": 106828164.0,
      "step": 61595
    },
    {
      "entropy": 5.978539562225341,
      "epoch": 4.79268624781171,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029250408983582527,
      "loss": 4.987,
      "mean_token_accuracy": 0.2154807835817337,
      "num_tokens": 106837545.0,
      "step": 61600
    },
    {
      "entropy": 5.998735761642456,
      "epoch": 4.793075277183427,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002924764581130333,
      "loss": 5.015,
      "mean_token_accuracy": 0.2064029887318611,
      "num_tokens": 106845864.0,
      "step": 61605
    },
    {
      "entropy": 6.050983047485351,
      "epoch": 4.793464306555145,
      "grad_norm": 1.578125,
      "learning_rate": 0.000292448826125064,
      "loss": 5.0466,
      "mean_token_accuracy": 0.20993514508008956,
      "num_tokens": 106853744.0,
      "step": 61610
    },
    {
      "entropy": 5.850368499755859,
      "epoch": 4.793853335926863,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002924211938723364,
      "loss": 4.8146,
      "mean_token_accuracy": 0.21976527571678162,
      "num_tokens": 106862847.0,
      "step": 61615
    },
    {
      "entropy": 5.928741312026977,
      "epoch": 4.79424236529858,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029239356135526994,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21686784625053407,
      "num_tokens": 106871361.0,
      "step": 61620
    },
    {
      "entropy": 5.9424224376678465,
      "epoch": 4.794631394670297,
      "grad_norm": 1.359375,
      "learning_rate": 0.000292365928574284,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21383305490016938,
      "num_tokens": 106880512.0,
      "step": 61625
    },
    {
      "entropy": 5.954893779754639,
      "epoch": 4.795020424042015,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002923382955297975,
      "loss": 4.9509,
      "mean_token_accuracy": 0.22274935990571976,
      "num_tokens": 106888602.0,
      "step": 61630
    },
    {
      "entropy": 5.981440687179566,
      "epoch": 4.795409453413733,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002923106622222301,
      "loss": 5.0443,
      "mean_token_accuracy": 0.21654405295848847,
      "num_tokens": 106898031.0,
      "step": 61635
    },
    {
      "entropy": 5.935976934432984,
      "epoch": 4.79579848278545,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002922830286520009,
      "loss": 4.9585,
      "mean_token_accuracy": 0.21700815707445145,
      "num_tokens": 106907340.0,
      "step": 61640
    },
    {
      "entropy": 5.9965332508087155,
      "epoch": 4.796187512157168,
      "grad_norm": 1.53125,
      "learning_rate": 0.00029225539481952934,
      "loss": 5.0099,
      "mean_token_accuracy": 0.21199132949113847,
      "num_tokens": 106915786.0,
      "step": 61645
    },
    {
      "entropy": 5.9836585521698,
      "epoch": 4.796576541528886,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002922277607252346,
      "loss": 5.0016,
      "mean_token_accuracy": 0.2111242264509201,
      "num_tokens": 106925094.0,
      "step": 61650
    },
    {
      "entropy": 5.984159851074219,
      "epoch": 4.796965570900603,
      "grad_norm": 1.421875,
      "learning_rate": 0.000292200126369536,
      "loss": 5.0768,
      "mean_token_accuracy": 0.20464636236429215,
      "num_tokens": 106933696.0,
      "step": 61655
    },
    {
      "entropy": 6.023592901229859,
      "epoch": 4.79735460027232,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029217249175285284,
      "loss": 5.0664,
      "mean_token_accuracy": 0.20999234169721603,
      "num_tokens": 106943280.0,
      "step": 61660
    },
    {
      "entropy": 5.980923795700074,
      "epoch": 4.797743629644038,
      "grad_norm": 1.375,
      "learning_rate": 0.00029214485687560437,
      "loss": 4.9509,
      "mean_token_accuracy": 0.20966610312461853,
      "num_tokens": 106951822.0,
      "step": 61665
    },
    {
      "entropy": 5.946805715560913,
      "epoch": 4.798132659015756,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002921172217382101,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21902274042367936,
      "num_tokens": 106959866.0,
      "step": 61670
    },
    {
      "entropy": 5.916334533691407,
      "epoch": 4.798521688387473,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029208958634108917,
      "loss": 4.9095,
      "mean_token_accuracy": 0.22238699346780777,
      "num_tokens": 106968664.0,
      "step": 61675
    },
    {
      "entropy": 6.013497972488404,
      "epoch": 4.798910717759191,
      "grad_norm": 1.40625,
      "learning_rate": 0.000292061950684661,
      "loss": 5.0218,
      "mean_token_accuracy": 0.2110625147819519,
      "num_tokens": 106977792.0,
      "step": 61680
    },
    {
      "entropy": 5.974016904830933,
      "epoch": 4.799299747130909,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029203431476934496,
      "loss": 5.0035,
      "mean_token_accuracy": 0.2198063999414444,
      "num_tokens": 106986877.0,
      "step": 61685
    },
    {
      "entropy": 5.917117309570313,
      "epoch": 4.7996887765026255,
      "grad_norm": 1.53125,
      "learning_rate": 0.00029200667859556026,
      "loss": 4.8676,
      "mean_token_accuracy": 0.2186676010489464,
      "num_tokens": 106995388.0,
      "step": 61690
    },
    {
      "entropy": 5.9187798500061035,
      "epoch": 4.800077805874343,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002919790421637262,
      "loss": 4.8634,
      "mean_token_accuracy": 0.21949893236160278,
      "num_tokens": 107004617.0,
      "step": 61695
    },
    {
      "entropy": 5.945383501052857,
      "epoch": 4.800466835246061,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002919514054742622,
      "loss": 4.9517,
      "mean_token_accuracy": 0.2155870497226715,
      "num_tokens": 107013892.0,
      "step": 61700
    },
    {
      "entropy": 5.8845862865448,
      "epoch": 4.800855864617779,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00029192376852758774,
      "loss": 4.8461,
      "mean_token_accuracy": 0.21898577213287354,
      "num_tokens": 107022258.0,
      "step": 61705
    },
    {
      "entropy": 5.8645405769348145,
      "epoch": 4.801244893989496,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002918961313241219,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2260574385523796,
      "num_tokens": 107031196.0,
      "step": 61710
    },
    {
      "entropy": 5.921511030197143,
      "epoch": 4.801633923361214,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002918684938642842,
      "loss": 4.8821,
      "mean_token_accuracy": 0.2251448392868042,
      "num_tokens": 107039662.0,
      "step": 61715
    },
    {
      "entropy": 5.941891241073608,
      "epoch": 4.802022952732932,
      "grad_norm": 1.375,
      "learning_rate": 0.00029184085614849403,
      "loss": 4.9632,
      "mean_token_accuracy": 0.2125189870595932,
      "num_tokens": 107048855.0,
      "step": 61720
    },
    {
      "entropy": 5.946382904052735,
      "epoch": 4.802411982104649,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002918132181771706,
      "loss": 4.9297,
      "mean_token_accuracy": 0.21203707456588744,
      "num_tokens": 107057687.0,
      "step": 61725
    },
    {
      "entropy": 6.003921031951904,
      "epoch": 4.802801011476366,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029178557995073335,
      "loss": 4.9811,
      "mean_token_accuracy": 0.2174824669957161,
      "num_tokens": 107065692.0,
      "step": 61730
    },
    {
      "entropy": 5.905394887924194,
      "epoch": 4.803190040848084,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002917579414696017,
      "loss": 4.8713,
      "mean_token_accuracy": 0.21708856970071794,
      "num_tokens": 107073975.0,
      "step": 61735
    },
    {
      "entropy": 5.9792173385620115,
      "epoch": 4.803579070219802,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029173030273419484,
      "loss": 5.001,
      "mean_token_accuracy": 0.21285705715417863,
      "num_tokens": 107082566.0,
      "step": 61740
    },
    {
      "entropy": 6.004631471633911,
      "epoch": 4.803968099591519,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002917026637449323,
      "loss": 5.0456,
      "mean_token_accuracy": 0.20810424238443376,
      "num_tokens": 107091293.0,
      "step": 61745
    },
    {
      "entropy": 5.974508810043335,
      "epoch": 4.804357128963237,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002916750245022335,
      "loss": 4.9306,
      "mean_token_accuracy": 0.22276796102523805,
      "num_tokens": 107099978.0,
      "step": 61750
    },
    {
      "entropy": 5.951584529876709,
      "epoch": 4.804746158334955,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002916473850065177,
      "loss": 5.0232,
      "mean_token_accuracy": 0.20538302809000014,
      "num_tokens": 107108467.0,
      "step": 61755
    },
    {
      "entropy": 5.925952196121216,
      "epoch": 4.8051351877066715,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029161974525820434,
      "loss": 4.893,
      "mean_token_accuracy": 0.2125195875763893,
      "num_tokens": 107117715.0,
      "step": 61760
    },
    {
      "entropy": 6.039210462570191,
      "epoch": 4.805524217078389,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002915921052577127,
      "loss": 5.0412,
      "mean_token_accuracy": 0.21319814771413803,
      "num_tokens": 107126693.0,
      "step": 61765
    },
    {
      "entropy": 5.997626781463623,
      "epoch": 4.805913246450107,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029156446500546243,
      "loss": 5.0173,
      "mean_token_accuracy": 0.21433499455451965,
      "num_tokens": 107136150.0,
      "step": 61770
    },
    {
      "entropy": 6.036182308197022,
      "epoch": 4.806302275821825,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029153682450187265,
      "loss": 4.9564,
      "mean_token_accuracy": 0.2097330555319786,
      "num_tokens": 107144776.0,
      "step": 61775
    },
    {
      "entropy": 5.929311609268188,
      "epoch": 4.806691305193542,
      "grad_norm": 1.5,
      "learning_rate": 0.00029150918374736294,
      "loss": 4.9013,
      "mean_token_accuracy": 0.21806749999523162,
      "num_tokens": 107152993.0,
      "step": 61780
    },
    {
      "entropy": 5.926703643798828,
      "epoch": 4.80708033456526,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002914815427423526,
      "loss": 4.9062,
      "mean_token_accuracy": 0.22964268922805786,
      "num_tokens": 107161871.0,
      "step": 61785
    },
    {
      "entropy": 6.020199918746949,
      "epoch": 4.807469363936978,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002914539014872612,
      "loss": 5.101,
      "mean_token_accuracy": 0.19928285777568816,
      "num_tokens": 107171145.0,
      "step": 61790
    },
    {
      "entropy": 6.011377239227295,
      "epoch": 4.8078583933086945,
      "grad_norm": 1.5,
      "learning_rate": 0.00029142625998250786,
      "loss": 5.0397,
      "mean_token_accuracy": 0.21381450742483138,
      "num_tokens": 107179718.0,
      "step": 61795
    },
    {
      "entropy": 5.971212005615234,
      "epoch": 4.808247422680412,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029139861822851233,
      "loss": 4.9696,
      "mean_token_accuracy": 0.21846384257078172,
      "num_tokens": 107187753.0,
      "step": 61800
    },
    {
      "entropy": 5.862743997573853,
      "epoch": 4.80863645205213,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029137097622569376,
      "loss": 4.9106,
      "mean_token_accuracy": 0.216135635972023,
      "num_tokens": 107196382.0,
      "step": 61805
    },
    {
      "entropy": 5.916265821456909,
      "epoch": 4.809025481423848,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002913433339744718,
      "loss": 4.9616,
      "mean_token_accuracy": 0.22321856021881104,
      "num_tokens": 107204750.0,
      "step": 61810
    },
    {
      "entropy": 5.94086594581604,
      "epoch": 4.809414510795565,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002913156914752656,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21400032341480255,
      "num_tokens": 107213860.0,
      "step": 61815
    },
    {
      "entropy": 6.041183567047119,
      "epoch": 4.809803540167283,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002912880487284948,
      "loss": 5.0554,
      "mean_token_accuracy": 0.2072896331548691,
      "num_tokens": 107222737.0,
      "step": 61820
    },
    {
      "entropy": 6.014329481124878,
      "epoch": 4.810192569539,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029126040573457886,
      "loss": 5.0074,
      "mean_token_accuracy": 0.21232164204120635,
      "num_tokens": 107232006.0,
      "step": 61825
    },
    {
      "entropy": 6.057813405990601,
      "epoch": 4.8105815989107175,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029123276249393707,
      "loss": 5.0503,
      "mean_token_accuracy": 0.20954277813434602,
      "num_tokens": 107240175.0,
      "step": 61830
    },
    {
      "entropy": 5.982154750823975,
      "epoch": 4.810970628282435,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029120511900698894,
      "loss": 5.0268,
      "mean_token_accuracy": 0.20864997953176498,
      "num_tokens": 107248344.0,
      "step": 61835
    },
    {
      "entropy": 5.906527185440064,
      "epoch": 4.811359657654153,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000291177475274154,
      "loss": 4.967,
      "mean_token_accuracy": 0.22124786227941512,
      "num_tokens": 107257849.0,
      "step": 61840
    },
    {
      "entropy": 5.934559059143067,
      "epoch": 4.8117486870258706,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002911498312958516,
      "loss": 4.9968,
      "mean_token_accuracy": 0.21305757910013198,
      "num_tokens": 107267065.0,
      "step": 61845
    },
    {
      "entropy": 6.020502042770386,
      "epoch": 4.812137716397588,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029112218707250115,
      "loss": 5.0174,
      "mean_token_accuracy": 0.21275440752506256,
      "num_tokens": 107275460.0,
      "step": 61850
    },
    {
      "entropy": 5.927441024780274,
      "epoch": 4.812526745769306,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029109454260452223,
      "loss": 4.9216,
      "mean_token_accuracy": 0.2153344377875328,
      "num_tokens": 107284187.0,
      "step": 61855
    },
    {
      "entropy": 5.911665296554565,
      "epoch": 4.812915775141024,
      "grad_norm": 1.375,
      "learning_rate": 0.00029106689789233425,
      "loss": 4.9459,
      "mean_token_accuracy": 0.21421818286180497,
      "num_tokens": 107292959.0,
      "step": 61860
    },
    {
      "entropy": 5.923187494277954,
      "epoch": 4.8133048045127405,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002910392529363566,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21480103731155395,
      "num_tokens": 107301936.0,
      "step": 61865
    },
    {
      "entropy": 5.966071462631225,
      "epoch": 4.813693833884458,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029101160773700887,
      "loss": 4.977,
      "mean_token_accuracy": 0.21877652257680893,
      "num_tokens": 107310996.0,
      "step": 61870
    },
    {
      "entropy": 5.933529710769653,
      "epoch": 4.814082863256176,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002909839622947104,
      "loss": 4.8599,
      "mean_token_accuracy": 0.22365160584449767,
      "num_tokens": 107319291.0,
      "step": 61875
    },
    {
      "entropy": 6.023220586776733,
      "epoch": 4.8144718926278935,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002909563166098809,
      "loss": 5.0791,
      "mean_token_accuracy": 0.1995217964053154,
      "num_tokens": 107328169.0,
      "step": 61880
    },
    {
      "entropy": 5.881319761276245,
      "epoch": 4.814860921999611,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029092867068293966,
      "loss": 4.9303,
      "mean_token_accuracy": 0.21597301065921784,
      "num_tokens": 107336992.0,
      "step": 61885
    },
    {
      "entropy": 5.920500898361206,
      "epoch": 4.815249951371328,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002909010245143061,
      "loss": 4.959,
      "mean_token_accuracy": 0.21946073472499847,
      "num_tokens": 107345292.0,
      "step": 61890
    },
    {
      "entropy": 5.871447992324829,
      "epoch": 4.815638980743046,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002908733781043998,
      "loss": 4.9281,
      "mean_token_accuracy": 0.22117433995008468,
      "num_tokens": 107353567.0,
      "step": 61895
    },
    {
      "entropy": 5.970368528366089,
      "epoch": 4.8160280101147634,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002908457314536403,
      "loss": 4.9977,
      "mean_token_accuracy": 0.21277628391981124,
      "num_tokens": 107361675.0,
      "step": 61900
    },
    {
      "entropy": 5.940203905105591,
      "epoch": 4.816417039486481,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00029081808456244703,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21450794637203216,
      "num_tokens": 107370198.0,
      "step": 61905
    },
    {
      "entropy": 5.951638317108154,
      "epoch": 4.816806068858199,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029079043743123953,
      "loss": 4.9851,
      "mean_token_accuracy": 0.21339751034975052,
      "num_tokens": 107379492.0,
      "step": 61910
    },
    {
      "entropy": 5.938049936294556,
      "epoch": 4.8171950982299165,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029076279006043723,
      "loss": 4.9214,
      "mean_token_accuracy": 0.22354572117328644,
      "num_tokens": 107388690.0,
      "step": 61915
    },
    {
      "entropy": 6.004018115997314,
      "epoch": 4.817584127601634,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002907351424504597,
      "loss": 5.0726,
      "mean_token_accuracy": 0.20593581199645997,
      "num_tokens": 107397158.0,
      "step": 61920
    },
    {
      "entropy": 5.868466901779175,
      "epoch": 4.817973156973352,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002907074946017265,
      "loss": 4.862,
      "mean_token_accuracy": 0.22442245930433274,
      "num_tokens": 107405832.0,
      "step": 61925
    },
    {
      "entropy": 5.933877563476562,
      "epoch": 4.818362186345069,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000290679846514657,
      "loss": 4.9592,
      "mean_token_accuracy": 0.21285081654787064,
      "num_tokens": 107414446.0,
      "step": 61930
    },
    {
      "entropy": 5.979132986068725,
      "epoch": 4.818751215716786,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029065219818967075,
      "loss": 4.9841,
      "mean_token_accuracy": 0.2191564619541168,
      "num_tokens": 107423402.0,
      "step": 61935
    },
    {
      "entropy": 5.995204019546509,
      "epoch": 4.819140245088504,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029062454962718736,
      "loss": 4.9744,
      "mean_token_accuracy": 0.21284471750259398,
      "num_tokens": 107432415.0,
      "step": 61940
    },
    {
      "entropy": 5.944805669784546,
      "epoch": 4.819529274460222,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002905969008276263,
      "loss": 4.9468,
      "mean_token_accuracy": 0.21257275342941284,
      "num_tokens": 107441170.0,
      "step": 61945
    },
    {
      "entropy": 6.004467630386353,
      "epoch": 4.8199183038319395,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002905692517914071,
      "loss": 5.0454,
      "mean_token_accuracy": 0.21161332130432128,
      "num_tokens": 107449523.0,
      "step": 61950
    },
    {
      "entropy": 5.945244359970093,
      "epoch": 4.820307333203657,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00029054160251894924,
      "loss": 4.9633,
      "mean_token_accuracy": 0.22114895284175873,
      "num_tokens": 107458629.0,
      "step": 61955
    },
    {
      "entropy": 5.966711235046387,
      "epoch": 4.820696362575374,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002905139530106724,
      "loss": 4.9803,
      "mean_token_accuracy": 0.2118236929178238,
      "num_tokens": 107467016.0,
      "step": 61960
    },
    {
      "entropy": 5.967395305633545,
      "epoch": 4.821085391947092,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00029048630326699594,
      "loss": 4.9067,
      "mean_token_accuracy": 0.21574630886316298,
      "num_tokens": 107475128.0,
      "step": 61965
    },
    {
      "entropy": 5.983470010757446,
      "epoch": 4.821474421318809,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029045865328833944,
      "loss": 5.0177,
      "mean_token_accuracy": 0.20763752460479737,
      "num_tokens": 107484344.0,
      "step": 61970
    },
    {
      "entropy": 5.8396485328674315,
      "epoch": 4.821863450690527,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00029043100307512247,
      "loss": 4.8813,
      "mean_token_accuracy": 0.22760993540287017,
      "num_tokens": 107493252.0,
      "step": 61975
    },
    {
      "entropy": 6.00921139717102,
      "epoch": 4.822252480062245,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029040335262776467,
      "loss": 5.0445,
      "mean_token_accuracy": 0.20700038075447083,
      "num_tokens": 107502335.0,
      "step": 61980
    },
    {
      "entropy": 6.002280282974243,
      "epoch": 4.8226415094339625,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002903757019466855,
      "loss": 4.9865,
      "mean_token_accuracy": 0.2076638326048851,
      "num_tokens": 107510532.0,
      "step": 61985
    },
    {
      "entropy": 5.989982843399048,
      "epoch": 4.82303053880568,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029034805103230453,
      "loss": 4.9744,
      "mean_token_accuracy": 0.22255508452653885,
      "num_tokens": 107518628.0,
      "step": 61990
    },
    {
      "entropy": 5.996124029159546,
      "epoch": 4.823419568177397,
      "grad_norm": 1.515625,
      "learning_rate": 0.00029032039988504126,
      "loss": 5.0435,
      "mean_token_accuracy": 0.20888043195009232,
      "num_tokens": 107527213.0,
      "step": 61995
    },
    {
      "entropy": 5.99093599319458,
      "epoch": 4.823808597549115,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00029029274850531534,
      "loss": 5.0301,
      "mean_token_accuracy": 0.2125764012336731,
      "num_tokens": 107536382.0,
      "step": 62000
    },
    {
      "entropy": 5.8976154804229735,
      "epoch": 4.824197626920832,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029026509689354635,
      "loss": 4.9352,
      "mean_token_accuracy": 0.22133806794881822,
      "num_tokens": 107545419.0,
      "step": 62005
    },
    {
      "entropy": 5.946125268936157,
      "epoch": 4.82458665629255,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029023744505015373,
      "loss": 4.975,
      "mean_token_accuracy": 0.2105588749051094,
      "num_tokens": 107554458.0,
      "step": 62010
    },
    {
      "entropy": 6.007529258728027,
      "epoch": 4.824975685664268,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029020979297555717,
      "loss": 5.021,
      "mean_token_accuracy": 0.20661982893943787,
      "num_tokens": 107563121.0,
      "step": 62015
    },
    {
      "entropy": 5.977300930023193,
      "epoch": 4.8253647150359855,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029018214067017624,
      "loss": 4.9726,
      "mean_token_accuracy": 0.21709932535886764,
      "num_tokens": 107571380.0,
      "step": 62020
    },
    {
      "entropy": 5.9770989418029785,
      "epoch": 4.825753744407702,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002901544881344305,
      "loss": 4.9627,
      "mean_token_accuracy": 0.21084933876991271,
      "num_tokens": 107580524.0,
      "step": 62025
    },
    {
      "entropy": 5.979213523864746,
      "epoch": 4.82614277377942,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002901268353687394,
      "loss": 5.029,
      "mean_token_accuracy": 0.2114427998661995,
      "num_tokens": 107589492.0,
      "step": 62030
    },
    {
      "entropy": 5.9769768714904785,
      "epoch": 4.826531803151138,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029009918237352274,
      "loss": 5.0003,
      "mean_token_accuracy": 0.20840680003166198,
      "num_tokens": 107598258.0,
      "step": 62035
    },
    {
      "entropy": 5.922567319869995,
      "epoch": 4.826920832522855,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002900715291492,
      "loss": 4.9139,
      "mean_token_accuracy": 0.22051929384469987,
      "num_tokens": 107606526.0,
      "step": 62040
    },
    {
      "entropy": 5.979802036285401,
      "epoch": 4.827309861894573,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002900438756961908,
      "loss": 5.001,
      "mean_token_accuracy": 0.20780059397220613,
      "num_tokens": 107615321.0,
      "step": 62045
    },
    {
      "entropy": 5.959657478332519,
      "epoch": 4.827698891266291,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002900162220149148,
      "loss": 5.0016,
      "mean_token_accuracy": 0.21107443571090698,
      "num_tokens": 107625295.0,
      "step": 62050
    },
    {
      "entropy": 5.9835145473480225,
      "epoch": 4.8280879206380085,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028998856810579144,
      "loss": 5.0281,
      "mean_token_accuracy": 0.21203786879777908,
      "num_tokens": 107634453.0,
      "step": 62055
    },
    {
      "entropy": 6.005669164657593,
      "epoch": 4.828476950009726,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002899609139692405,
      "loss": 4.9424,
      "mean_token_accuracy": 0.21608620136976242,
      "num_tokens": 107642987.0,
      "step": 62060
    },
    {
      "entropy": 6.065689468383789,
      "epoch": 4.828865979381443,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028993325960568133,
      "loss": 5.0876,
      "mean_token_accuracy": 0.21227670162916185,
      "num_tokens": 107651614.0,
      "step": 62065
    },
    {
      "entropy": 5.97391905784607,
      "epoch": 4.829255008753161,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002899056050155339,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22286701202392578,
      "num_tokens": 107660367.0,
      "step": 62070
    },
    {
      "entropy": 5.94419903755188,
      "epoch": 4.829644038124878,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002898779501992176,
      "loss": 5.0187,
      "mean_token_accuracy": 0.21665442436933519,
      "num_tokens": 107668763.0,
      "step": 62075
    },
    {
      "entropy": 5.968116283416748,
      "epoch": 4.830033067496596,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000289850295157152,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20946889519691467,
      "num_tokens": 107677846.0,
      "step": 62080
    },
    {
      "entropy": 5.956566047668457,
      "epoch": 4.830422096868314,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028982263988975687,
      "loss": 4.9048,
      "mean_token_accuracy": 0.2181496575474739,
      "num_tokens": 107687049.0,
      "step": 62085
    },
    {
      "entropy": 5.938007068634033,
      "epoch": 4.8308111262400315,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002897949843974518,
      "loss": 4.9777,
      "mean_token_accuracy": 0.21508435606956483,
      "num_tokens": 107695487.0,
      "step": 62090
    },
    {
      "entropy": 5.963528394699097,
      "epoch": 4.831200155611748,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002897673286806562,
      "loss": 4.969,
      "mean_token_accuracy": 0.21040670722723007,
      "num_tokens": 107703814.0,
      "step": 62095
    },
    {
      "entropy": 6.019762992858887,
      "epoch": 4.831589184983466,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002897396727397901,
      "loss": 5.0015,
      "mean_token_accuracy": 0.21294335126876832,
      "num_tokens": 107712402.0,
      "step": 62100
    },
    {
      "entropy": 5.979795169830322,
      "epoch": 4.831978214355184,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002897120165752729,
      "loss": 4.9942,
      "mean_token_accuracy": 0.20664868205785752,
      "num_tokens": 107720863.0,
      "step": 62105
    },
    {
      "entropy": 5.956272983551026,
      "epoch": 4.832367243726901,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002896843601875243,
      "loss": 5.0308,
      "mean_token_accuracy": 0.2085858851671219,
      "num_tokens": 107730079.0,
      "step": 62110
    },
    {
      "entropy": 5.959002542495727,
      "epoch": 4.832756273098619,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028965670357696384,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21357775777578353,
      "num_tokens": 107738328.0,
      "step": 62115
    },
    {
      "entropy": 5.886544322967529,
      "epoch": 4.833145302470337,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028962904674401124,
      "loss": 4.8679,
      "mean_token_accuracy": 0.2238164708018303,
      "num_tokens": 107747159.0,
      "step": 62120
    },
    {
      "entropy": 5.887013483047485,
      "epoch": 4.8335343318420545,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002896013896890862,
      "loss": 5.0186,
      "mean_token_accuracy": 0.20866921544075012,
      "num_tokens": 107756282.0,
      "step": 62125
    },
    {
      "entropy": 5.933960866928101,
      "epoch": 4.833923361213771,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028957373241260816,
      "loss": 4.88,
      "mean_token_accuracy": 0.22534471601247788,
      "num_tokens": 107764967.0,
      "step": 62130
    },
    {
      "entropy": 5.982131052017212,
      "epoch": 4.834312390585489,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000289546074914997,
      "loss": 4.9792,
      "mean_token_accuracy": 0.2151278018951416,
      "num_tokens": 107773544.0,
      "step": 62135
    },
    {
      "entropy": 5.849381303787231,
      "epoch": 4.834701419957207,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002895184171966724,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22136432081460952,
      "num_tokens": 107781996.0,
      "step": 62140
    },
    {
      "entropy": 5.962338924407959,
      "epoch": 4.835090449328924,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002894907592580538,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21616612374782562,
      "num_tokens": 107790708.0,
      "step": 62145
    },
    {
      "entropy": 5.987463045120239,
      "epoch": 4.835479478700642,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028946310109956107,
      "loss": 4.9212,
      "mean_token_accuracy": 0.22406015694141387,
      "num_tokens": 107798829.0,
      "step": 62150
    },
    {
      "entropy": 5.958460807800293,
      "epoch": 4.83586850807236,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002894354427216138,
      "loss": 5.0071,
      "mean_token_accuracy": 0.21203849017620086,
      "num_tokens": 107808498.0,
      "step": 62155
    },
    {
      "entropy": 6.004882860183716,
      "epoch": 4.836257537444077,
      "grad_norm": 1.65625,
      "learning_rate": 0.00028940778412463164,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21032360345125198,
      "num_tokens": 107815831.0,
      "step": 62160
    },
    {
      "entropy": 5.9539214134216305,
      "epoch": 4.836646566815794,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028938012530903425,
      "loss": 4.9942,
      "mean_token_accuracy": 0.20935427248477936,
      "num_tokens": 107825258.0,
      "step": 62165
    },
    {
      "entropy": 5.884432411193847,
      "epoch": 4.837035596187512,
      "grad_norm": 1.375,
      "learning_rate": 0.0002893524662752414,
      "loss": 4.8889,
      "mean_token_accuracy": 0.21961808949708939,
      "num_tokens": 107833725.0,
      "step": 62170
    },
    {
      "entropy": 5.88515944480896,
      "epoch": 4.83742462555923,
      "grad_norm": 1.46875,
      "learning_rate": 0.00028932480702367264,
      "loss": 4.9014,
      "mean_token_accuracy": 0.21450790911912918,
      "num_tokens": 107841906.0,
      "step": 62175
    },
    {
      "entropy": 5.951255464553833,
      "epoch": 4.837813654930947,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002892971475547478,
      "loss": 5.0282,
      "mean_token_accuracy": 0.2220454692840576,
      "num_tokens": 107850629.0,
      "step": 62180
    },
    {
      "entropy": 6.044350624084473,
      "epoch": 4.838202684302665,
      "grad_norm": 1.46875,
      "learning_rate": 0.00028926948786888655,
      "loss": 5.0734,
      "mean_token_accuracy": 0.20971916615962982,
      "num_tokens": 107859829.0,
      "step": 62185
    },
    {
      "entropy": 5.98042311668396,
      "epoch": 4.838591713674383,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028924182796650854,
      "loss": 4.9722,
      "mean_token_accuracy": 0.21009718924760817,
      "num_tokens": 107867859.0,
      "step": 62190
    },
    {
      "entropy": 5.8776692867279055,
      "epoch": 4.8389807430461,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028921416784803336,
      "loss": 4.8555,
      "mean_token_accuracy": 0.21767117232084274,
      "num_tokens": 107876460.0,
      "step": 62195
    },
    {
      "entropy": 5.919812726974487,
      "epoch": 4.839369772417817,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002891865075138809,
      "loss": 4.948,
      "mean_token_accuracy": 0.21607521325349807,
      "num_tokens": 107884904.0,
      "step": 62200
    },
    {
      "entropy": 5.856677103042602,
      "epoch": 4.839758801789535,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00028915884696447064,
      "loss": 4.8787,
      "mean_token_accuracy": 0.22695459127426149,
      "num_tokens": 107893397.0,
      "step": 62205
    },
    {
      "entropy": 5.998841667175293,
      "epoch": 4.840147831161253,
      "grad_norm": 1.5,
      "learning_rate": 0.0002891311862002225,
      "loss": 5.0509,
      "mean_token_accuracy": 0.21474215984344483,
      "num_tokens": 107901736.0,
      "step": 62210
    },
    {
      "entropy": 5.9551084518432615,
      "epoch": 4.84053686053297,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002891035252215561,
      "loss": 4.9274,
      "mean_token_accuracy": 0.21660346984863282,
      "num_tokens": 107910370.0,
      "step": 62215
    },
    {
      "entropy": 5.921539545059204,
      "epoch": 4.840925889904688,
      "grad_norm": 1.5,
      "learning_rate": 0.00028907586402889115,
      "loss": 4.9692,
      "mean_token_accuracy": 0.21637522131204606,
      "num_tokens": 107918732.0,
      "step": 62220
    },
    {
      "entropy": 5.9289031505584715,
      "epoch": 4.841314919276405,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002890482026226473,
      "loss": 4.9406,
      "mean_token_accuracy": 0.219663405418396,
      "num_tokens": 107927854.0,
      "step": 62225
    },
    {
      "entropy": 5.915649557113648,
      "epoch": 4.841703948648123,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028902054100324445,
      "loss": 4.8708,
      "mean_token_accuracy": 0.22470742017030715,
      "num_tokens": 107937144.0,
      "step": 62230
    },
    {
      "entropy": 5.980217599868775,
      "epoch": 4.84209297801984,
      "grad_norm": 1.46875,
      "learning_rate": 0.00028899287917110214,
      "loss": 5.0161,
      "mean_token_accuracy": 0.20845099538564682,
      "num_tokens": 107945590.0,
      "step": 62235
    },
    {
      "entropy": 5.930459594726562,
      "epoch": 4.842482007391558,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002889652171266402,
      "loss": 4.9478,
      "mean_token_accuracy": 0.2215537741780281,
      "num_tokens": 107954191.0,
      "step": 62240
    },
    {
      "entropy": 5.997185564041137,
      "epoch": 4.842871036763276,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028893755487027827,
      "loss": 4.9618,
      "mean_token_accuracy": 0.21665312796831132,
      "num_tokens": 107963530.0,
      "step": 62245
    },
    {
      "entropy": 6.032745742797852,
      "epoch": 4.843260066134993,
      "grad_norm": 1.5,
      "learning_rate": 0.00028890989240243614,
      "loss": 5.066,
      "mean_token_accuracy": 0.20527476221323013,
      "num_tokens": 107971540.0,
      "step": 62250
    },
    {
      "entropy": 5.887638568878174,
      "epoch": 4.843649095506711,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028888222972353356,
      "loss": 4.9055,
      "mean_token_accuracy": 0.2207385316491127,
      "num_tokens": 107980013.0,
      "step": 62255
    },
    {
      "entropy": 5.956586122512817,
      "epoch": 4.844038124878429,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028885456683399024,
      "loss": 5.0035,
      "mean_token_accuracy": 0.21161157488822938,
      "num_tokens": 107989061.0,
      "step": 62260
    },
    {
      "entropy": 5.868087339401245,
      "epoch": 4.844427154250146,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028882690373422586,
      "loss": 4.8815,
      "mean_token_accuracy": 0.21387384980916976,
      "num_tokens": 107997491.0,
      "step": 62265
    },
    {
      "entropy": 5.954627609252929,
      "epoch": 4.844816183621863,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002887992404246603,
      "loss": 4.9491,
      "mean_token_accuracy": 0.21530840545892715,
      "num_tokens": 108006564.0,
      "step": 62270
    },
    {
      "entropy": 5.919207620620727,
      "epoch": 4.845205212993581,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002887715769057132,
      "loss": 4.909,
      "mean_token_accuracy": 0.22007119804620742,
      "num_tokens": 108015355.0,
      "step": 62275
    },
    {
      "entropy": 5.925663852691651,
      "epoch": 4.845594242365299,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028874391317780437,
      "loss": 4.9719,
      "mean_token_accuracy": 0.2168294295668602,
      "num_tokens": 108023786.0,
      "step": 62280
    },
    {
      "entropy": 5.8906255722045895,
      "epoch": 4.845983271737016,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002887162492413534,
      "loss": 4.8084,
      "mean_token_accuracy": 0.2257682979106903,
      "num_tokens": 108031919.0,
      "step": 62285
    },
    {
      "entropy": 5.9154726505279545,
      "epoch": 4.846372301108734,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002886885850967803,
      "loss": 4.9052,
      "mean_token_accuracy": 0.22301119565963745,
      "num_tokens": 108040572.0,
      "step": 62290
    },
    {
      "entropy": 5.998310375213623,
      "epoch": 4.846761330480451,
      "grad_norm": 1.5,
      "learning_rate": 0.00028866092074450467,
      "loss": 5.0143,
      "mean_token_accuracy": 0.21536397337913513,
      "num_tokens": 108048293.0,
      "step": 62295
    },
    {
      "entropy": 6.007065486907959,
      "epoch": 4.847150359852169,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002886332561849464,
      "loss": 5.0765,
      "mean_token_accuracy": 0.20775685906410218,
      "num_tokens": 108056862.0,
      "step": 62300
    },
    {
      "entropy": 6.010410976409912,
      "epoch": 4.847539389223886,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028860559141852504,
      "loss": 5.0429,
      "mean_token_accuracy": 0.21159214079380034,
      "num_tokens": 108065466.0,
      "step": 62305
    },
    {
      "entropy": 5.963889312744141,
      "epoch": 4.847928418595604,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028857792644566057,
      "loss": 4.8511,
      "mean_token_accuracy": 0.21954134851694107,
      "num_tokens": 108073558.0,
      "step": 62310
    },
    {
      "entropy": 5.833923244476319,
      "epoch": 4.848317447967322,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002885502612667727,
      "loss": 4.8274,
      "mean_token_accuracy": 0.22944421023130418,
      "num_tokens": 108083181.0,
      "step": 62315
    },
    {
      "entropy": 5.818351554870605,
      "epoch": 4.848706477339039,
      "grad_norm": 1.46875,
      "learning_rate": 0.00028852259588228106,
      "loss": 4.8429,
      "mean_token_accuracy": 0.22114098370075225,
      "num_tokens": 108091643.0,
      "step": 62320
    },
    {
      "entropy": 5.923635339736938,
      "epoch": 4.849095506710757,
      "grad_norm": 1.546875,
      "learning_rate": 0.00028849493029260573,
      "loss": 4.866,
      "mean_token_accuracy": 0.23055962920188905,
      "num_tokens": 108099815.0,
      "step": 62325
    },
    {
      "entropy": 5.930892419815064,
      "epoch": 4.849484536082474,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002884672644981661,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21286232322454451,
      "num_tokens": 108108360.0,
      "step": 62330
    },
    {
      "entropy": 5.965675115585327,
      "epoch": 4.849873565454192,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002884395984993823,
      "loss": 4.94,
      "mean_token_accuracy": 0.2166549175977707,
      "num_tokens": 108116368.0,
      "step": 62335
    },
    {
      "entropy": 5.917623138427734,
      "epoch": 4.850262594825909,
      "grad_norm": 1.515625,
      "learning_rate": 0.000288411932296674,
      "loss": 4.9039,
      "mean_token_accuracy": 0.21243095993995667,
      "num_tokens": 108124997.0,
      "step": 62340
    },
    {
      "entropy": 5.954190254211426,
      "epoch": 4.850651624197627,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002883842658904609,
      "loss": 4.9669,
      "mean_token_accuracy": 0.21762391477823256,
      "num_tokens": 108133440.0,
      "step": 62345
    },
    {
      "entropy": 5.957757091522216,
      "epoch": 4.851040653569345,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002883565992811629,
      "loss": 4.9507,
      "mean_token_accuracy": 0.21837241053581238,
      "num_tokens": 108142730.0,
      "step": 62350
    },
    {
      "entropy": 5.965146589279175,
      "epoch": 4.851429682941062,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002883289324691998,
      "loss": 4.961,
      "mean_token_accuracy": 0.21059280037879943,
      "num_tokens": 108151470.0,
      "step": 62355
    },
    {
      "entropy": 5.878978061676025,
      "epoch": 4.851818712312779,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002883012654549913,
      "loss": 4.8867,
      "mean_token_accuracy": 0.22210302203893661,
      "num_tokens": 108160165.0,
      "step": 62360
    },
    {
      "entropy": 5.910700178146362,
      "epoch": 4.852207741684497,
      "grad_norm": 1.546875,
      "learning_rate": 0.00028827359823895726,
      "loss": 4.9231,
      "mean_token_accuracy": 0.2111850008368492,
      "num_tokens": 108168124.0,
      "step": 62365
    },
    {
      "entropy": 6.002094268798828,
      "epoch": 4.852596771056215,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028824593082151754,
      "loss": 5.0707,
      "mean_token_accuracy": 0.2012113645672798,
      "num_tokens": 108176579.0,
      "step": 62370
    },
    {
      "entropy": 5.949891090393066,
      "epoch": 4.852985800427932,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002882182632030919,
      "loss": 4.884,
      "mean_token_accuracy": 0.22614724040031434,
      "num_tokens": 108184638.0,
      "step": 62375
    },
    {
      "entropy": 5.966904163360596,
      "epoch": 4.85337482979965,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002881905953841002,
      "loss": 4.9248,
      "mean_token_accuracy": 0.2189707487821579,
      "num_tokens": 108192751.0,
      "step": 62380
    },
    {
      "entropy": 5.952373743057251,
      "epoch": 4.853763859171368,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002881629273649621,
      "loss": 4.9683,
      "mean_token_accuracy": 0.2153223693370819,
      "num_tokens": 108201219.0,
      "step": 62385
    },
    {
      "entropy": 5.95755352973938,
      "epoch": 4.854152888543085,
      "grad_norm": 1.546875,
      "learning_rate": 0.00028813525914609767,
      "loss": 5.0831,
      "mean_token_accuracy": 0.20762702077627182,
      "num_tokens": 108210184.0,
      "step": 62390
    },
    {
      "entropy": 5.9866392612457275,
      "epoch": 4.854541917914803,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028810759072792644,
      "loss": 4.9893,
      "mean_token_accuracy": 0.21291426420211793,
      "num_tokens": 108218549.0,
      "step": 62395
    },
    {
      "entropy": 6.039851951599121,
      "epoch": 4.85493094728652,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002880799221108684,
      "loss": 5.037,
      "mean_token_accuracy": 0.21474067121744156,
      "num_tokens": 108226608.0,
      "step": 62400
    },
    {
      "entropy": 5.9445085525512695,
      "epoch": 4.855319976658238,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002880522532953435,
      "loss": 4.9754,
      "mean_token_accuracy": 0.21942467838525773,
      "num_tokens": 108234384.0,
      "step": 62405
    },
    {
      "entropy": 5.980475950241089,
      "epoch": 4.855709006029955,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002880245842817713,
      "loss": 5.0181,
      "mean_token_accuracy": 0.21448771357536317,
      "num_tokens": 108243534.0,
      "step": 62410
    },
    {
      "entropy": 5.9517323017120365,
      "epoch": 4.856098035401673,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002879969150705718,
      "loss": 5.0141,
      "mean_token_accuracy": 0.2178867295384407,
      "num_tokens": 108252658.0,
      "step": 62415
    },
    {
      "entropy": 5.948013877868652,
      "epoch": 4.856487064773391,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002879692456621649,
      "loss": 4.8856,
      "mean_token_accuracy": 0.22678062617778777,
      "num_tokens": 108261464.0,
      "step": 62420
    },
    {
      "entropy": 5.898688745498657,
      "epoch": 4.8568760941451075,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002879415760569702,
      "loss": 4.9158,
      "mean_token_accuracy": 0.22375869750976562,
      "num_tokens": 108269936.0,
      "step": 62425
    },
    {
      "entropy": 5.9854659080505375,
      "epoch": 4.857265123516825,
      "grad_norm": 1.5,
      "learning_rate": 0.0002879139062554077,
      "loss": 5.0327,
      "mean_token_accuracy": 0.20757278949022293,
      "num_tokens": 108279112.0,
      "step": 62430
    },
    {
      "entropy": 6.002867364883423,
      "epoch": 4.857654152888543,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00028788623625789725,
      "loss": 4.9752,
      "mean_token_accuracy": 0.20916147381067277,
      "num_tokens": 108287205.0,
      "step": 62435
    },
    {
      "entropy": 6.021507787704468,
      "epoch": 4.8580431822602606,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002878585660648587,
      "loss": 5.0096,
      "mean_token_accuracy": 0.20732629895210267,
      "num_tokens": 108296118.0,
      "step": 62440
    },
    {
      "entropy": 5.995490312576294,
      "epoch": 4.858432211631978,
      "grad_norm": 1.375,
      "learning_rate": 0.0002878308956767118,
      "loss": 5.014,
      "mean_token_accuracy": 0.20312580317258835,
      "num_tokens": 108306012.0,
      "step": 62445
    },
    {
      "entropy": 5.963175821304321,
      "epoch": 4.858821241003696,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002878032250938766,
      "loss": 4.9777,
      "mean_token_accuracy": 0.21770790815353394,
      "num_tokens": 108315359.0,
      "step": 62450
    },
    {
      "entropy": 5.945657253265381,
      "epoch": 4.859210270375414,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00028777555431677276,
      "loss": 4.9137,
      "mean_token_accuracy": 0.2229872465133667,
      "num_tokens": 108323093.0,
      "step": 62455
    },
    {
      "entropy": 5.867759752273559,
      "epoch": 4.859599299747131,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028774788334582025,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22571109235286713,
      "num_tokens": 108332556.0,
      "step": 62460
    },
    {
      "entropy": 5.920924091339112,
      "epoch": 4.859988329118848,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002877202121814389,
      "loss": 4.957,
      "mean_token_accuracy": 0.20899031907320023,
      "num_tokens": 108341794.0,
      "step": 62465
    },
    {
      "entropy": 6.068184804916382,
      "epoch": 4.860377358490566,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002876925408240485,
      "loss": 5.0418,
      "mean_token_accuracy": 0.20976990312337876,
      "num_tokens": 108349772.0,
      "step": 62470
    },
    {
      "entropy": 5.9050750732421875,
      "epoch": 4.8607663878622835,
      "grad_norm": 1.375,
      "learning_rate": 0.0002876648692740691,
      "loss": 4.8507,
      "mean_token_accuracy": 0.2298590362071991,
      "num_tokens": 108358406.0,
      "step": 62475
    },
    {
      "entropy": 5.983275127410889,
      "epoch": 4.861155417234001,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028763719753192044,
      "loss": 5.0652,
      "mean_token_accuracy": 0.20897759050130843,
      "num_tokens": 108368203.0,
      "step": 62480
    },
    {
      "entropy": 5.9673772811889645,
      "epoch": 4.861544446605719,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002876095255980224,
      "loss": 4.998,
      "mean_token_accuracy": 0.21131344884634018,
      "num_tokens": 108377482.0,
      "step": 62485
    },
    {
      "entropy": 5.870056390762329,
      "epoch": 4.861933475977437,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00028758185347279485,
      "loss": 4.8055,
      "mean_token_accuracy": 0.23057566583156586,
      "num_tokens": 108385993.0,
      "step": 62490
    },
    {
      "entropy": 5.9704900741577145,
      "epoch": 4.8623225053491534,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002875541811566578,
      "loss": 5.0557,
      "mean_token_accuracy": 0.2104988396167755,
      "num_tokens": 108394143.0,
      "step": 62495
    },
    {
      "entropy": 5.959816598892212,
      "epoch": 4.862711534720871,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002875265086500309,
      "loss": 4.937,
      "mean_token_accuracy": 0.22604523599147797,
      "num_tokens": 108402531.0,
      "step": 62500
    },
    {
      "entropy": 5.931360149383545,
      "epoch": 4.863100564092589,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028749883595333424,
      "loss": 5.0073,
      "mean_token_accuracy": 0.20964014828205108,
      "num_tokens": 108411884.0,
      "step": 62505
    },
    {
      "entropy": 5.919588470458985,
      "epoch": 4.8634895934643065,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002874711630669876,
      "loss": 5.0328,
      "mean_token_accuracy": 0.20710403323173524,
      "num_tokens": 108421246.0,
      "step": 62510
    },
    {
      "entropy": 5.920648097991943,
      "epoch": 4.863878622836024,
      "grad_norm": 1.421875,
      "learning_rate": 0.000287443489991411,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22473251074552536,
      "num_tokens": 108429858.0,
      "step": 62515
    },
    {
      "entropy": 6.0476991653442385,
      "epoch": 4.864267652207742,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002874158167270242,
      "loss": 5.0342,
      "mean_token_accuracy": 0.2124841332435608,
      "num_tokens": 108438647.0,
      "step": 62520
    },
    {
      "entropy": 6.004568147659302,
      "epoch": 4.86465668157946,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028738814327424714,
      "loss": 4.9433,
      "mean_token_accuracy": 0.21430870294570922,
      "num_tokens": 108446695.0,
      "step": 62525
    },
    {
      "entropy": 5.937493085861206,
      "epoch": 4.865045710951176,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002873604696334997,
      "loss": 4.9047,
      "mean_token_accuracy": 0.21397738456726073,
      "num_tokens": 108455299.0,
      "step": 62530
    },
    {
      "entropy": 5.966624212265015,
      "epoch": 4.865434740322894,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028733279580520185,
      "loss": 5.0051,
      "mean_token_accuracy": 0.21181025952100754,
      "num_tokens": 108464227.0,
      "step": 62535
    },
    {
      "entropy": 5.926171588897705,
      "epoch": 4.865823769694612,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028730512178977343,
      "loss": 4.9295,
      "mean_token_accuracy": 0.2216956064105034,
      "num_tokens": 108473806.0,
      "step": 62540
    },
    {
      "entropy": 5.970411252975464,
      "epoch": 4.8662127990663295,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028727744758763434,
      "loss": 5.027,
      "mean_token_accuracy": 0.214263716340065,
      "num_tokens": 108482291.0,
      "step": 62545
    },
    {
      "entropy": 5.931823301315307,
      "epoch": 4.866601828438047,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028724977319920453,
      "loss": 4.9864,
      "mean_token_accuracy": 0.21358804553747177,
      "num_tokens": 108490662.0,
      "step": 62550
    },
    {
      "entropy": 5.895310640335083,
      "epoch": 4.866990857809765,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028722209862490393,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21827817857265472,
      "num_tokens": 108500005.0,
      "step": 62555
    },
    {
      "entropy": 5.888722085952759,
      "epoch": 4.867379887181482,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028719442386515245,
      "loss": 4.8908,
      "mean_token_accuracy": 0.22265540808439255,
      "num_tokens": 108509226.0,
      "step": 62560
    },
    {
      "entropy": 5.856849479675293,
      "epoch": 4.867768916553199,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028716674892036994,
      "loss": 4.8679,
      "mean_token_accuracy": 0.22694025933742523,
      "num_tokens": 108518417.0,
      "step": 62565
    },
    {
      "entropy": 5.949336576461792,
      "epoch": 4.868157945924917,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002871390737909765,
      "loss": 4.9557,
      "mean_token_accuracy": 0.2165167361497879,
      "num_tokens": 108527603.0,
      "step": 62570
    },
    {
      "entropy": 5.95817084312439,
      "epoch": 4.868546975296635,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028711139847739187,
      "loss": 4.9871,
      "mean_token_accuracy": 0.21476602405309678,
      "num_tokens": 108535908.0,
      "step": 62575
    },
    {
      "entropy": 5.945937252044677,
      "epoch": 4.8689360046683525,
      "grad_norm": 1.46875,
      "learning_rate": 0.00028708372298003607,
      "loss": 4.9324,
      "mean_token_accuracy": 0.2162817656993866,
      "num_tokens": 108544250.0,
      "step": 62580
    },
    {
      "entropy": 5.926062059402466,
      "epoch": 4.86932503404007,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00028705604729932895,
      "loss": 4.9264,
      "mean_token_accuracy": 0.21510962694883345,
      "num_tokens": 108553053.0,
      "step": 62585
    },
    {
      "entropy": 5.9675208568573,
      "epoch": 4.869714063411788,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028702837143569057,
      "loss": 4.9551,
      "mean_token_accuracy": 0.21661992371082306,
      "num_tokens": 108561686.0,
      "step": 62590
    },
    {
      "entropy": 5.860761451721191,
      "epoch": 4.870103092783506,
      "grad_norm": 1.515625,
      "learning_rate": 0.00028700069538954064,
      "loss": 4.8524,
      "mean_token_accuracy": 0.22031694799661636,
      "num_tokens": 108570219.0,
      "step": 62595
    },
    {
      "entropy": 5.982944488525391,
      "epoch": 4.870492122155222,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00028697301916129944,
      "loss": 4.9912,
      "mean_token_accuracy": 0.2181403696537018,
      "num_tokens": 108579056.0,
      "step": 62600
    },
    {
      "entropy": 5.947331523895263,
      "epoch": 4.87088115152694,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002869453427513866,
      "loss": 4.895,
      "mean_token_accuracy": 0.22354386895895004,
      "num_tokens": 108587486.0,
      "step": 62605
    },
    {
      "entropy": 5.956013250350952,
      "epoch": 4.871270180898658,
      "grad_norm": 1.5,
      "learning_rate": 0.00028691766616022234,
      "loss": 4.9729,
      "mean_token_accuracy": 0.2121080458164215,
      "num_tokens": 108595989.0,
      "step": 62610
    },
    {
      "entropy": 5.950033855438233,
      "epoch": 4.8716592102703755,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002868899893882264,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2227548986673355,
      "num_tokens": 108604807.0,
      "step": 62615
    },
    {
      "entropy": 5.97179069519043,
      "epoch": 4.872048239642093,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028686231243581873,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20783150792121888,
      "num_tokens": 108613338.0,
      "step": 62620
    },
    {
      "entropy": 5.883927536010742,
      "epoch": 4.872437269013811,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028683463530341944,
      "loss": 4.8902,
      "mean_token_accuracy": 0.22053857296705245,
      "num_tokens": 108621695.0,
      "step": 62625
    },
    {
      "entropy": 5.896273422241211,
      "epoch": 4.872826298385528,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002868069579914483,
      "loss": 4.9411,
      "mean_token_accuracy": 0.22312623560428618,
      "num_tokens": 108631238.0,
      "step": 62630
    },
    {
      "entropy": 5.895962476730347,
      "epoch": 4.873215327757245,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028677928050032547,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23048595190048218,
      "num_tokens": 108639829.0,
      "step": 62635
    },
    {
      "entropy": 6.011116409301758,
      "epoch": 4.873604357128963,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028675160283047073,
      "loss": 5.0198,
      "mean_token_accuracy": 0.21055786311626434,
      "num_tokens": 108648244.0,
      "step": 62640
    },
    {
      "entropy": 5.994265174865722,
      "epoch": 4.873993386500681,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028672392498230414,
      "loss": 4.9938,
      "mean_token_accuracy": 0.20839695185422896,
      "num_tokens": 108656642.0,
      "step": 62645
    },
    {
      "entropy": 5.992451763153076,
      "epoch": 4.8743824158723985,
      "grad_norm": 1.5625,
      "learning_rate": 0.00028669624695624573,
      "loss": 5.0347,
      "mean_token_accuracy": 0.21017993241548538,
      "num_tokens": 108664720.0,
      "step": 62650
    },
    {
      "entropy": 5.920983409881591,
      "epoch": 4.874771445244116,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028666856875271535,
      "loss": 4.8525,
      "mean_token_accuracy": 0.23205921798944473,
      "num_tokens": 108674145.0,
      "step": 62655
    },
    {
      "entropy": 5.982768106460571,
      "epoch": 4.875160474615834,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002866408903721329,
      "loss": 5.0258,
      "mean_token_accuracy": 0.21012409329414367,
      "num_tokens": 108682914.0,
      "step": 62660
    },
    {
      "entropy": 5.934384965896607,
      "epoch": 4.875549503987551,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00028661321181491854,
      "loss": 4.8957,
      "mean_token_accuracy": 0.22287616580724717,
      "num_tokens": 108691449.0,
      "step": 62665
    },
    {
      "entropy": 5.959488821029663,
      "epoch": 4.875938533359268,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002865855330814922,
      "loss": 5.0099,
      "mean_token_accuracy": 0.2124689757823944,
      "num_tokens": 108699978.0,
      "step": 62670
    },
    {
      "entropy": 5.932666158676147,
      "epoch": 4.876327562730986,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002865578541722738,
      "loss": 4.9884,
      "mean_token_accuracy": 0.2206720605492592,
      "num_tokens": 108707864.0,
      "step": 62675
    },
    {
      "entropy": 5.931645345687866,
      "epoch": 4.876716592102704,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002865301750876834,
      "loss": 4.9714,
      "mean_token_accuracy": 0.224133962392807,
      "num_tokens": 108716781.0,
      "step": 62680
    },
    {
      "entropy": 6.007809162139893,
      "epoch": 4.8771056214744215,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002865024958281409,
      "loss": 5.0464,
      "mean_token_accuracy": 0.21893274337053298,
      "num_tokens": 108725909.0,
      "step": 62685
    },
    {
      "entropy": 5.992783164978027,
      "epoch": 4.877494650846139,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028647481639406644,
      "loss": 4.9941,
      "mean_token_accuracy": 0.21300367563962935,
      "num_tokens": 108735548.0,
      "step": 62690
    },
    {
      "entropy": 5.953064584732056,
      "epoch": 4.877883680217856,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028644713678587983,
      "loss": 4.9242,
      "mean_token_accuracy": 0.2172428086400032,
      "num_tokens": 108744006.0,
      "step": 62695
    },
    {
      "entropy": 5.959431314468384,
      "epoch": 4.878272709589574,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028641945700400104,
      "loss": 4.9515,
      "mean_token_accuracy": 0.21521011292934417,
      "num_tokens": 108752490.0,
      "step": 62700
    },
    {
      "entropy": 5.933838605880737,
      "epoch": 4.878661738961291,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028639177704885033,
      "loss": 4.9974,
      "mean_token_accuracy": 0.218491593003273,
      "num_tokens": 108761685.0,
      "step": 62705
    },
    {
      "entropy": 5.93937315940857,
      "epoch": 4.879050768333009,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028636409692084743,
      "loss": 4.9618,
      "mean_token_accuracy": 0.21762053668498993,
      "num_tokens": 108769880.0,
      "step": 62710
    },
    {
      "entropy": 5.982127857208252,
      "epoch": 4.879439797704727,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002863364166204125,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20779570788145066,
      "num_tokens": 108779018.0,
      "step": 62715
    },
    {
      "entropy": 5.9794846534729,
      "epoch": 4.8798288270764445,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00028630873614796543,
      "loss": 4.9535,
      "mean_token_accuracy": 0.21243578791618348,
      "num_tokens": 108786908.0,
      "step": 62720
    },
    {
      "entropy": 5.946661138534546,
      "epoch": 4.880217856448162,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00028628105550392637,
      "loss": 4.8812,
      "mean_token_accuracy": 0.2158359944820404,
      "num_tokens": 108795156.0,
      "step": 62725
    },
    {
      "entropy": 5.940703582763672,
      "epoch": 4.88060688581988,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002862533746887152,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22232444882392882,
      "num_tokens": 108803175.0,
      "step": 62730
    },
    {
      "entropy": 5.925836372375488,
      "epoch": 4.880995915191597,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028622569370275196,
      "loss": 4.916,
      "mean_token_accuracy": 0.21965106427669526,
      "num_tokens": 108813187.0,
      "step": 62735
    },
    {
      "entropy": 5.966361045837402,
      "epoch": 4.881384944563314,
      "grad_norm": 1.5625,
      "learning_rate": 0.00028619801254645675,
      "loss": 5.0037,
      "mean_token_accuracy": 0.21276121586561203,
      "num_tokens": 108820913.0,
      "step": 62740
    },
    {
      "entropy": 5.9602337837219235,
      "epoch": 4.881773973935032,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002861703312202495,
      "loss": 4.9954,
      "mean_token_accuracy": 0.21729689836502075,
      "num_tokens": 108829927.0,
      "step": 62745
    },
    {
      "entropy": 5.9583131790161135,
      "epoch": 4.88216300330675,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028614264972455023,
      "loss": 5.022,
      "mean_token_accuracy": 0.21362959146499633,
      "num_tokens": 108838556.0,
      "step": 62750
    },
    {
      "entropy": 5.969022464752197,
      "epoch": 4.8825520326784675,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00028611496805977906,
      "loss": 5.031,
      "mean_token_accuracy": 0.20862161368131638,
      "num_tokens": 108846640.0,
      "step": 62755
    },
    {
      "entropy": 5.959452867507935,
      "epoch": 4.882941062050184,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028608728622635585,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22041326016187668,
      "num_tokens": 108855008.0,
      "step": 62760
    },
    {
      "entropy": 6.044925498962402,
      "epoch": 4.883330091421902,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028605960422470073,
      "loss": 5.0686,
      "mean_token_accuracy": 0.21455166041851043,
      "num_tokens": 108863824.0,
      "step": 62765
    },
    {
      "entropy": 6.001237058639527,
      "epoch": 4.88371912079362,
      "grad_norm": 1.5,
      "learning_rate": 0.0002860319220552338,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20919059216976166,
      "num_tokens": 108872365.0,
      "step": 62770
    },
    {
      "entropy": 5.918903350830078,
      "epoch": 4.884108150165337,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028600423971837484,
      "loss": 4.8444,
      "mean_token_accuracy": 0.2256878137588501,
      "num_tokens": 108881024.0,
      "step": 62775
    },
    {
      "entropy": 5.870365905761719,
      "epoch": 4.884497179537055,
      "grad_norm": 1.546875,
      "learning_rate": 0.00028597655721454427,
      "loss": 4.9456,
      "mean_token_accuracy": 0.21443799287080764,
      "num_tokens": 108889049.0,
      "step": 62780
    },
    {
      "entropy": 5.93803768157959,
      "epoch": 4.884886208908773,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028594887454416176,
      "loss": 4.9423,
      "mean_token_accuracy": 0.22198717892169953,
      "num_tokens": 108897819.0,
      "step": 62785
    },
    {
      "entropy": 5.966743803024292,
      "epoch": 4.8852752382804905,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002859211917076476,
      "loss": 4.9534,
      "mean_token_accuracy": 0.22078824192285537,
      "num_tokens": 108906249.0,
      "step": 62790
    },
    {
      "entropy": 5.942706060409546,
      "epoch": 4.885664267652208,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028589350870542165,
      "loss": 5.0175,
      "mean_token_accuracy": 0.21060277223587037,
      "num_tokens": 108914502.0,
      "step": 62795
    },
    {
      "entropy": 5.982306909561157,
      "epoch": 4.886053297023925,
      "grad_norm": 1.5,
      "learning_rate": 0.0002858658255379041,
      "loss": 5.0997,
      "mean_token_accuracy": 0.2100283607840538,
      "num_tokens": 108922945.0,
      "step": 62800
    },
    {
      "entropy": 5.949176597595215,
      "epoch": 4.886442326395643,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028583814220551495,
      "loss": 4.9531,
      "mean_token_accuracy": 0.2168431207537651,
      "num_tokens": 108931920.0,
      "step": 62805
    },
    {
      "entropy": 6.020411968231201,
      "epoch": 4.88683135576736,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002858104587086742,
      "loss": 5.0994,
      "mean_token_accuracy": 0.20849640369415284,
      "num_tokens": 108940529.0,
      "step": 62810
    },
    {
      "entropy": 5.954330825805664,
      "epoch": 4.887220385139078,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000285782775047802,
      "loss": 4.9284,
      "mean_token_accuracy": 0.21844000667333602,
      "num_tokens": 108949623.0,
      "step": 62815
    },
    {
      "entropy": 6.012961721420288,
      "epoch": 4.887609414510796,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002857550912233183,
      "loss": 5.0506,
      "mean_token_accuracy": 0.20945561677217484,
      "num_tokens": 108958324.0,
      "step": 62820
    },
    {
      "entropy": 5.850955867767334,
      "epoch": 4.8879984438825135,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002857274072356432,
      "loss": 4.9204,
      "mean_token_accuracy": 0.22435320168733597,
      "num_tokens": 108967419.0,
      "step": 62825
    },
    {
      "entropy": 5.876270914077759,
      "epoch": 4.88838747325423,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028569972308519674,
      "loss": 4.896,
      "mean_token_accuracy": 0.22698958814144135,
      "num_tokens": 108976570.0,
      "step": 62830
    },
    {
      "entropy": 5.940917253494263,
      "epoch": 4.888776502625948,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028567203877239914,
      "loss": 5.0416,
      "mean_token_accuracy": 0.21107975095510484,
      "num_tokens": 108985494.0,
      "step": 62835
    },
    {
      "entropy": 6.046118211746216,
      "epoch": 4.889165531997666,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028564435429767035,
      "loss": 5.0312,
      "mean_token_accuracy": 0.2123110443353653,
      "num_tokens": 108994252.0,
      "step": 62840
    },
    {
      "entropy": 5.948751926422119,
      "epoch": 4.889554561369383,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002856166696614303,
      "loss": 4.9698,
      "mean_token_accuracy": 0.2239857643842697,
      "num_tokens": 109003995.0,
      "step": 62845
    },
    {
      "entropy": 5.964430618286133,
      "epoch": 4.889943590741101,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002855889848640992,
      "loss": 4.924,
      "mean_token_accuracy": 0.22492528706789017,
      "num_tokens": 109012914.0,
      "step": 62850
    },
    {
      "entropy": 6.027150058746338,
      "epoch": 4.890332620112819,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002855612999060971,
      "loss": 5.0474,
      "mean_token_accuracy": 0.2087967649102211,
      "num_tokens": 109021543.0,
      "step": 62855
    },
    {
      "entropy": 5.891846323013306,
      "epoch": 4.8907216494845365,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002855336147878442,
      "loss": 4.9439,
      "mean_token_accuracy": 0.22011609375476837,
      "num_tokens": 109030228.0,
      "step": 62860
    },
    {
      "entropy": 5.8612734317779545,
      "epoch": 4.891110678856253,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028550592950976036,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21889870166778563,
      "num_tokens": 109040203.0,
      "step": 62865
    },
    {
      "entropy": 6.011459922790527,
      "epoch": 4.891499708227971,
      "grad_norm": 1.578125,
      "learning_rate": 0.00028547824407226574,
      "loss": 4.9769,
      "mean_token_accuracy": 0.21611387580633162,
      "num_tokens": 109048718.0,
      "step": 62870
    },
    {
      "entropy": 5.932864189147949,
      "epoch": 4.891888737599689,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002854505584757805,
      "loss": 4.9353,
      "mean_token_accuracy": 0.21402130275964737,
      "num_tokens": 109057006.0,
      "step": 62875
    },
    {
      "entropy": 5.965393018722534,
      "epoch": 4.892277766971406,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028542287272072475,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20908736139535905,
      "num_tokens": 109065547.0,
      "step": 62880
    },
    {
      "entropy": 5.888929986953736,
      "epoch": 4.892666796343124,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002853951868075184,
      "loss": 4.9034,
      "mean_token_accuracy": 0.21744290739297867,
      "num_tokens": 109074856.0,
      "step": 62885
    },
    {
      "entropy": 5.950581789016724,
      "epoch": 4.893055825714842,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002853675007365816,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21567222028970717,
      "num_tokens": 109082810.0,
      "step": 62890
    },
    {
      "entropy": 5.8925902366638185,
      "epoch": 4.893444855086559,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002853398145083345,
      "loss": 4.9158,
      "mean_token_accuracy": 0.22758617848157883,
      "num_tokens": 109091737.0,
      "step": 62895
    },
    {
      "entropy": 5.963476610183716,
      "epoch": 4.893833884458276,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028531212812319724,
      "loss": 5.0543,
      "mean_token_accuracy": 0.20738180875778198,
      "num_tokens": 109101145.0,
      "step": 62900
    },
    {
      "entropy": 5.998518323898315,
      "epoch": 4.894222913829994,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002852844415815898,
      "loss": 4.9943,
      "mean_token_accuracy": 0.2150094226002693,
      "num_tokens": 109110633.0,
      "step": 62905
    },
    {
      "entropy": 5.991554594039917,
      "epoch": 4.894611943201712,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028525675488393234,
      "loss": 5.0588,
      "mean_token_accuracy": 0.206807379424572,
      "num_tokens": 109120496.0,
      "step": 62910
    },
    {
      "entropy": 5.925271940231323,
      "epoch": 4.895000972573429,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028522906803064494,
      "loss": 4.9205,
      "mean_token_accuracy": 0.2188255399465561,
      "num_tokens": 109129270.0,
      "step": 62915
    },
    {
      "entropy": 5.9234319686889645,
      "epoch": 4.895390001945147,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002852013810221477,
      "loss": 4.9214,
      "mean_token_accuracy": 0.22282865196466445,
      "num_tokens": 109137285.0,
      "step": 62920
    },
    {
      "entropy": 5.917309427261353,
      "epoch": 4.895779031316865,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002851736938588607,
      "loss": 4.9705,
      "mean_token_accuracy": 0.21477615237236022,
      "num_tokens": 109145613.0,
      "step": 62925
    },
    {
      "entropy": 5.920075130462647,
      "epoch": 4.8961680606885825,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002851460065412042,
      "loss": 5.0531,
      "mean_token_accuracy": 0.21613599509000778,
      "num_tokens": 109154894.0,
      "step": 62930
    },
    {
      "entropy": 5.972373914718628,
      "epoch": 4.896557090060299,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002851183190695981,
      "loss": 4.995,
      "mean_token_accuracy": 0.2169308543205261,
      "num_tokens": 109163588.0,
      "step": 62935
    },
    {
      "entropy": 6.050361156463623,
      "epoch": 4.896946119432017,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002850906314444627,
      "loss": 4.9696,
      "mean_token_accuracy": 0.20923201143741607,
      "num_tokens": 109172120.0,
      "step": 62940
    },
    {
      "entropy": 5.988979339599609,
      "epoch": 4.897335148803735,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00028506294366621796,
      "loss": 4.9137,
      "mean_token_accuracy": 0.22414956539869307,
      "num_tokens": 109180474.0,
      "step": 62945
    },
    {
      "entropy": 5.924572944641113,
      "epoch": 4.897724178175452,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028503525573528413,
      "loss": 4.8542,
      "mean_token_accuracy": 0.22324741035699844,
      "num_tokens": 109188832.0,
      "step": 62950
    },
    {
      "entropy": 5.848375701904297,
      "epoch": 4.89811320754717,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002850075676520812,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22773393392562866,
      "num_tokens": 109197508.0,
      "step": 62955
    },
    {
      "entropy": 5.919560050964355,
      "epoch": 4.898502236918888,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002849798794170294,
      "loss": 5.0212,
      "mean_token_accuracy": 0.21003840863704681,
      "num_tokens": 109205980.0,
      "step": 62960
    },
    {
      "entropy": 5.880718421936035,
      "epoch": 4.898891266290605,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002849521910305488,
      "loss": 4.8845,
      "mean_token_accuracy": 0.21703977584838868,
      "num_tokens": 109215169.0,
      "step": 62965
    },
    {
      "entropy": 5.916193962097168,
      "epoch": 4.899280295662322,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028492450249305957,
      "loss": 4.9266,
      "mean_token_accuracy": 0.2168121874332428,
      "num_tokens": 109224539.0,
      "step": 62970
    },
    {
      "entropy": 6.097114944458008,
      "epoch": 4.89966932503404,
      "grad_norm": 1.46875,
      "learning_rate": 0.00028489681380498177,
      "loss": 5.1377,
      "mean_token_accuracy": 0.20167911052703857,
      "num_tokens": 109233455.0,
      "step": 62975
    },
    {
      "entropy": 5.892354583740234,
      "epoch": 4.900058354405758,
      "grad_norm": 1.515625,
      "learning_rate": 0.00028486912496673557,
      "loss": 4.8298,
      "mean_token_accuracy": 0.22719240039587021,
      "num_tokens": 109241711.0,
      "step": 62980
    },
    {
      "entropy": 5.9986247539520265,
      "epoch": 4.900447383777475,
      "grad_norm": 1.546875,
      "learning_rate": 0.00028484143597874115,
      "loss": 5.0373,
      "mean_token_accuracy": 0.2098922997713089,
      "num_tokens": 109249459.0,
      "step": 62985
    },
    {
      "entropy": 5.91586742401123,
      "epoch": 4.900836413149193,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002848137468414186,
      "loss": 4.95,
      "mean_token_accuracy": 0.20841717422008516,
      "num_tokens": 109258404.0,
      "step": 62990
    },
    {
      "entropy": 5.974041748046875,
      "epoch": 4.901225442520911,
      "grad_norm": 1.484375,
      "learning_rate": 0.000284786057555188,
      "loss": 4.9111,
      "mean_token_accuracy": 0.2193846434354782,
      "num_tokens": 109266519.0,
      "step": 62995
    },
    {
      "entropy": 5.930045032501221,
      "epoch": 4.9016144718926276,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002847583681204696,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21822151839733123,
      "num_tokens": 109275177.0,
      "step": 63000
    },
    {
      "epoch": 4.9016144718926276,
      "eval_entropy": 5.652714310665332,
      "eval_loss": 5.0592145919799805,
      "eval_mean_token_accuracy": 0.22115875099919186,
      "eval_num_tokens": 109275177.0,
      "eval_runtime": 21.6139,
      "eval_samples_per_second": 1922.745,
      "eval_steps_per_second": 240.355,
      "step": 63000
    },
    {
      "entropy": 5.85461311340332,
      "epoch": 4.902003501264345,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028473067853768347,
      "loss": 4.8542,
      "mean_token_accuracy": 0.22625658363103868,
      "num_tokens": 109284009.0,
      "step": 63005
    },
    {
      "entropy": 5.888583612442017,
      "epoch": 4.902392530636063,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028470298880724985,
      "loss": 4.9368,
      "mean_token_accuracy": 0.22174277454614638,
      "num_tokens": 109292197.0,
      "step": 63010
    },
    {
      "entropy": 5.957393980026245,
      "epoch": 4.902781560007781,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002846752989295888,
      "loss": 4.9689,
      "mean_token_accuracy": 0.21476148813962936,
      "num_tokens": 109300949.0,
      "step": 63015
    },
    {
      "entropy": 5.9886205196380615,
      "epoch": 4.903170589379498,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002846476089051204,
      "loss": 4.9704,
      "mean_token_accuracy": 0.21919180899858476,
      "num_tokens": 109309331.0,
      "step": 63020
    },
    {
      "entropy": 5.962155771255493,
      "epoch": 4.903559618751216,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00028461991873426494,
      "loss": 4.9842,
      "mean_token_accuracy": 0.21331615447998048,
      "num_tokens": 109319025.0,
      "step": 63025
    },
    {
      "entropy": 5.929145860671997,
      "epoch": 4.903948648122933,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002845922284174426,
      "loss": 4.8825,
      "mean_token_accuracy": 0.21861564069986344,
      "num_tokens": 109328063.0,
      "step": 63030
    },
    {
      "entropy": 5.902605152130127,
      "epoch": 4.9043376774946505,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028456453795507335,
      "loss": 4.8306,
      "mean_token_accuracy": 0.22251228541135787,
      "num_tokens": 109336214.0,
      "step": 63035
    },
    {
      "entropy": 5.940720176696777,
      "epoch": 4.904726706866368,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002845368473475776,
      "loss": 5.008,
      "mean_token_accuracy": 0.21293063461780548,
      "num_tokens": 109345479.0,
      "step": 63040
    },
    {
      "entropy": 5.951022672653198,
      "epoch": 4.905115736238086,
      "grad_norm": 1.546875,
      "learning_rate": 0.00028450915659537524,
      "loss": 4.9215,
      "mean_token_accuracy": 0.22367617189884187,
      "num_tokens": 109353881.0,
      "step": 63045
    },
    {
      "entropy": 5.927049350738526,
      "epoch": 4.905504765609804,
      "grad_norm": 1.375,
      "learning_rate": 0.0002844814656988866,
      "loss": 4.972,
      "mean_token_accuracy": 0.21864169836044312,
      "num_tokens": 109362693.0,
      "step": 63050
    },
    {
      "entropy": 5.853742027282715,
      "epoch": 4.905893794981521,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028445377465853176,
      "loss": 4.8163,
      "mean_token_accuracy": 0.2237616539001465,
      "num_tokens": 109372187.0,
      "step": 63055
    },
    {
      "entropy": 5.916268730163575,
      "epoch": 4.906282824353239,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00028442608347473093,
      "loss": 4.9517,
      "mean_token_accuracy": 0.21607140302658082,
      "num_tokens": 109380773.0,
      "step": 63060
    },
    {
      "entropy": 5.841884803771973,
      "epoch": 4.906671853724957,
      "grad_norm": 1.515625,
      "learning_rate": 0.00028439839214790437,
      "loss": 4.8816,
      "mean_token_accuracy": 0.23068843632936478,
      "num_tokens": 109390005.0,
      "step": 63065
    },
    {
      "entropy": 5.952748727798462,
      "epoch": 4.9070608830966735,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002843707006784722,
      "loss": 4.9273,
      "mean_token_accuracy": 0.21825793087482454,
      "num_tokens": 109398524.0,
      "step": 63070
    },
    {
      "entropy": 6.007682991027832,
      "epoch": 4.907449912468391,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002843430090668545,
      "loss": 5.0055,
      "mean_token_accuracy": 0.2087055191397667,
      "num_tokens": 109407966.0,
      "step": 63075
    },
    {
      "entropy": 5.9114298820495605,
      "epoch": 4.907838941840109,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028431531731347155,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21217501908540726,
      "num_tokens": 109416735.0,
      "step": 63080
    },
    {
      "entropy": 5.9625335216522215,
      "epoch": 4.908227971211827,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002842876254187434,
      "loss": 4.9963,
      "mean_token_accuracy": 0.21428932398557662,
      "num_tokens": 109425770.0,
      "step": 63085
    },
    {
      "entropy": 5.956892299652099,
      "epoch": 4.908617000583544,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028425993338309036,
      "loss": 4.9174,
      "mean_token_accuracy": 0.21090444028377534,
      "num_tokens": 109433631.0,
      "step": 63090
    },
    {
      "entropy": 5.980113172531128,
      "epoch": 4.909006029955261,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002842322412069325,
      "loss": 5.0563,
      "mean_token_accuracy": 0.20872304439544678,
      "num_tokens": 109441692.0,
      "step": 63095
    },
    {
      "entropy": 5.955261468887329,
      "epoch": 4.909395059326979,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002842045488906902,
      "loss": 4.9924,
      "mean_token_accuracy": 0.21723778396844864,
      "num_tokens": 109450805.0,
      "step": 63100
    },
    {
      "entropy": 5.934364461898804,
      "epoch": 4.9097840886986965,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028417685643478347,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22776860892772674,
      "num_tokens": 109459500.0,
      "step": 63105
    },
    {
      "entropy": 5.962911939620971,
      "epoch": 4.910173118070414,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00028414916383963265,
      "loss": 4.9898,
      "mean_token_accuracy": 0.21713587492704392,
      "num_tokens": 109467303.0,
      "step": 63110
    },
    {
      "entropy": 5.908255958557129,
      "epoch": 4.910562147442132,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002841214711056577,
      "loss": 4.978,
      "mean_token_accuracy": 0.2208307832479477,
      "num_tokens": 109475752.0,
      "step": 63115
    },
    {
      "entropy": 5.965822315216064,
      "epoch": 4.91095117681385,
      "grad_norm": 1.4375,
      "learning_rate": 0.000284093778233279,
      "loss": 4.9269,
      "mean_token_accuracy": 0.22186021208763124,
      "num_tokens": 109484385.0,
      "step": 63120
    },
    {
      "entropy": 5.914725589752197,
      "epoch": 4.911340206185567,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002840660852229166,
      "loss": 4.9932,
      "mean_token_accuracy": 0.21080339848995208,
      "num_tokens": 109492674.0,
      "step": 63125
    },
    {
      "entropy": 6.014460945129395,
      "epoch": 4.911729235557285,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002840383920749908,
      "loss": 5.0322,
      "mean_token_accuracy": 0.21515663266181945,
      "num_tokens": 109502027.0,
      "step": 63130
    },
    {
      "entropy": 6.028676080703735,
      "epoch": 4.912118264929002,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002840106987899219,
      "loss": 4.9586,
      "mean_token_accuracy": 0.21193777918815612,
      "num_tokens": 109510324.0,
      "step": 63135
    },
    {
      "entropy": 5.919425535202026,
      "epoch": 4.9125072943007195,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002839830053681299,
      "loss": 4.9391,
      "mean_token_accuracy": 0.2134452596306801,
      "num_tokens": 109519097.0,
      "step": 63140
    },
    {
      "entropy": 5.941795063018799,
      "epoch": 4.912896323672437,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028395531181003516,
      "loss": 4.9431,
      "mean_token_accuracy": 0.21991322040557862,
      "num_tokens": 109527841.0,
      "step": 63145
    },
    {
      "entropy": 5.9953711986541744,
      "epoch": 4.913285353044155,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002839276181160577,
      "loss": 4.9777,
      "mean_token_accuracy": 0.21783761382102967,
      "num_tokens": 109536222.0,
      "step": 63150
    },
    {
      "entropy": 6.038424491882324,
      "epoch": 4.913674382415873,
      "grad_norm": 1.359375,
      "learning_rate": 0.000283899924286618,
      "loss": 5.0843,
      "mean_token_accuracy": 0.20836381763219833,
      "num_tokens": 109545727.0,
      "step": 63155
    },
    {
      "entropy": 5.993653154373169,
      "epoch": 4.91406341178759,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000283872230322136,
      "loss": 5.0282,
      "mean_token_accuracy": 0.20641112476587295,
      "num_tokens": 109554350.0,
      "step": 63160
    },
    {
      "entropy": 6.023221015930176,
      "epoch": 4.914452441159307,
      "grad_norm": 1.546875,
      "learning_rate": 0.00028384453622303207,
      "loss": 5.0713,
      "mean_token_accuracy": 0.20121698081493378,
      "num_tokens": 109562950.0,
      "step": 63165
    },
    {
      "entropy": 6.006545925140381,
      "epoch": 4.914841470531025,
      "grad_norm": 1.59375,
      "learning_rate": 0.00028381684198972636,
      "loss": 4.9943,
      "mean_token_accuracy": 0.21690357476472855,
      "num_tokens": 109570956.0,
      "step": 63170
    },
    {
      "entropy": 5.959226036071778,
      "epoch": 4.9152304999027425,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028378914762263905,
      "loss": 4.9924,
      "mean_token_accuracy": 0.21950844824314117,
      "num_tokens": 109579991.0,
      "step": 63175
    },
    {
      "entropy": 5.920270681381226,
      "epoch": 4.91561952927446,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028376145312219054,
      "loss": 4.972,
      "mean_token_accuracy": 0.21749282032251357,
      "num_tokens": 109588006.0,
      "step": 63180
    },
    {
      "entropy": 5.895156145095825,
      "epoch": 4.916008558646178,
      "grad_norm": 1.640625,
      "learning_rate": 0.0002837337584888009,
      "loss": 4.8598,
      "mean_token_accuracy": 0.2223282963037491,
      "num_tokens": 109595568.0,
      "step": 63185
    },
    {
      "entropy": 5.984888744354248,
      "epoch": 4.916397588017896,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002837060637228903,
      "loss": 5.0061,
      "mean_token_accuracy": 0.21437182873487473,
      "num_tokens": 109603654.0,
      "step": 63190
    },
    {
      "entropy": 6.0535430908203125,
      "epoch": 4.916786617389613,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028367836882487905,
      "loss": 5.084,
      "mean_token_accuracy": 0.20416779518127443,
      "num_tokens": 109612303.0,
      "step": 63195
    },
    {
      "entropy": 6.0062463760375975,
      "epoch": 4.91717564676133,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002836506737951875,
      "loss": 5.0496,
      "mean_token_accuracy": 0.2100728929042816,
      "num_tokens": 109620829.0,
      "step": 63200
    },
    {
      "entropy": 6.015235900878906,
      "epoch": 4.917564676133048,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002836229786342356,
      "loss": 5.0459,
      "mean_token_accuracy": 0.20695452243089676,
      "num_tokens": 109629140.0,
      "step": 63205
    },
    {
      "entropy": 5.911878442764282,
      "epoch": 4.9179537055047655,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002835952833424438,
      "loss": 4.9595,
      "mean_token_accuracy": 0.21258497834205628,
      "num_tokens": 109637951.0,
      "step": 63210
    },
    {
      "entropy": 5.953508138656616,
      "epoch": 4.918342734876483,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028356758792023217,
      "loss": 4.9155,
      "mean_token_accuracy": 0.22351914048194885,
      "num_tokens": 109646243.0,
      "step": 63215
    },
    {
      "entropy": 5.981290674209594,
      "epoch": 4.918731764248201,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028353989236802114,
      "loss": 4.9649,
      "mean_token_accuracy": 0.212136909365654,
      "num_tokens": 109655426.0,
      "step": 63220
    },
    {
      "entropy": 5.998653078079224,
      "epoch": 4.919120793619919,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028351219668623076,
      "loss": 5.0198,
      "mean_token_accuracy": 0.20141818821430207,
      "num_tokens": 109664131.0,
      "step": 63225
    },
    {
      "entropy": 5.969208717346191,
      "epoch": 4.919509822991635,
      "grad_norm": 1.5,
      "learning_rate": 0.0002834845008752813,
      "loss": 4.9109,
      "mean_token_accuracy": 0.21544253677129746,
      "num_tokens": 109673252.0,
      "step": 63230
    },
    {
      "entropy": 5.954246282577515,
      "epoch": 4.919898852363353,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028345680493559313,
      "loss": 4.9093,
      "mean_token_accuracy": 0.22585375010967254,
      "num_tokens": 109681224.0,
      "step": 63235
    },
    {
      "entropy": 5.983716726303101,
      "epoch": 4.920287881735071,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002834291088675864,
      "loss": 4.9866,
      "mean_token_accuracy": 0.21000488996505737,
      "num_tokens": 109690187.0,
      "step": 63240
    },
    {
      "entropy": 5.950297737121582,
      "epoch": 4.9206769111067885,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00028340141267168135,
      "loss": 4.9598,
      "mean_token_accuracy": 0.22097481340169906,
      "num_tokens": 109699204.0,
      "step": 63245
    },
    {
      "entropy": 5.952619314193726,
      "epoch": 4.921065940478506,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028337371634829824,
      "loss": 4.9662,
      "mean_token_accuracy": 0.2160538002848625,
      "num_tokens": 109708151.0,
      "step": 63250
    },
    {
      "entropy": 5.976229810714722,
      "epoch": 4.921454969850224,
      "grad_norm": 1.5,
      "learning_rate": 0.0002833460198978573,
      "loss": 5.0209,
      "mean_token_accuracy": 0.21549562811851503,
      "num_tokens": 109717304.0,
      "step": 63255
    },
    {
      "entropy": 5.995584487915039,
      "epoch": 4.921843999221942,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002833183233207788,
      "loss": 5.0198,
      "mean_token_accuracy": 0.2153089627623558,
      "num_tokens": 109727088.0,
      "step": 63260
    },
    {
      "entropy": 6.0054210186004635,
      "epoch": 4.922233028593659,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028329062661748295,
      "loss": 4.9669,
      "mean_token_accuracy": 0.2194717973470688,
      "num_tokens": 109735699.0,
      "step": 63265
    },
    {
      "entropy": 6.041084718704224,
      "epoch": 4.922622057965376,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028326292978839005,
      "loss": 4.969,
      "mean_token_accuracy": 0.2195608824491501,
      "num_tokens": 109743596.0,
      "step": 63270
    },
    {
      "entropy": 5.901253080368042,
      "epoch": 4.923011087337094,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00028323523283392034,
      "loss": 4.8931,
      "mean_token_accuracy": 0.22465574592351914,
      "num_tokens": 109751699.0,
      "step": 63275
    },
    {
      "entropy": 5.961084604263306,
      "epoch": 4.9234001167088115,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028320753575449415,
      "loss": 4.9923,
      "mean_token_accuracy": 0.20973095297813416,
      "num_tokens": 109759890.0,
      "step": 63280
    },
    {
      "entropy": 6.010225868225097,
      "epoch": 4.923789146080529,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002831798385505316,
      "loss": 5.1123,
      "mean_token_accuracy": 0.20896685123443604,
      "num_tokens": 109768428.0,
      "step": 63285
    },
    {
      "entropy": 5.966949415206909,
      "epoch": 4.924178175452247,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00028315214122245295,
      "loss": 4.9139,
      "mean_token_accuracy": 0.2185269147157669,
      "num_tokens": 109776758.0,
      "step": 63290
    },
    {
      "entropy": 6.007488489151001,
      "epoch": 4.924567204823965,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002831244437706786,
      "loss": 5.0092,
      "mean_token_accuracy": 0.2021022543311119,
      "num_tokens": 109785572.0,
      "step": 63295
    },
    {
      "entropy": 5.950564575195313,
      "epoch": 4.924956234195681,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002830967461956288,
      "loss": 4.9732,
      "mean_token_accuracy": 0.21577254980802535,
      "num_tokens": 109794395.0,
      "step": 63300
    },
    {
      "entropy": 6.005304145812988,
      "epoch": 4.925345263567399,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028306904849772374,
      "loss": 4.9539,
      "mean_token_accuracy": 0.21857646554708482,
      "num_tokens": 109802238.0,
      "step": 63305
    },
    {
      "entropy": 5.958049964904785,
      "epoch": 4.925734292939117,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002830413506773837,
      "loss": 4.9613,
      "mean_token_accuracy": 0.21227618604898452,
      "num_tokens": 109811151.0,
      "step": 63310
    },
    {
      "entropy": 5.917121934890747,
      "epoch": 4.9261233223108345,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000283013652735029,
      "loss": 4.9732,
      "mean_token_accuracy": 0.22661359310150148,
      "num_tokens": 109819622.0,
      "step": 63315
    },
    {
      "entropy": 6.0441161632537845,
      "epoch": 4.926512351682552,
      "grad_norm": 1.5,
      "learning_rate": 0.00028298595467107985,
      "loss": 5.0571,
      "mean_token_accuracy": 0.2033882573246956,
      "num_tokens": 109827993.0,
      "step": 63320
    },
    {
      "entropy": 6.068973064422607,
      "epoch": 4.92690138105427,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002829582564859565,
      "loss": 5.0779,
      "mean_token_accuracy": 0.20832820534706115,
      "num_tokens": 109836731.0,
      "step": 63325
    },
    {
      "entropy": 6.009917449951172,
      "epoch": 4.927290410425988,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002829305581800793,
      "loss": 5.0101,
      "mean_token_accuracy": 0.21852977573871613,
      "num_tokens": 109845083.0,
      "step": 63330
    },
    {
      "entropy": 6.019495820999145,
      "epoch": 4.927679439797704,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002829028597538685,
      "loss": 4.9186,
      "mean_token_accuracy": 0.21844810396432876,
      "num_tokens": 109853118.0,
      "step": 63335
    },
    {
      "entropy": 5.976355695724488,
      "epoch": 4.928068469169422,
      "grad_norm": 1.546875,
      "learning_rate": 0.00028287516120774445,
      "loss": 5.0445,
      "mean_token_accuracy": 0.21278872191905976,
      "num_tokens": 109861212.0,
      "step": 63340
    },
    {
      "entropy": 5.9101625919342045,
      "epoch": 4.92845749854114,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002828474625421273,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21823876053094865,
      "num_tokens": 109868969.0,
      "step": 63345
    },
    {
      "entropy": 5.988764142990112,
      "epoch": 4.9288465279128575,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002828197637574374,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2206467255949974,
      "num_tokens": 109878284.0,
      "step": 63350
    },
    {
      "entropy": 5.98193850517273,
      "epoch": 4.929235557284575,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028279206485409505,
      "loss": 4.9813,
      "mean_token_accuracy": 0.21575236916542054,
      "num_tokens": 109886815.0,
      "step": 63355
    },
    {
      "entropy": 6.007997179031372,
      "epoch": 4.929624586656293,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002827643658325205,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21635722219944,
      "num_tokens": 109896163.0,
      "step": 63360
    },
    {
      "entropy": 6.005470228195191,
      "epoch": 4.93001361602801,
      "grad_norm": 1.609375,
      "learning_rate": 0.000282736666693134,
      "loss": 4.976,
      "mean_token_accuracy": 0.21511476784944533,
      "num_tokens": 109904995.0,
      "step": 63365
    },
    {
      "entropy": 6.000383090972901,
      "epoch": 4.930402645399727,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002827089674363559,
      "loss": 4.9335,
      "mean_token_accuracy": 0.2165600150823593,
      "num_tokens": 109913168.0,
      "step": 63370
    },
    {
      "entropy": 5.893802499771118,
      "epoch": 4.930791674771445,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002826812680626066,
      "loss": 4.9933,
      "mean_token_accuracy": 0.21067523658275605,
      "num_tokens": 109921915.0,
      "step": 63375
    },
    {
      "entropy": 5.914792156219482,
      "epoch": 4.931180704143163,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002826535685723062,
      "loss": 4.977,
      "mean_token_accuracy": 0.2129411831498146,
      "num_tokens": 109931294.0,
      "step": 63380
    },
    {
      "entropy": 6.013063955307007,
      "epoch": 4.9315697335148805,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028262586896587517,
      "loss": 4.9669,
      "mean_token_accuracy": 0.2112464725971222,
      "num_tokens": 109939864.0,
      "step": 63385
    },
    {
      "entropy": 5.958077096939087,
      "epoch": 4.931958762886598,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002825981692437336,
      "loss": 4.9713,
      "mean_token_accuracy": 0.218314266204834,
      "num_tokens": 109948715.0,
      "step": 63390
    },
    {
      "entropy": 6.015986490249634,
      "epoch": 4.932347792258316,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028257046940630197,
      "loss": 5.0668,
      "mean_token_accuracy": 0.2084370583295822,
      "num_tokens": 109957763.0,
      "step": 63395
    },
    {
      "entropy": 5.95835313796997,
      "epoch": 4.932736821630033,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028254276945400056,
      "loss": 4.9215,
      "mean_token_accuracy": 0.21456012427806853,
      "num_tokens": 109966215.0,
      "step": 63400
    },
    {
      "entropy": 5.978005313873291,
      "epoch": 4.93312585100175,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028251506938724953,
      "loss": 5.017,
      "mean_token_accuracy": 0.21322564035654068,
      "num_tokens": 109975393.0,
      "step": 63405
    },
    {
      "entropy": 5.932893705368042,
      "epoch": 4.933514880373468,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028248736920646927,
      "loss": 4.9176,
      "mean_token_accuracy": 0.22542205601930618,
      "num_tokens": 109984749.0,
      "step": 63410
    },
    {
      "entropy": 5.930654907226563,
      "epoch": 4.933903909745186,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028245966891208017,
      "loss": 4.9304,
      "mean_token_accuracy": 0.21855888664722442,
      "num_tokens": 109993397.0,
      "step": 63415
    },
    {
      "entropy": 5.911979007720947,
      "epoch": 4.9342929391169035,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002824319685045025,
      "loss": 4.9293,
      "mean_token_accuracy": 0.21808079034090042,
      "num_tokens": 110002261.0,
      "step": 63420
    },
    {
      "entropy": 5.9960957050323485,
      "epoch": 4.934681968488621,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028240426798415645,
      "loss": 4.9833,
      "mean_token_accuracy": 0.21386766582727432,
      "num_tokens": 110011276.0,
      "step": 63425
    },
    {
      "entropy": 5.9283815860748295,
      "epoch": 4.935070997860338,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002823765673514625,
      "loss": 4.9492,
      "mean_token_accuracy": 0.20741480886936187,
      "num_tokens": 110020066.0,
      "step": 63430
    },
    {
      "entropy": 6.040676784515381,
      "epoch": 4.935460027232056,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028234886660684084,
      "loss": 5.0929,
      "mean_token_accuracy": 0.19967034906148912,
      "num_tokens": 110028650.0,
      "step": 63435
    },
    {
      "entropy": 5.977835035324096,
      "epoch": 4.935849056603773,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002823211657507118,
      "loss": 4.9468,
      "mean_token_accuracy": 0.21339952945709229,
      "num_tokens": 110037422.0,
      "step": 63440
    },
    {
      "entropy": 5.945356321334839,
      "epoch": 4.936238085975491,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002822934647834958,
      "loss": 4.945,
      "mean_token_accuracy": 0.21767951548099518,
      "num_tokens": 110046449.0,
      "step": 63445
    },
    {
      "entropy": 5.992082405090332,
      "epoch": 4.936627115347209,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028226576370561304,
      "loss": 5.0431,
      "mean_token_accuracy": 0.20125000923871994,
      "num_tokens": 110055089.0,
      "step": 63450
    },
    {
      "entropy": 5.957101678848266,
      "epoch": 4.9370161447189265,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028223806251748384,
      "loss": 4.9868,
      "mean_token_accuracy": 0.2128631815314293,
      "num_tokens": 110064512.0,
      "step": 63455
    },
    {
      "entropy": 6.015191984176636,
      "epoch": 4.937405174090644,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002822103612195286,
      "loss": 5.0449,
      "mean_token_accuracy": 0.2126172110438347,
      "num_tokens": 110073043.0,
      "step": 63460
    },
    {
      "entropy": 5.991922807693482,
      "epoch": 4.937794203462362,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002821826598121676,
      "loss": 4.9766,
      "mean_token_accuracy": 0.21445770114660262,
      "num_tokens": 110080846.0,
      "step": 63465
    },
    {
      "entropy": 6.005674552917481,
      "epoch": 4.938183232834079,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002821549582958212,
      "loss": 5.0235,
      "mean_token_accuracy": 0.2090216636657715,
      "num_tokens": 110090442.0,
      "step": 63470
    },
    {
      "entropy": 5.947021770477295,
      "epoch": 4.938572262205796,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002821272566709096,
      "loss": 4.881,
      "mean_token_accuracy": 0.21784914880990983,
      "num_tokens": 110098982.0,
      "step": 63475
    },
    {
      "entropy": 6.060851192474365,
      "epoch": 4.938961291577514,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028209955493785333,
      "loss": 5.1225,
      "mean_token_accuracy": 0.20657771974802017,
      "num_tokens": 110107912.0,
      "step": 63480
    },
    {
      "entropy": 5.914357376098633,
      "epoch": 4.939350320949232,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002820718530970726,
      "loss": 4.871,
      "mean_token_accuracy": 0.22238297313451766,
      "num_tokens": 110116484.0,
      "step": 63485
    },
    {
      "entropy": 5.97326889038086,
      "epoch": 4.9397393503209495,
      "grad_norm": 1.515625,
      "learning_rate": 0.00028204415114898765,
      "loss": 4.971,
      "mean_token_accuracy": 0.21657685488462447,
      "num_tokens": 110125412.0,
      "step": 63490
    },
    {
      "entropy": 5.946655130386352,
      "epoch": 4.940128379692667,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028201644909401903,
      "loss": 4.9024,
      "mean_token_accuracy": 0.2187240019440651,
      "num_tokens": 110133159.0,
      "step": 63495
    },
    {
      "entropy": 5.955028009414673,
      "epoch": 4.940517409064384,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028198874693258696,
      "loss": 4.9828,
      "mean_token_accuracy": 0.21358607411384584,
      "num_tokens": 110142167.0,
      "step": 63500
    },
    {
      "entropy": 6.044516468048096,
      "epoch": 4.940906438436102,
      "grad_norm": 1.375,
      "learning_rate": 0.0002819610446651117,
      "loss": 5.1063,
      "mean_token_accuracy": 0.20724042356014252,
      "num_tokens": 110151097.0,
      "step": 63505
    },
    {
      "entropy": 5.938747215270996,
      "epoch": 4.941295467807819,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028193334229201374,
      "loss": 4.9301,
      "mean_token_accuracy": 0.2138589709997177,
      "num_tokens": 110160982.0,
      "step": 63510
    },
    {
      "entropy": 6.0448261260986325,
      "epoch": 4.941684497179537,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028190563981371334,
      "loss": 5.0798,
      "mean_token_accuracy": 0.218965807557106,
      "num_tokens": 110169572.0,
      "step": 63515
    },
    {
      "entropy": 5.979134225845337,
      "epoch": 4.942073526551255,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002818779372306308,
      "loss": 5.0086,
      "mean_token_accuracy": 0.20818604081869124,
      "num_tokens": 110177901.0,
      "step": 63520
    },
    {
      "entropy": 6.055978727340698,
      "epoch": 4.9424625559229725,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028185023454318646,
      "loss": 5.1014,
      "mean_token_accuracy": 0.21146378219127654,
      "num_tokens": 110187153.0,
      "step": 63525
    },
    {
      "entropy": 6.060267925262451,
      "epoch": 4.94285158529469,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028182253175180084,
      "loss": 5.0698,
      "mean_token_accuracy": 0.20995452702045442,
      "num_tokens": 110195445.0,
      "step": 63530
    },
    {
      "entropy": 5.959158277511596,
      "epoch": 4.943240614666407,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002817948288568942,
      "loss": 4.9071,
      "mean_token_accuracy": 0.2203369230031967,
      "num_tokens": 110204219.0,
      "step": 63535
    },
    {
      "entropy": 5.931695461273193,
      "epoch": 4.943629644038125,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002817671258588868,
      "loss": 5.0088,
      "mean_token_accuracy": 0.21358156055212021,
      "num_tokens": 110213242.0,
      "step": 63540
    },
    {
      "entropy": 5.922355604171753,
      "epoch": 4.944018673409842,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002817394227581989,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21864015311002732,
      "num_tokens": 110221557.0,
      "step": 63545
    },
    {
      "entropy": 6.030595302581787,
      "epoch": 4.94440770278156,
      "grad_norm": 1.90625,
      "learning_rate": 0.0002817117195552511,
      "loss": 5.0969,
      "mean_token_accuracy": 0.201298888027668,
      "num_tokens": 110230229.0,
      "step": 63550
    },
    {
      "entropy": 5.948713111877441,
      "epoch": 4.944796732153278,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028168401625046363,
      "loss": 4.9552,
      "mean_token_accuracy": 0.21887002140283585,
      "num_tokens": 110238950.0,
      "step": 63555
    },
    {
      "entropy": 5.996391010284424,
      "epoch": 4.9451857615249954,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002816563128442568,
      "loss": 4.9626,
      "mean_token_accuracy": 0.21717875599861144,
      "num_tokens": 110248216.0,
      "step": 63560
    },
    {
      "entropy": 5.966849517822266,
      "epoch": 4.945574790896712,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028162860933705105,
      "loss": 4.9671,
      "mean_token_accuracy": 0.20692503005266188,
      "num_tokens": 110257830.0,
      "step": 63565
    },
    {
      "entropy": 5.955839061737061,
      "epoch": 4.94596382026843,
      "grad_norm": 1.375,
      "learning_rate": 0.0002816009057292667,
      "loss": 4.9609,
      "mean_token_accuracy": 0.21407763808965682,
      "num_tokens": 110266432.0,
      "step": 63570
    },
    {
      "entropy": 5.941349744796753,
      "epoch": 4.946352849640148,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028157320202132417,
      "loss": 4.9632,
      "mean_token_accuracy": 0.21761271208524705,
      "num_tokens": 110275268.0,
      "step": 63575
    },
    {
      "entropy": 5.946577882766723,
      "epoch": 4.946741879011865,
      "grad_norm": 1.375,
      "learning_rate": 0.00028154549821364375,
      "loss": 4.9321,
      "mean_token_accuracy": 0.22142809331417085,
      "num_tokens": 110284100.0,
      "step": 63580
    },
    {
      "entropy": 5.941385173797608,
      "epoch": 4.947130908383583,
      "grad_norm": 1.375,
      "learning_rate": 0.00028151779430664575,
      "loss": 4.992,
      "mean_token_accuracy": 0.2147132560610771,
      "num_tokens": 110292672.0,
      "step": 63585
    },
    {
      "entropy": 5.989567613601684,
      "epoch": 4.947519937755301,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028149009030075056,
      "loss": 4.9508,
      "mean_token_accuracy": 0.2168251484632492,
      "num_tokens": 110301160.0,
      "step": 63590
    },
    {
      "entropy": 5.971408891677856,
      "epoch": 4.947908967127018,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028146238619637865,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21731097549200057,
      "num_tokens": 110310585.0,
      "step": 63595
    },
    {
      "entropy": 5.956007766723633,
      "epoch": 4.948297996498736,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028143468199395036,
      "loss": 4.9764,
      "mean_token_accuracy": 0.2153783842921257,
      "num_tokens": 110318754.0,
      "step": 63600
    },
    {
      "entropy": 5.929397535324097,
      "epoch": 4.948687025870453,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028140697769388597,
      "loss": 5.0067,
      "mean_token_accuracy": 0.21503014713525773,
      "num_tokens": 110328144.0,
      "step": 63605
    },
    {
      "entropy": 5.9350017547607425,
      "epoch": 4.949076055242171,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002813792732966059,
      "loss": 4.9191,
      "mean_token_accuracy": 0.21589315980672835,
      "num_tokens": 110337151.0,
      "step": 63610
    },
    {
      "entropy": 5.946207714080811,
      "epoch": 4.949465084613888,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002813515688025305,
      "loss": 4.9217,
      "mean_token_accuracy": 0.21406424790620804,
      "num_tokens": 110344899.0,
      "step": 63615
    },
    {
      "entropy": 6.006354236602784,
      "epoch": 4.949854113985606,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002813238642120802,
      "loss": 5.0176,
      "mean_token_accuracy": 0.21048103272914886,
      "num_tokens": 110354745.0,
      "step": 63620
    },
    {
      "entropy": 5.918613481521606,
      "epoch": 4.950243143357324,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028129615952567517,
      "loss": 4.9478,
      "mean_token_accuracy": 0.21566716134548186,
      "num_tokens": 110363226.0,
      "step": 63625
    },
    {
      "entropy": 5.973747205734253,
      "epoch": 4.9506321727290405,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002812684547437361,
      "loss": 4.9908,
      "mean_token_accuracy": 0.21028875112533568,
      "num_tokens": 110371775.0,
      "step": 63630
    },
    {
      "entropy": 6.033539962768555,
      "epoch": 4.951021202100758,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002812407498666831,
      "loss": 5.0665,
      "mean_token_accuracy": 0.2063036948442459,
      "num_tokens": 110380769.0,
      "step": 63635
    },
    {
      "entropy": 5.9688104629516605,
      "epoch": 4.951410231472476,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002812130448949367,
      "loss": 4.948,
      "mean_token_accuracy": 0.21649263203144073,
      "num_tokens": 110388858.0,
      "step": 63640
    },
    {
      "entropy": 5.90745964050293,
      "epoch": 4.951799260844194,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028118533982891727,
      "loss": 4.8258,
      "mean_token_accuracy": 0.22260864377021788,
      "num_tokens": 110397285.0,
      "step": 63645
    },
    {
      "entropy": 5.999681091308593,
      "epoch": 4.952188290215911,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002811576346690451,
      "loss": 4.9624,
      "mean_token_accuracy": 0.2195064753293991,
      "num_tokens": 110404954.0,
      "step": 63650
    },
    {
      "entropy": 5.897418451309204,
      "epoch": 4.952577319587629,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00028112992941574074,
      "loss": 4.9254,
      "mean_token_accuracy": 0.22009748667478563,
      "num_tokens": 110413546.0,
      "step": 63655
    },
    {
      "entropy": 5.929078912734985,
      "epoch": 4.952966348959347,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00028110222406942424,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2209790751338005,
      "num_tokens": 110421670.0,
      "step": 63660
    },
    {
      "entropy": 5.986019039154053,
      "epoch": 4.953355378331064,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002810745186305164,
      "loss": 4.9839,
      "mean_token_accuracy": 0.21751054972410203,
      "num_tokens": 110430362.0,
      "step": 63665
    },
    {
      "entropy": 5.991124153137207,
      "epoch": 4.953744407702781,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002810468130994373,
      "loss": 5.0026,
      "mean_token_accuracy": 0.22032052874565125,
      "num_tokens": 110438632.0,
      "step": 63670
    },
    {
      "entropy": 6.024920463562012,
      "epoch": 4.954133437074499,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028101910747660745,
      "loss": 4.9777,
      "mean_token_accuracy": 0.20974259823560715,
      "num_tokens": 110448080.0,
      "step": 63675
    },
    {
      "entropy": 5.980934286117554,
      "epoch": 4.954522466446217,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002809914017624472,
      "loss": 4.903,
      "mean_token_accuracy": 0.2216262251138687,
      "num_tokens": 110457028.0,
      "step": 63680
    },
    {
      "entropy": 5.9656037330627445,
      "epoch": 4.954911495817934,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002809636959573771,
      "loss": 4.972,
      "mean_token_accuracy": 0.2158194825053215,
      "num_tokens": 110466671.0,
      "step": 63685
    },
    {
      "entropy": 6.050493574142456,
      "epoch": 4.955300525189652,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002809359900618173,
      "loss": 5.1307,
      "mean_token_accuracy": 0.1987726017832756,
      "num_tokens": 110475679.0,
      "step": 63690
    },
    {
      "entropy": 6.0023431301116945,
      "epoch": 4.95568955456137,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028090828407618837,
      "loss": 4.9867,
      "mean_token_accuracy": 0.21258708834648132,
      "num_tokens": 110484280.0,
      "step": 63695
    },
    {
      "entropy": 6.004178142547607,
      "epoch": 4.9560785839330865,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00028088057800091054,
      "loss": 5.041,
      "mean_token_accuracy": 0.21251711398363113,
      "num_tokens": 110493160.0,
      "step": 63700
    },
    {
      "entropy": 5.971774101257324,
      "epoch": 4.956467613304804,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002808528718364043,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2203864723443985,
      "num_tokens": 110501409.0,
      "step": 63705
    },
    {
      "entropy": 5.986966276168824,
      "epoch": 4.956856642676522,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028082516558309005,
      "loss": 5.0166,
      "mean_token_accuracy": 0.20873988717794417,
      "num_tokens": 110510342.0,
      "step": 63710
    },
    {
      "entropy": 5.92267336845398,
      "epoch": 4.95724567204824,
      "grad_norm": 1.5,
      "learning_rate": 0.0002807974592413883,
      "loss": 4.9176,
      "mean_token_accuracy": 0.21992526650428773,
      "num_tokens": 110519226.0,
      "step": 63715
    },
    {
      "entropy": 5.944930791854858,
      "epoch": 4.957634701419957,
      "grad_norm": 1.46875,
      "learning_rate": 0.00028076975281171926,
      "loss": 4.9072,
      "mean_token_accuracy": 0.22562885880470276,
      "num_tokens": 110528116.0,
      "step": 63720
    },
    {
      "entropy": 5.992868328094483,
      "epoch": 4.958023730791675,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028074204629450334,
      "loss": 4.9913,
      "mean_token_accuracy": 0.22008632272481918,
      "num_tokens": 110536853.0,
      "step": 63725
    },
    {
      "entropy": 5.929089069366455,
      "epoch": 4.958412760163393,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002807143396901611,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21657758057117463,
      "num_tokens": 110547111.0,
      "step": 63730
    },
    {
      "entropy": 5.9626435279846195,
      "epoch": 4.9588017895351095,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028068663299911286,
      "loss": 4.9313,
      "mean_token_accuracy": 0.2258359596133232,
      "num_tokens": 110556223.0,
      "step": 63735
    },
    {
      "entropy": 5.950687742233276,
      "epoch": 4.959190818906827,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002806589262217789,
      "loss": 4.8906,
      "mean_token_accuracy": 0.22261662930250167,
      "num_tokens": 110564531.0,
      "step": 63740
    },
    {
      "entropy": 5.943018102645874,
      "epoch": 4.959579848278545,
      "grad_norm": 1.59375,
      "learning_rate": 0.00028063121935857985,
      "loss": 4.9269,
      "mean_token_accuracy": 0.22053411602973938,
      "num_tokens": 110572298.0,
      "step": 63745
    },
    {
      "entropy": 5.933963489532471,
      "epoch": 4.959968877650263,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002806035124099359,
      "loss": 4.9507,
      "mean_token_accuracy": 0.21394215673208236,
      "num_tokens": 110580774.0,
      "step": 63750
    },
    {
      "entropy": 5.898367500305175,
      "epoch": 4.96035790702198,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00028057580537626766,
      "loss": 4.9627,
      "mean_token_accuracy": 0.21935340911149978,
      "num_tokens": 110589681.0,
      "step": 63755
    },
    {
      "entropy": 6.004323482513428,
      "epoch": 4.960746936393698,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028054809825799537,
      "loss": 4.9782,
      "mean_token_accuracy": 0.2105149194598198,
      "num_tokens": 110598409.0,
      "step": 63760
    },
    {
      "entropy": 5.9494309425354,
      "epoch": 4.961135965765415,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028052039105553965,
      "loss": 4.9682,
      "mean_token_accuracy": 0.21629618257284164,
      "num_tokens": 110607235.0,
      "step": 63765
    },
    {
      "entropy": 6.0056249618530275,
      "epoch": 4.9615249951371325,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002804926837693207,
      "loss": 4.9682,
      "mean_token_accuracy": 0.22260719984769822,
      "num_tokens": 110616444.0,
      "step": 63770
    },
    {
      "entropy": 5.976556634902954,
      "epoch": 4.96191402450885,
      "grad_norm": 1.5,
      "learning_rate": 0.000280464976399759,
      "loss": 4.8952,
      "mean_token_accuracy": 0.22020761519670487,
      "num_tokens": 110624766.0,
      "step": 63775
    },
    {
      "entropy": 5.939637136459351,
      "epoch": 4.962303053880568,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000280437268947275,
      "loss": 4.8935,
      "mean_token_accuracy": 0.21726289242506028,
      "num_tokens": 110633197.0,
      "step": 63780
    },
    {
      "entropy": 5.903435707092285,
      "epoch": 4.962692083252286,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002804095614122891,
      "loss": 5.0021,
      "mean_token_accuracy": 0.21757557094097138,
      "num_tokens": 110641883.0,
      "step": 63785
    },
    {
      "entropy": 5.906747341156006,
      "epoch": 4.963081112624003,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00028038185379522176,
      "loss": 4.8889,
      "mean_token_accuracy": 0.22155261039733887,
      "num_tokens": 110650285.0,
      "step": 63790
    },
    {
      "entropy": 5.959514617919922,
      "epoch": 4.963470141995721,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028035414609649327,
      "loss": 4.9497,
      "mean_token_accuracy": 0.21106860488653184,
      "num_tokens": 110658418.0,
      "step": 63795
    },
    {
      "entropy": 5.954483222961426,
      "epoch": 4.963859171367439,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028032643831652415,
      "loss": 4.9752,
      "mean_token_accuracy": 0.21498009711503982,
      "num_tokens": 110667148.0,
      "step": 63800
    },
    {
      "entropy": 5.939287328720093,
      "epoch": 4.9642482007391555,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002802987304557349,
      "loss": 4.9828,
      "mean_token_accuracy": 0.2187413290143013,
      "num_tokens": 110675244.0,
      "step": 63805
    },
    {
      "entropy": 5.962346410751342,
      "epoch": 4.964637230110873,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002802710225145458,
      "loss": 4.8989,
      "mean_token_accuracy": 0.22940557301044465,
      "num_tokens": 110683201.0,
      "step": 63810
    },
    {
      "entropy": 5.964638900756836,
      "epoch": 4.965026259482591,
      "grad_norm": 1.515625,
      "learning_rate": 0.00028024331449337726,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20865247249603272,
      "num_tokens": 110692306.0,
      "step": 63815
    },
    {
      "entropy": 5.965317344665527,
      "epoch": 4.965415288854309,
      "grad_norm": 1.375,
      "learning_rate": 0.00028021560639264985,
      "loss": 5.0511,
      "mean_token_accuracy": 0.21114179342985154,
      "num_tokens": 110702483.0,
      "step": 63820
    },
    {
      "entropy": 6.059542036056518,
      "epoch": 4.965804318226026,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002801878982127838,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21155906468629837,
      "num_tokens": 110711426.0,
      "step": 63825
    },
    {
      "entropy": 6.022303104400635,
      "epoch": 4.966193347597744,
      "grad_norm": 1.796875,
      "learning_rate": 0.0002801601899541998,
      "loss": 4.9509,
      "mean_token_accuracy": 0.22528134435415267,
      "num_tokens": 110720368.0,
      "step": 63830
    },
    {
      "entropy": 6.010675621032715,
      "epoch": 4.966582376969461,
      "grad_norm": 1.4375,
      "learning_rate": 0.000280132481617318,
      "loss": 5.0271,
      "mean_token_accuracy": 0.21558926701545716,
      "num_tokens": 110728910.0,
      "step": 63835
    },
    {
      "entropy": 5.958311462402344,
      "epoch": 4.9669714063411785,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028010477320255904,
      "loss": 5.0236,
      "mean_token_accuracy": 0.20842650830745696,
      "num_tokens": 110738248.0,
      "step": 63840
    },
    {
      "entropy": 5.9994734764099125,
      "epoch": 4.967360435712896,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002800770647103433,
      "loss": 4.9974,
      "mean_token_accuracy": 0.21460507065057755,
      "num_tokens": 110747255.0,
      "step": 63845
    },
    {
      "entropy": 5.9364762783050535,
      "epoch": 4.967749465084614,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002800493561410911,
      "loss": 4.9538,
      "mean_token_accuracy": 0.22037217766046524,
      "num_tokens": 110756828.0,
      "step": 63850
    },
    {
      "entropy": 5.999535274505615,
      "epoch": 4.968138494456332,
      "grad_norm": 1.390625,
      "learning_rate": 0.000280021647495223,
      "loss": 5.0123,
      "mean_token_accuracy": 0.21505232751369477,
      "num_tokens": 110766364.0,
      "step": 63855
    },
    {
      "entropy": 5.987196636199951,
      "epoch": 4.968527523828049,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027999393877315944,
      "loss": 5.0471,
      "mean_token_accuracy": 0.2185453236103058,
      "num_tokens": 110774656.0,
      "step": 63860
    },
    {
      "entropy": 5.9618542194366455,
      "epoch": 4.968916553199767,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002799662299753207,
      "loss": 4.9512,
      "mean_token_accuracy": 0.21635394245386125,
      "num_tokens": 110783762.0,
      "step": 63865
    },
    {
      "entropy": 5.954960107803345,
      "epoch": 4.969305582571484,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027993852110212735,
      "loss": 4.9586,
      "mean_token_accuracy": 0.2182486593723297,
      "num_tokens": 110792978.0,
      "step": 63870
    },
    {
      "entropy": 6.029923963546753,
      "epoch": 4.9696946119432015,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002799108121539999,
      "loss": 5.0832,
      "mean_token_accuracy": 0.2027270182967186,
      "num_tokens": 110801572.0,
      "step": 63875
    },
    {
      "entropy": 5.916785717010498,
      "epoch": 4.970083641314919,
      "grad_norm": 1.546875,
      "learning_rate": 0.00027988310313135857,
      "loss": 4.8254,
      "mean_token_accuracy": 0.22330797016620635,
      "num_tokens": 110809362.0,
      "step": 63880
    },
    {
      "entropy": 5.9976640224456785,
      "epoch": 4.970472670686637,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00027985539403462407,
      "loss": 5.0228,
      "mean_token_accuracy": 0.217745079100132,
      "num_tokens": 110818078.0,
      "step": 63885
    },
    {
      "entropy": 5.872941970825195,
      "epoch": 4.970861700058355,
      "grad_norm": 1.59375,
      "learning_rate": 0.00027982768486421657,
      "loss": 4.8597,
      "mean_token_accuracy": 0.21133479326963425,
      "num_tokens": 110825622.0,
      "step": 63890
    },
    {
      "entropy": 5.8495588302612305,
      "epoch": 4.971250729430072,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027979997562055663,
      "loss": 4.9949,
      "mean_token_accuracy": 0.21648041009902955,
      "num_tokens": 110833985.0,
      "step": 63895
    },
    {
      "entropy": 5.91601710319519,
      "epoch": 4.971639758801789,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027977226630406476,
      "loss": 4.9324,
      "mean_token_accuracy": 0.22367036044597627,
      "num_tokens": 110842565.0,
      "step": 63900
    },
    {
      "entropy": 6.029841518402099,
      "epoch": 4.972028788173507,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002797445569151613,
      "loss": 5.0861,
      "mean_token_accuracy": 0.20379938185214996,
      "num_tokens": 110851463.0,
      "step": 63905
    },
    {
      "entropy": 6.018596506118774,
      "epoch": 4.9724178175452245,
      "grad_norm": 1.5,
      "learning_rate": 0.0002797168474542668,
      "loss": 5.0474,
      "mean_token_accuracy": 0.2038809299468994,
      "num_tokens": 110860317.0,
      "step": 63910
    },
    {
      "entropy": 6.0174633979797365,
      "epoch": 4.972806846916942,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027968913792180154,
      "loss": 5.0382,
      "mean_token_accuracy": 0.21298862844705582,
      "num_tokens": 110868940.0,
      "step": 63915
    },
    {
      "entropy": 5.960891675949097,
      "epoch": 4.97319587628866,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002796614283181862,
      "loss": 4.9265,
      "mean_token_accuracy": 0.2182511031627655,
      "num_tokens": 110877300.0,
      "step": 63920
    },
    {
      "entropy": 5.903309106826782,
      "epoch": 4.973584905660378,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000279633718643841,
      "loss": 4.9703,
      "mean_token_accuracy": 0.21575938910245895,
      "num_tokens": 110887016.0,
      "step": 63925
    },
    {
      "entropy": 6.043300533294678,
      "epoch": 4.973973935032095,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027960600889918657,
      "loss": 5.0762,
      "mean_token_accuracy": 0.20491232872009277,
      "num_tokens": 110897299.0,
      "step": 63930
    },
    {
      "entropy": 5.9703610897064205,
      "epoch": 4.974362964403813,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002795782990846433,
      "loss": 4.9768,
      "mean_token_accuracy": 0.2096459150314331,
      "num_tokens": 110906379.0,
      "step": 63935
    },
    {
      "entropy": 5.959058237075806,
      "epoch": 4.97475199377553,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002795505892006316,
      "loss": 4.9409,
      "mean_token_accuracy": 0.2128370836377144,
      "num_tokens": 110915007.0,
      "step": 63940
    },
    {
      "entropy": 6.022095680236816,
      "epoch": 4.9751410231472475,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002795228792475719,
      "loss": 5.073,
      "mean_token_accuracy": 0.20872470140457153,
      "num_tokens": 110923878.0,
      "step": 63945
    },
    {
      "entropy": 5.977816867828369,
      "epoch": 4.975530052518965,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002794951692258848,
      "loss": 5.0059,
      "mean_token_accuracy": 0.20403165221214295,
      "num_tokens": 110932262.0,
      "step": 63950
    },
    {
      "entropy": 5.940880346298218,
      "epoch": 4.975919081890683,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002794674591359906,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22647271752357484,
      "num_tokens": 110940494.0,
      "step": 63955
    },
    {
      "entropy": 5.891310548782348,
      "epoch": 4.976308111262401,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002794397489783098,
      "loss": 4.8407,
      "mean_token_accuracy": 0.22248669862747192,
      "num_tokens": 110949067.0,
      "step": 63960
    },
    {
      "entropy": 5.948409366607666,
      "epoch": 4.976697140634117,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002794120387532629,
      "loss": 4.9772,
      "mean_token_accuracy": 0.21739549934864044,
      "num_tokens": 110957799.0,
      "step": 63965
    },
    {
      "entropy": 5.959265279769897,
      "epoch": 4.977086170005835,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002793843284612703,
      "loss": 5.002,
      "mean_token_accuracy": 0.21712651550769807,
      "num_tokens": 110965887.0,
      "step": 63970
    },
    {
      "entropy": 6.031672859191895,
      "epoch": 4.977475199377553,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002793566181027526,
      "loss": 5.0892,
      "mean_token_accuracy": 0.20941001176834106,
      "num_tokens": 110974068.0,
      "step": 63975
    },
    {
      "entropy": 5.884252452850342,
      "epoch": 4.9778642287492705,
      "grad_norm": 1.5,
      "learning_rate": 0.00027932890767813005,
      "loss": 4.9071,
      "mean_token_accuracy": 0.21988178193569183,
      "num_tokens": 110982264.0,
      "step": 63980
    },
    {
      "entropy": 5.927508687973022,
      "epoch": 4.978253258120988,
      "grad_norm": 1.5,
      "learning_rate": 0.0002793011971878233,
      "loss": 4.9358,
      "mean_token_accuracy": 0.22059098929166793,
      "num_tokens": 110989882.0,
      "step": 63985
    },
    {
      "entropy": 5.961652231216431,
      "epoch": 4.978642287492706,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002792734866322526,
      "loss": 4.9383,
      "mean_token_accuracy": 0.21650320142507554,
      "num_tokens": 110998081.0,
      "step": 63990
    },
    {
      "entropy": 5.9721049785614015,
      "epoch": 4.979031316864424,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002792457760118387,
      "loss": 4.9907,
      "mean_token_accuracy": 0.2134132519364357,
      "num_tokens": 111007229.0,
      "step": 63995
    },
    {
      "entropy": 6.02068943977356,
      "epoch": 4.979420346236141,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027921806532700186,
      "loss": 4.9952,
      "mean_token_accuracy": 0.2170492649078369,
      "num_tokens": 111016097.0,
      "step": 64000
    },
    {
      "entropy": 6.007446002960205,
      "epoch": 4.979809375607858,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002791903545781625,
      "loss": 5.0088,
      "mean_token_accuracy": 0.2150631546974182,
      "num_tokens": 111024078.0,
      "step": 64005
    },
    {
      "entropy": 5.93270959854126,
      "epoch": 4.980198404979576,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027916264376574124,
      "loss": 4.9587,
      "mean_token_accuracy": 0.2165970116853714,
      "num_tokens": 111033214.0,
      "step": 64010
    },
    {
      "entropy": 5.965338230133057,
      "epoch": 4.9805874343512935,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027913493289015857,
      "loss": 4.9926,
      "mean_token_accuracy": 0.21466201841831206,
      "num_tokens": 111042307.0,
      "step": 64015
    },
    {
      "entropy": 5.999902772903442,
      "epoch": 4.980976463723011,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027910722195183474,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21665821820497513,
      "num_tokens": 111050490.0,
      "step": 64020
    },
    {
      "entropy": 5.940768194198609,
      "epoch": 4.981365493094729,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002790795109511904,
      "loss": 4.9545,
      "mean_token_accuracy": 0.21786734759807586,
      "num_tokens": 111059378.0,
      "step": 64025
    },
    {
      "entropy": 5.939095830917358,
      "epoch": 4.981754522466447,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00027905179988864593,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20884962677955626,
      "num_tokens": 111067157.0,
      "step": 64030
    },
    {
      "entropy": 6.009640836715699,
      "epoch": 4.982143551838163,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002790240887646219,
      "loss": 5.0224,
      "mean_token_accuracy": 0.21580564528703688,
      "num_tokens": 111076016.0,
      "step": 64035
    },
    {
      "entropy": 5.950236034393311,
      "epoch": 4.982532581209881,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00027899637757953876,
      "loss": 4.9087,
      "mean_token_accuracy": 0.22437747716903686,
      "num_tokens": 111084413.0,
      "step": 64040
    },
    {
      "entropy": 5.953349018096924,
      "epoch": 4.982921610581599,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00027896866633381694,
      "loss": 5.0252,
      "mean_token_accuracy": 0.21033919155597686,
      "num_tokens": 111092000.0,
      "step": 64045
    },
    {
      "entropy": 5.938661193847656,
      "epoch": 4.9833106399533165,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002789409550278769,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20722602158784867,
      "num_tokens": 111100918.0,
      "step": 64050
    },
    {
      "entropy": 5.931330919265747,
      "epoch": 4.983699669325034,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002789132436621391,
      "loss": 4.9851,
      "mean_token_accuracy": 0.22071402072906493,
      "num_tokens": 111109653.0,
      "step": 64055
    },
    {
      "entropy": 6.030182361602783,
      "epoch": 4.984088698696752,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002788855322370241,
      "loss": 4.9934,
      "mean_token_accuracy": 0.21908103227615355,
      "num_tokens": 111118101.0,
      "step": 64060
    },
    {
      "entropy": 5.99071249961853,
      "epoch": 4.98447772806847,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00027885782075295234,
      "loss": 4.9506,
      "mean_token_accuracy": 0.2212925747036934,
      "num_tokens": 111126552.0,
      "step": 64065
    },
    {
      "entropy": 5.983301210403442,
      "epoch": 4.984866757440186,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0002788301092103443,
      "loss": 5.0137,
      "mean_token_accuracy": 0.2152522921562195,
      "num_tokens": 111135285.0,
      "step": 64070
    },
    {
      "entropy": 5.866214752197266,
      "epoch": 4.985255786811904,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027880239760962044,
      "loss": 4.8684,
      "mean_token_accuracy": 0.22803619056940078,
      "num_tokens": 111144358.0,
      "step": 64075
    },
    {
      "entropy": 5.966524076461792,
      "epoch": 4.985644816183622,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00027877468595120125,
      "loss": 5.0675,
      "mean_token_accuracy": 0.20582932829856873,
      "num_tokens": 111152396.0,
      "step": 64080
    },
    {
      "entropy": 6.087909412384033,
      "epoch": 4.9860338455553395,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002787469742355072,
      "loss": 5.0396,
      "mean_token_accuracy": 0.2090340331196785,
      "num_tokens": 111161452.0,
      "step": 64085
    },
    {
      "entropy": 5.999317073822022,
      "epoch": 4.986422874927057,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002787192624629587,
      "loss": 5.0069,
      "mean_token_accuracy": 0.20891609638929368,
      "num_tokens": 111169773.0,
      "step": 64090
    },
    {
      "entropy": 5.982045888900757,
      "epoch": 4.986811904298775,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002786915506339764,
      "loss": 4.8924,
      "mean_token_accuracy": 0.21881015300750734,
      "num_tokens": 111178202.0,
      "step": 64095
    },
    {
      "entropy": 5.943727207183838,
      "epoch": 4.987200933670492,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002786638387489807,
      "loss": 5.0163,
      "mean_token_accuracy": 0.2100713849067688,
      "num_tokens": 111186384.0,
      "step": 64100
    },
    {
      "entropy": 5.960705757141113,
      "epoch": 4.987589963042209,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027863612680839204,
      "loss": 5.0156,
      "mean_token_accuracy": 0.21157492101192474,
      "num_tokens": 111195086.0,
      "step": 64105
    },
    {
      "entropy": 5.99471492767334,
      "epoch": 4.987978992413927,
      "grad_norm": 1.4375,
      "learning_rate": 0.000278608414812631,
      "loss": 5.009,
      "mean_token_accuracy": 0.21238764822483064,
      "num_tokens": 111203537.0,
      "step": 64110
    },
    {
      "entropy": 5.997534465789795,
      "epoch": 4.988368021785645,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027858070276211804,
      "loss": 4.9071,
      "mean_token_accuracy": 0.22279173135757446,
      "num_tokens": 111212518.0,
      "step": 64115
    },
    {
      "entropy": 6.032884407043457,
      "epoch": 4.9887570511573625,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002785529906572735,
      "loss": 5.0121,
      "mean_token_accuracy": 0.21396758556365966,
      "num_tokens": 111220747.0,
      "step": 64120
    },
    {
      "entropy": 6.016614723205566,
      "epoch": 4.98914608052908,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000278525278498518,
      "loss": 5.0516,
      "mean_token_accuracy": 0.20340076237916946,
      "num_tokens": 111229938.0,
      "step": 64125
    },
    {
      "entropy": 6.05537691116333,
      "epoch": 4.989535109900798,
      "grad_norm": 1.515625,
      "learning_rate": 0.00027849756628627205,
      "loss": 5.0159,
      "mean_token_accuracy": 0.20467608869075776,
      "num_tokens": 111237601.0,
      "step": 64130
    },
    {
      "entropy": 5.968055343627929,
      "epoch": 4.9899241392725155,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002784698540209561,
      "loss": 5.0144,
      "mean_token_accuracy": 0.20931281000375748,
      "num_tokens": 111245439.0,
      "step": 64135
    },
    {
      "entropy": 5.889980983734131,
      "epoch": 4.990313168644232,
      "grad_norm": 1.53125,
      "learning_rate": 0.00027844214170299065,
      "loss": 4.9325,
      "mean_token_accuracy": 0.22199618965387344,
      "num_tokens": 111254661.0,
      "step": 64140
    },
    {
      "entropy": 5.992573833465576,
      "epoch": 4.99070219801595,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027841442933279614,
      "loss": 4.9828,
      "mean_token_accuracy": 0.22415275126695633,
      "num_tokens": 111263038.0,
      "step": 64145
    },
    {
      "entropy": 5.907088136672973,
      "epoch": 4.991091227387668,
      "grad_norm": 1.5,
      "learning_rate": 0.00027838671691079314,
      "loss": 4.8612,
      "mean_token_accuracy": 0.22346449494361878,
      "num_tokens": 111270781.0,
      "step": 64150
    },
    {
      "entropy": 5.848456478118896,
      "epoch": 4.9914802567593854,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002783590044374021,
      "loss": 4.9659,
      "mean_token_accuracy": 0.212752628326416,
      "num_tokens": 111279555.0,
      "step": 64155
    },
    {
      "entropy": 5.9197510242462155,
      "epoch": 4.991869286131103,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002783312919130435,
      "loss": 4.8847,
      "mean_token_accuracy": 0.222040556371212,
      "num_tokens": 111288252.0,
      "step": 64160
    },
    {
      "entropy": 5.8765068531036375,
      "epoch": 4.992258315502821,
      "grad_norm": 1.375,
      "learning_rate": 0.0002783035793381378,
      "loss": 4.949,
      "mean_token_accuracy": 0.2156123548746109,
      "num_tokens": 111298293.0,
      "step": 64165
    },
    {
      "entropy": 6.02159423828125,
      "epoch": 4.992647344874538,
      "grad_norm": 1.5,
      "learning_rate": 0.00027827586671310556,
      "loss": 4.971,
      "mean_token_accuracy": 0.21450741291046144,
      "num_tokens": 111306408.0,
      "step": 64170
    },
    {
      "entropy": 5.933205080032349,
      "epoch": 4.993036374246255,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002782481540383673,
      "loss": 4.8883,
      "mean_token_accuracy": 0.22526173889636994,
      "num_tokens": 111315412.0,
      "step": 64175
    },
    {
      "entropy": 5.887360954284668,
      "epoch": 4.993425403617973,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027822044131434344,
      "loss": 4.9428,
      "mean_token_accuracy": 0.22040002942085266,
      "num_tokens": 111324331.0,
      "step": 64180
    },
    {
      "entropy": 5.914601182937622,
      "epoch": 4.993814432989691,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002781927285414545,
      "loss": 4.9825,
      "mean_token_accuracy": 0.21791246384382248,
      "num_tokens": 111333327.0,
      "step": 64185
    },
    {
      "entropy": 5.964279651641846,
      "epoch": 4.994203462361408,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027816501572012104,
      "loss": 4.9502,
      "mean_token_accuracy": 0.22117865979671478,
      "num_tokens": 111342191.0,
      "step": 64190
    },
    {
      "entropy": 5.9876556396484375,
      "epoch": 4.994592491733126,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002781373028507634,
      "loss": 4.9998,
      "mean_token_accuracy": 0.2147589221596718,
      "num_tokens": 111350647.0,
      "step": 64195
    },
    {
      "entropy": 5.903028869628907,
      "epoch": 4.994981521104844,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002781095899338023,
      "loss": 4.868,
      "mean_token_accuracy": 0.22256578654050826,
      "num_tokens": 111358745.0,
      "step": 64200
    },
    {
      "entropy": 5.978062677383423,
      "epoch": 4.995370550476561,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002780818769696581,
      "loss": 5.0254,
      "mean_token_accuracy": 0.20791832804679872,
      "num_tokens": 111368369.0,
      "step": 64205
    },
    {
      "entropy": 5.975418567657471,
      "epoch": 4.995759579848278,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027805416395875127,
      "loss": 4.9756,
      "mean_token_accuracy": 0.21357603669166564,
      "num_tokens": 111377637.0,
      "step": 64210
    },
    {
      "entropy": 6.00588583946228,
      "epoch": 4.996148609219996,
      "grad_norm": 1.5,
      "learning_rate": 0.0002780264509015024,
      "loss": 4.9776,
      "mean_token_accuracy": 0.213634891808033,
      "num_tokens": 111386016.0,
      "step": 64215
    },
    {
      "entropy": 5.966055488586425,
      "epoch": 4.996537638591714,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027799873779833193,
      "loss": 4.9536,
      "mean_token_accuracy": 0.22317275404930115,
      "num_tokens": 111394811.0,
      "step": 64220
    },
    {
      "entropy": 5.986238622665406,
      "epoch": 4.996926667963431,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002779710246496604,
      "loss": 4.9753,
      "mean_token_accuracy": 0.2201543465256691,
      "num_tokens": 111403842.0,
      "step": 64225
    },
    {
      "entropy": 5.949134683609008,
      "epoch": 4.997315697335149,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002779433114559083,
      "loss": 4.9426,
      "mean_token_accuracy": 0.222234508395195,
      "num_tokens": 111412164.0,
      "step": 64230
    },
    {
      "entropy": 5.928938198089599,
      "epoch": 4.997704726706866,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027791559821749615,
      "loss": 4.9462,
      "mean_token_accuracy": 0.21307792216539384,
      "num_tokens": 111421202.0,
      "step": 64235
    },
    {
      "entropy": 5.964136028289795,
      "epoch": 4.998093756078584,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027788788493484447,
      "loss": 4.8999,
      "mean_token_accuracy": 0.22490315288305282,
      "num_tokens": 111429531.0,
      "step": 64240
    },
    {
      "entropy": 6.036328649520874,
      "epoch": 4.998482785450301,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002778601716083736,
      "loss": 4.9645,
      "mean_token_accuracy": 0.21246591359376907,
      "num_tokens": 111438090.0,
      "step": 64245
    },
    {
      "entropy": 5.958028554916382,
      "epoch": 4.998871814822019,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002778324582385043,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20755317211151122,
      "num_tokens": 111447011.0,
      "step": 64250
    },
    {
      "entropy": 5.918970775604248,
      "epoch": 4.999260844193737,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00027780474482565687,
      "loss": 4.9621,
      "mean_token_accuracy": 0.2196969985961914,
      "num_tokens": 111455486.0,
      "step": 64255
    },
    {
      "entropy": 5.94116735458374,
      "epoch": 4.999649873565454,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027777703137025193,
      "loss": 4.9605,
      "mean_token_accuracy": 0.21686898171901703,
      "num_tokens": 111464053.0,
      "step": 64260
    },
    {
      "entropy": 5.938394069671631,
      "epoch": 5.0,
      "grad_norm": 2.953125,
      "learning_rate": 0.00027774931787271,
      "loss": 4.9299,
      "mean_token_accuracy": 0.2189635154273775,
      "num_tokens": 111471835.0,
      "step": 64265
    },
    {
      "entropy": 5.9992913722991945,
      "epoch": 5.000389029371718,
      "grad_norm": 1.375,
      "learning_rate": 0.0002777216043334516,
      "loss": 4.9992,
      "mean_token_accuracy": 0.21412764936685563,
      "num_tokens": 111480800.0,
      "step": 64270
    },
    {
      "entropy": 5.915005874633789,
      "epoch": 5.000778058743435,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027769389075289714,
      "loss": 4.9261,
      "mean_token_accuracy": 0.2255018264055252,
      "num_tokens": 111490128.0,
      "step": 64275
    },
    {
      "entropy": 5.963856554031372,
      "epoch": 5.001167088115153,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002776661771314671,
      "loss": 4.9199,
      "mean_token_accuracy": 0.22687022536993026,
      "num_tokens": 111498196.0,
      "step": 64280
    },
    {
      "entropy": 5.948149299621582,
      "epoch": 5.00155611748687,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002776384634695821,
      "loss": 4.9102,
      "mean_token_accuracy": 0.22363985627889632,
      "num_tokens": 111505892.0,
      "step": 64285
    },
    {
      "entropy": 6.00820779800415,
      "epoch": 5.001945146858588,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002776107497676626,
      "loss": 5.0162,
      "mean_token_accuracy": 0.21780850291252135,
      "num_tokens": 111514945.0,
      "step": 64290
    },
    {
      "entropy": 5.922317600250244,
      "epoch": 5.002334176230305,
      "grad_norm": 1.375,
      "learning_rate": 0.0002775830360261291,
      "loss": 4.8823,
      "mean_token_accuracy": 0.21582133024930955,
      "num_tokens": 111523113.0,
      "step": 64295
    },
    {
      "entropy": 5.89165768623352,
      "epoch": 5.002723205602023,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027755532224540226,
      "loss": 4.8797,
      "mean_token_accuracy": 0.2207957312464714,
      "num_tokens": 111531396.0,
      "step": 64300
    },
    {
      "entropy": 5.9533556461334225,
      "epoch": 5.003112234973741,
      "grad_norm": 1.640625,
      "learning_rate": 0.00027752760842590245,
      "loss": 5.034,
      "mean_token_accuracy": 0.20952605158090593,
      "num_tokens": 111539600.0,
      "step": 64305
    },
    {
      "entropy": 5.912472677230835,
      "epoch": 5.003501264345458,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002774998945680502,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22143614739179612,
      "num_tokens": 111547837.0,
      "step": 64310
    },
    {
      "entropy": 5.9206352710723875,
      "epoch": 5.003890293717176,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000277472180672266,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21412252336740495,
      "num_tokens": 111556914.0,
      "step": 64315
    },
    {
      "entropy": 5.991730260848999,
      "epoch": 5.004279323088893,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00027744446673897034,
      "loss": 5.0099,
      "mean_token_accuracy": 0.2173427537083626,
      "num_tokens": 111566373.0,
      "step": 64320
    },
    {
      "entropy": 5.916161060333252,
      "epoch": 5.004668352460611,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00027741675276858374,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2203942820429802,
      "num_tokens": 111575155.0,
      "step": 64325
    },
    {
      "entropy": 6.028845357894897,
      "epoch": 5.005057381832328,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00027738903876152693,
      "loss": 5.0101,
      "mean_token_accuracy": 0.21571316868066787,
      "num_tokens": 111584448.0,
      "step": 64330
    },
    {
      "entropy": 5.901907682418823,
      "epoch": 5.005446411204046,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002773613247182202,
      "loss": 4.8943,
      "mean_token_accuracy": 0.2256663054227829,
      "num_tokens": 111592761.0,
      "step": 64335
    },
    {
      "entropy": 5.941567611694336,
      "epoch": 5.005835440575764,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00027733361063908405,
      "loss": 4.8848,
      "mean_token_accuracy": 0.22712364941835403,
      "num_tokens": 111600591.0,
      "step": 64340
    },
    {
      "entropy": 5.8213623523712155,
      "epoch": 5.006224469947481,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002773058965245393,
      "loss": 4.8417,
      "mean_token_accuracy": 0.22544692158699037,
      "num_tokens": 111609089.0,
      "step": 64345
    },
    {
      "entropy": 5.825745868682861,
      "epoch": 5.006613499319198,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000277278182375006,
      "loss": 4.8483,
      "mean_token_accuracy": 0.2243263989686966,
      "num_tokens": 111617211.0,
      "step": 64350
    },
    {
      "entropy": 5.9841516494750975,
      "epoch": 5.007002528690916,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002772504681909049,
      "loss": 4.9988,
      "mean_token_accuracy": 0.2145882233977318,
      "num_tokens": 111625562.0,
      "step": 64355
    },
    {
      "entropy": 5.963325500488281,
      "epoch": 5.007391558062634,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002772227539726566,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2157747268676758,
      "num_tokens": 111634095.0,
      "step": 64360
    },
    {
      "entropy": 5.9477873802185055,
      "epoch": 5.007780587434351,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002771950397206816,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2181429848074913,
      "num_tokens": 111642386.0,
      "step": 64365
    },
    {
      "entropy": 5.895689630508423,
      "epoch": 5.008169616806069,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002771673254354003,
      "loss": 4.8883,
      "mean_token_accuracy": 0.22305791825056076,
      "num_tokens": 111650561.0,
      "step": 64370
    },
    {
      "entropy": 5.9402069568634035,
      "epoch": 5.008558646177787,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027713961111723335,
      "loss": 4.9836,
      "mean_token_accuracy": 0.21357118636369704,
      "num_tokens": 111659264.0,
      "step": 64375
    },
    {
      "entropy": 5.942184686660767,
      "epoch": 5.008947675549504,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027711189676660116,
      "loss": 4.9512,
      "mean_token_accuracy": 0.2165623798966408,
      "num_tokens": 111667254.0,
      "step": 64380
    },
    {
      "entropy": 6.023373460769653,
      "epoch": 5.009336704921221,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027708418238392435,
      "loss": 5.0173,
      "mean_token_accuracy": 0.21320536732673645,
      "num_tokens": 111676160.0,
      "step": 64385
    },
    {
      "entropy": 5.897787237167359,
      "epoch": 5.009725734292939,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002770564679696233,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22351074814796448,
      "num_tokens": 111685109.0,
      "step": 64390
    },
    {
      "entropy": 5.878963375091553,
      "epoch": 5.010114763664657,
      "grad_norm": 1.5,
      "learning_rate": 0.00027702875352411866,
      "loss": 4.8196,
      "mean_token_accuracy": 0.22820911705493926,
      "num_tokens": 111694949.0,
      "step": 64395
    },
    {
      "entropy": 5.997205114364624,
      "epoch": 5.010503793036374,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002770010390478309,
      "loss": 4.9989,
      "mean_token_accuracy": 0.22195437997579576,
      "num_tokens": 111704141.0,
      "step": 64400
    },
    {
      "entropy": 5.954021453857422,
      "epoch": 5.010892822408092,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027697332454118057,
      "loss": 4.9566,
      "mean_token_accuracy": 0.21605437397956848,
      "num_tokens": 111713776.0,
      "step": 64405
    },
    {
      "entropy": 5.911357259750366,
      "epoch": 5.01128185177981,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027694561000458825,
      "loss": 4.8786,
      "mean_token_accuracy": 0.22575565576553344,
      "num_tokens": 111721914.0,
      "step": 64410
    },
    {
      "entropy": 5.966030693054199,
      "epoch": 5.011670881151527,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002769178954384743,
      "loss": 4.9869,
      "mean_token_accuracy": 0.2218691036105156,
      "num_tokens": 111730835.0,
      "step": 64415
    },
    {
      "entropy": 5.966979026794434,
      "epoch": 5.012059910523244,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002768901808432594,
      "loss": 4.9947,
      "mean_token_accuracy": 0.21712421774864196,
      "num_tokens": 111739621.0,
      "step": 64420
    },
    {
      "entropy": 5.936514854431152,
      "epoch": 5.012448939894962,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000276862466219364,
      "loss": 4.9595,
      "mean_token_accuracy": 0.22356475591659547,
      "num_tokens": 111748818.0,
      "step": 64425
    },
    {
      "entropy": 6.013345718383789,
      "epoch": 5.01283796926668,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002768347515672086,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21295957416296005,
      "num_tokens": 111757234.0,
      "step": 64430
    },
    {
      "entropy": 5.936593341827392,
      "epoch": 5.013226998638397,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00027680703688721383,
      "loss": 4.8315,
      "mean_token_accuracy": 0.22767816334962845,
      "num_tokens": 111765991.0,
      "step": 64435
    },
    {
      "entropy": 5.885225915908814,
      "epoch": 5.013616028010115,
      "grad_norm": 1.5,
      "learning_rate": 0.00027677932217980004,
      "loss": 4.8883,
      "mean_token_accuracy": 0.22330691069364547,
      "num_tokens": 111774752.0,
      "step": 64440
    },
    {
      "entropy": 5.982660007476807,
      "epoch": 5.014005057381833,
      "grad_norm": 1.46875,
      "learning_rate": 0.000276751607445388,
      "loss": 5.0153,
      "mean_token_accuracy": 0.21027986258268355,
      "num_tokens": 111783433.0,
      "step": 64445
    },
    {
      "entropy": 5.961589574813843,
      "epoch": 5.0143940867535495,
      "grad_norm": 1.390625,
      "learning_rate": 0.000276723892684398,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2248132646083832,
      "num_tokens": 111793284.0,
      "step": 64450
    },
    {
      "entropy": 5.982715511322022,
      "epoch": 5.014783116125267,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00027669617789725074,
      "loss": 4.9588,
      "mean_token_accuracy": 0.21696071326732635,
      "num_tokens": 111801079.0,
      "step": 64455
    },
    {
      "entropy": 5.870068883895874,
      "epoch": 5.015172145496985,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002766684630843667,
      "loss": 4.8416,
      "mean_token_accuracy": 0.231181101500988,
      "num_tokens": 111809832.0,
      "step": 64460
    },
    {
      "entropy": 5.954134845733643,
      "epoch": 5.015561174868703,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002766407482461663,
      "loss": 5.0056,
      "mean_token_accuracy": 0.21122754365205765,
      "num_tokens": 111817846.0,
      "step": 64465
    },
    {
      "entropy": 5.952762031555176,
      "epoch": 5.01595020424042,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002766130333830702,
      "loss": 4.9416,
      "mean_token_accuracy": 0.21512737721204758,
      "num_tokens": 111826340.0,
      "step": 64470
    },
    {
      "entropy": 5.8644458770751955,
      "epoch": 5.016339233612138,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027658531849549894,
      "loss": 4.8634,
      "mean_token_accuracy": 0.21851956248283386,
      "num_tokens": 111834965.0,
      "step": 64475
    },
    {
      "entropy": 6.00278263092041,
      "epoch": 5.016728262983856,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002765576035838729,
      "loss": 4.959,
      "mean_token_accuracy": 0.21304924488067628,
      "num_tokens": 111843809.0,
      "step": 64480
    },
    {
      "entropy": 5.941182851791382,
      "epoch": 5.0171172923555725,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002765298886486128,
      "loss": 4.9299,
      "mean_token_accuracy": 0.22365241646766662,
      "num_tokens": 111853102.0,
      "step": 64485
    },
    {
      "entropy": 5.945188713073731,
      "epoch": 5.01750632172729,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002765021736901389,
      "loss": 4.9546,
      "mean_token_accuracy": 0.21909375339746476,
      "num_tokens": 111861909.0,
      "step": 64490
    },
    {
      "entropy": 5.913717412948609,
      "epoch": 5.017895351099008,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002764744587088721,
      "loss": 4.8783,
      "mean_token_accuracy": 0.22588928043842316,
      "num_tokens": 111870253.0,
      "step": 64495
    },
    {
      "entropy": 5.955865287780762,
      "epoch": 5.018284380470726,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002764467437052327,
      "loss": 4.8727,
      "mean_token_accuracy": 0.22777946442365646,
      "num_tokens": 111878295.0,
      "step": 64500
    },
    {
      "entropy": 5.950929117202759,
      "epoch": 5.018673409842443,
      "grad_norm": 1.5625,
      "learning_rate": 0.00027641902867964126,
      "loss": 4.9397,
      "mean_token_accuracy": 0.22387677878141404,
      "num_tokens": 111886684.0,
      "step": 64505
    },
    {
      "entropy": 5.900404453277588,
      "epoch": 5.019062439214161,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027639131363251825,
      "loss": 4.9023,
      "mean_token_accuracy": 0.22259812504053117,
      "num_tokens": 111895746.0,
      "step": 64510
    },
    {
      "entropy": 5.938545799255371,
      "epoch": 5.019451468585879,
      "grad_norm": 1.609375,
      "learning_rate": 0.00027636359856428436,
      "loss": 4.9136,
      "mean_token_accuracy": 0.21792158335447312,
      "num_tokens": 111903662.0,
      "step": 64515
    },
    {
      "entropy": 5.93454875946045,
      "epoch": 5.0198404979575955,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002763358834753599,
      "loss": 4.9697,
      "mean_token_accuracy": 0.216750305891037,
      "num_tokens": 111912990.0,
      "step": 64520
    },
    {
      "entropy": 5.984649133682251,
      "epoch": 5.020229527329313,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00027630816836616564,
      "loss": 4.9522,
      "mean_token_accuracy": 0.2210889220237732,
      "num_tokens": 111921417.0,
      "step": 64525
    },
    {
      "entropy": 5.9259730815887455,
      "epoch": 5.020618556701031,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00027628045323712194,
      "loss": 4.8637,
      "mean_token_accuracy": 0.22165046036243438,
      "num_tokens": 111930625.0,
      "step": 64530
    },
    {
      "entropy": 5.953272438049316,
      "epoch": 5.0210075860727486,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027625273808864955,
      "loss": 4.9263,
      "mean_token_accuracy": 0.2228071391582489,
      "num_tokens": 111939962.0,
      "step": 64535
    },
    {
      "entropy": 5.984951448440552,
      "epoch": 5.021396615444466,
      "grad_norm": 1.5,
      "learning_rate": 0.0002762250229211687,
      "loss": 5.0491,
      "mean_token_accuracy": 0.2120636761188507,
      "num_tokens": 111948461.0,
      "step": 64540
    },
    {
      "entropy": 5.932856512069702,
      "epoch": 5.021785644816184,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027619730773510015,
      "loss": 4.9265,
      "mean_token_accuracy": 0.22442819476127623,
      "num_tokens": 111956806.0,
      "step": 64545
    },
    {
      "entropy": 5.996646881103516,
      "epoch": 5.022174674187901,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027616959253086437,
      "loss": 5.0321,
      "mean_token_accuracy": 0.21168937236070634,
      "num_tokens": 111965947.0,
      "step": 64550
    },
    {
      "entropy": 5.965319728851318,
      "epoch": 5.0225637035596185,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002761418773088818,
      "loss": 4.9307,
      "mean_token_accuracy": 0.2172708258032799,
      "num_tokens": 111974729.0,
      "step": 64555
    },
    {
      "entropy": 5.999999570846557,
      "epoch": 5.022952732931336,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002761141620695731,
      "loss": 4.9736,
      "mean_token_accuracy": 0.2131114736199379,
      "num_tokens": 111983010.0,
      "step": 64560
    },
    {
      "entropy": 5.8825568675994875,
      "epoch": 5.023341762303054,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00027608644681335876,
      "loss": 4.8295,
      "mean_token_accuracy": 0.229107004404068,
      "num_tokens": 111990739.0,
      "step": 64565
    },
    {
      "entropy": 5.853581476211548,
      "epoch": 5.0237307916747715,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027605873154065935,
      "loss": 4.7885,
      "mean_token_accuracy": 0.23045054078102112,
      "num_tokens": 111998532.0,
      "step": 64570
    },
    {
      "entropy": 5.801084804534912,
      "epoch": 5.024119821046489,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027603101625189543,
      "loss": 4.7826,
      "mean_token_accuracy": 0.22840965837240218,
      "num_tokens": 112008592.0,
      "step": 64575
    },
    {
      "entropy": 5.948647117614746,
      "epoch": 5.024508850418207,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002760033009474874,
      "loss": 4.918,
      "mean_token_accuracy": 0.22756849527359008,
      "num_tokens": 112016700.0,
      "step": 64580
    },
    {
      "entropy": 6.027149677276611,
      "epoch": 5.024897879789924,
      "grad_norm": 1.5,
      "learning_rate": 0.0002759755856278558,
      "loss": 5.0566,
      "mean_token_accuracy": 0.20929763913154603,
      "num_tokens": 112026129.0,
      "step": 64585
    },
    {
      "entropy": 5.993504428863526,
      "epoch": 5.0252869091616414,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002759478702934213,
      "loss": 5.0431,
      "mean_token_accuracy": 0.20621503591537477,
      "num_tokens": 112035784.0,
      "step": 64590
    },
    {
      "entropy": 5.951185703277588,
      "epoch": 5.025675938533359,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027592015494460443,
      "loss": 4.8844,
      "mean_token_accuracy": 0.22701186686754227,
      "num_tokens": 112044901.0,
      "step": 64595
    },
    {
      "entropy": 5.981384134292602,
      "epoch": 5.026064967905077,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027589243958182566,
      "loss": 4.9562,
      "mean_token_accuracy": 0.21973519772291183,
      "num_tokens": 112054608.0,
      "step": 64600
    },
    {
      "entropy": 5.995902156829834,
      "epoch": 5.0264539972767945,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027586472420550553,
      "loss": 4.9001,
      "mean_token_accuracy": 0.2242453619837761,
      "num_tokens": 112063305.0,
      "step": 64605
    },
    {
      "entropy": 5.943789100646972,
      "epoch": 5.026843026648512,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002758370088160646,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21922565698623658,
      "num_tokens": 112071953.0,
      "step": 64610
    },
    {
      "entropy": 5.910513639450073,
      "epoch": 5.02723205602023,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002758092934139234,
      "loss": 4.9037,
      "mean_token_accuracy": 0.2183284655213356,
      "num_tokens": 112080912.0,
      "step": 64615
    },
    {
      "entropy": 5.999034309387207,
      "epoch": 5.027621085391947,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002757815779995023,
      "loss": 4.9346,
      "mean_token_accuracy": 0.21927718222141265,
      "num_tokens": 112089670.0,
      "step": 64620
    },
    {
      "entropy": 5.9748927593231205,
      "epoch": 5.028010114763664,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002757538625732222,
      "loss": 4.9864,
      "mean_token_accuracy": 0.2168762430548668,
      "num_tokens": 112098534.0,
      "step": 64625
    },
    {
      "entropy": 5.889771461486816,
      "epoch": 5.028399144135382,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002757261471355034,
      "loss": 4.8314,
      "mean_token_accuracy": 0.22442008256912233,
      "num_tokens": 112107494.0,
      "step": 64630
    },
    {
      "entropy": 5.944204473495484,
      "epoch": 5.0287881735071,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027569843168676647,
      "loss": 4.9554,
      "mean_token_accuracy": 0.2106885641813278,
      "num_tokens": 112115831.0,
      "step": 64635
    },
    {
      "entropy": 5.932636070251465,
      "epoch": 5.0291772028788175,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002756707162274319,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22630784064531326,
      "num_tokens": 112124683.0,
      "step": 64640
    },
    {
      "entropy": 5.920764970779419,
      "epoch": 5.029566232250535,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00027564300075792026,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22869690060615538,
      "num_tokens": 112133221.0,
      "step": 64645
    },
    {
      "entropy": 5.90343542098999,
      "epoch": 5.029955261622252,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027561528527865215,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21364986449480056,
      "num_tokens": 112142295.0,
      "step": 64650
    },
    {
      "entropy": 5.931903982162476,
      "epoch": 5.03034429099397,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002755875697900481,
      "loss": 4.9117,
      "mean_token_accuracy": 0.2198319599032402,
      "num_tokens": 112150959.0,
      "step": 64655
    },
    {
      "entropy": 5.912628984451294,
      "epoch": 5.030733320365687,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027555985429252857,
      "loss": 4.8281,
      "mean_token_accuracy": 0.22266353368759156,
      "num_tokens": 112159314.0,
      "step": 64660
    },
    {
      "entropy": 5.945159196853638,
      "epoch": 5.031122349737405,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002755321387865141,
      "loss": 4.9523,
      "mean_token_accuracy": 0.2211337774991989,
      "num_tokens": 112168010.0,
      "step": 64665
    },
    {
      "entropy": 5.969783306121826,
      "epoch": 5.031511379109123,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002755044232724253,
      "loss": 4.8866,
      "mean_token_accuracy": 0.22506098002195357,
      "num_tokens": 112175743.0,
      "step": 64670
    },
    {
      "entropy": 6.0086658000946045,
      "epoch": 5.0319004084808405,
      "grad_norm": 1.578125,
      "learning_rate": 0.00027547670775068274,
      "loss": 4.9966,
      "mean_token_accuracy": 0.2057319313287735,
      "num_tokens": 112184148.0,
      "step": 64675
    },
    {
      "entropy": 6.047475910186767,
      "epoch": 5.032289437852558,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027544899222170684,
      "loss": 4.985,
      "mean_token_accuracy": 0.2184407204389572,
      "num_tokens": 112193534.0,
      "step": 64680
    },
    {
      "entropy": 5.954745149612426,
      "epoch": 5.032678467224275,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027542127668591824,
      "loss": 4.9295,
      "mean_token_accuracy": 0.225139619410038,
      "num_tokens": 112202007.0,
      "step": 64685
    },
    {
      "entropy": 5.868213796615601,
      "epoch": 5.033067496595993,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002753935611437374,
      "loss": 4.8409,
      "mean_token_accuracy": 0.2285771444439888,
      "num_tokens": 112210305.0,
      "step": 64690
    },
    {
      "entropy": 5.961578464508056,
      "epoch": 5.03345652596771,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027536584559558494,
      "loss": 4.9254,
      "mean_token_accuracy": 0.21526242941617965,
      "num_tokens": 112218744.0,
      "step": 64695
    },
    {
      "entropy": 5.919012498855591,
      "epoch": 5.033845555339428,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002753381300418813,
      "loss": 4.9252,
      "mean_token_accuracy": 0.21993912607431412,
      "num_tokens": 112227807.0,
      "step": 64700
    },
    {
      "entropy": 5.908535432815552,
      "epoch": 5.034234584711146,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002753104144830471,
      "loss": 4.9588,
      "mean_token_accuracy": 0.21832275688648223,
      "num_tokens": 112236668.0,
      "step": 64705
    },
    {
      "entropy": 5.988567066192627,
      "epoch": 5.0346236140828635,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027528269891950285,
      "loss": 5.0206,
      "mean_token_accuracy": 0.2114863485097885,
      "num_tokens": 112245190.0,
      "step": 64710
    },
    {
      "entropy": 5.9327507495880125,
      "epoch": 5.035012643454581,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002752549833516691,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21545108109712602,
      "num_tokens": 112253558.0,
      "step": 64715
    },
    {
      "entropy": 5.911591625213623,
      "epoch": 5.035401672826298,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002752272677799664,
      "loss": 4.8536,
      "mean_token_accuracy": 0.2289659634232521,
      "num_tokens": 112262669.0,
      "step": 64720
    },
    {
      "entropy": 5.903499937057495,
      "epoch": 5.035790702198016,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00027519955220481525,
      "loss": 4.8807,
      "mean_token_accuracy": 0.2215979963541031,
      "num_tokens": 112270750.0,
      "step": 64725
    },
    {
      "entropy": 5.980356645584107,
      "epoch": 5.036179731569733,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002751718366266362,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21776337474584578,
      "num_tokens": 112279317.0,
      "step": 64730
    },
    {
      "entropy": 5.919329214096069,
      "epoch": 5.036568760941451,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002751441210458498,
      "loss": 4.9642,
      "mean_token_accuracy": 0.22425423115491866,
      "num_tokens": 112288343.0,
      "step": 64735
    },
    {
      "entropy": 5.893426370620728,
      "epoch": 5.036957790313169,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027511640546287665,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22952343970537187,
      "num_tokens": 112296651.0,
      "step": 64740
    },
    {
      "entropy": 5.920495080947876,
      "epoch": 5.0373468196848865,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002750886898781372,
      "loss": 4.8849,
      "mean_token_accuracy": 0.23480149060487748,
      "num_tokens": 112305716.0,
      "step": 64745
    },
    {
      "entropy": 5.84849066734314,
      "epoch": 5.037735849056604,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000275060974292052,
      "loss": 4.7411,
      "mean_token_accuracy": 0.2382361650466919,
      "num_tokens": 112314250.0,
      "step": 64750
    },
    {
      "entropy": 5.935089826583862,
      "epoch": 5.038124878428321,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002750332587050416,
      "loss": 4.9653,
      "mean_token_accuracy": 0.21852075308561325,
      "num_tokens": 112322621.0,
      "step": 64755
    },
    {
      "entropy": 5.932641696929932,
      "epoch": 5.038513907800039,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002750055431175266,
      "loss": 4.9424,
      "mean_token_accuracy": 0.21677245199680328,
      "num_tokens": 112331471.0,
      "step": 64760
    },
    {
      "entropy": 6.006406259536743,
      "epoch": 5.038902937171756,
      "grad_norm": 1.53125,
      "learning_rate": 0.00027497782752992743,
      "loss": 4.9709,
      "mean_token_accuracy": 0.21237770318984986,
      "num_tokens": 112339686.0,
      "step": 64765
    },
    {
      "entropy": 5.902475452423095,
      "epoch": 5.039291966543474,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002749501119426647,
      "loss": 4.8457,
      "mean_token_accuracy": 0.22895268797874452,
      "num_tokens": 112347926.0,
      "step": 64770
    },
    {
      "entropy": 5.895757055282592,
      "epoch": 5.039680995915192,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027492239635615903,
      "loss": 4.9725,
      "mean_token_accuracy": 0.2224930316209793,
      "num_tokens": 112357051.0,
      "step": 64775
    },
    {
      "entropy": 5.859316635131836,
      "epoch": 5.0400700252869095,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002748946807708308,
      "loss": 4.8941,
      "mean_token_accuracy": 0.2247803807258606,
      "num_tokens": 112366330.0,
      "step": 64780
    },
    {
      "entropy": 5.984780359268188,
      "epoch": 5.040459054658626,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002748669651871006,
      "loss": 5.0118,
      "mean_token_accuracy": 0.21367742419242858,
      "num_tokens": 112375198.0,
      "step": 64785
    },
    {
      "entropy": 5.960772371292114,
      "epoch": 5.040848084030344,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027483924960538907,
      "loss": 4.8844,
      "mean_token_accuracy": 0.21841208189725875,
      "num_tokens": 112384176.0,
      "step": 64790
    },
    {
      "entropy": 5.994722843170166,
      "epoch": 5.041237113402062,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027481153402611664,
      "loss": 4.9905,
      "mean_token_accuracy": 0.21865815073251724,
      "num_tokens": 112392861.0,
      "step": 64795
    },
    {
      "entropy": 5.956846475601196,
      "epoch": 5.041626142773779,
      "grad_norm": 1.515625,
      "learning_rate": 0.00027478381844970386,
      "loss": 4.952,
      "mean_token_accuracy": 0.21239129304885865,
      "num_tokens": 112402198.0,
      "step": 64800
    },
    {
      "entropy": 5.922763681411743,
      "epoch": 5.042015172145497,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002747561028765713,
      "loss": 4.8689,
      "mean_token_accuracy": 0.22311809659004211,
      "num_tokens": 112410175.0,
      "step": 64805
    },
    {
      "entropy": 5.939531326293945,
      "epoch": 5.042404201517215,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027472838730713956,
      "loss": 4.9861,
      "mean_token_accuracy": 0.21517639011144638,
      "num_tokens": 112418955.0,
      "step": 64810
    },
    {
      "entropy": 5.83917202949524,
      "epoch": 5.0427932308889325,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027470067174182905,
      "loss": 4.8501,
      "mean_token_accuracy": 0.22274995744228362,
      "num_tokens": 112427794.0,
      "step": 64815
    },
    {
      "entropy": 6.0189189434051515,
      "epoch": 5.043182260260649,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002746729561810604,
      "loss": 5.0025,
      "mean_token_accuracy": 0.21766810715198517,
      "num_tokens": 112437156.0,
      "step": 64820
    },
    {
      "entropy": 5.923266887664795,
      "epoch": 5.043571289632367,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027464524062525405,
      "loss": 4.8653,
      "mean_token_accuracy": 0.22640130966901778,
      "num_tokens": 112446082.0,
      "step": 64825
    },
    {
      "entropy": 6.014674234390259,
      "epoch": 5.043960319004085,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002746175250748306,
      "loss": 5.0424,
      "mean_token_accuracy": 0.2133150354027748,
      "num_tokens": 112454983.0,
      "step": 64830
    },
    {
      "entropy": 6.002985334396362,
      "epoch": 5.044349348375802,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002745898095302106,
      "loss": 5.0392,
      "mean_token_accuracy": 0.21668250858783722,
      "num_tokens": 112464933.0,
      "step": 64835
    },
    {
      "entropy": 6.016051959991455,
      "epoch": 5.04473837774752,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027456209399181466,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21497328579425812,
      "num_tokens": 112473834.0,
      "step": 64840
    },
    {
      "entropy": 5.960535001754761,
      "epoch": 5.045127407119238,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027453437846006325,
      "loss": 4.8884,
      "mean_token_accuracy": 0.22180876582860948,
      "num_tokens": 112482113.0,
      "step": 64845
    },
    {
      "entropy": 5.976833534240723,
      "epoch": 5.0455164364909555,
      "grad_norm": 1.578125,
      "learning_rate": 0.00027450666293537683,
      "loss": 4.931,
      "mean_token_accuracy": 0.2208910256624222,
      "num_tokens": 112490425.0,
      "step": 64850
    },
    {
      "entropy": 5.985898685455322,
      "epoch": 5.045905465862672,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000274478947418176,
      "loss": 5.0334,
      "mean_token_accuracy": 0.2147115707397461,
      "num_tokens": 112499456.0,
      "step": 64855
    },
    {
      "entropy": 5.955044841766357,
      "epoch": 5.04629449523439,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002744512319088814,
      "loss": 4.8936,
      "mean_token_accuracy": 0.22036911100149154,
      "num_tokens": 112508692.0,
      "step": 64860
    },
    {
      "entropy": 5.978382921218872,
      "epoch": 5.046683524606108,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002744235164079135,
      "loss": 4.9916,
      "mean_token_accuracy": 0.21495700776576995,
      "num_tokens": 112517262.0,
      "step": 64865
    },
    {
      "entropy": 5.952741956710815,
      "epoch": 5.047072553977825,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002743958009156928,
      "loss": 4.8855,
      "mean_token_accuracy": 0.2230050280690193,
      "num_tokens": 112526464.0,
      "step": 64870
    },
    {
      "entropy": 5.988785266876221,
      "epoch": 5.047461583349543,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002743680854326398,
      "loss": 5.0682,
      "mean_token_accuracy": 0.20602674931287765,
      "num_tokens": 112534582.0,
      "step": 64875
    },
    {
      "entropy": 6.000503587722778,
      "epoch": 5.047850612721261,
      "grad_norm": 1.375,
      "learning_rate": 0.0002743403699591751,
      "loss": 5.0605,
      "mean_token_accuracy": 0.20553330034017564,
      "num_tokens": 112543561.0,
      "step": 64880
    },
    {
      "entropy": 5.98208703994751,
      "epoch": 5.048239642092978,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027431265449571937,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21677876859903336,
      "num_tokens": 112552500.0,
      "step": 64885
    },
    {
      "entropy": 6.008733415603638,
      "epoch": 5.048628671464695,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00027428493904269285,
      "loss": 4.9758,
      "mean_token_accuracy": 0.20962157249450683,
      "num_tokens": 112561008.0,
      "step": 64890
    },
    {
      "entropy": 5.929378509521484,
      "epoch": 5.049017700836413,
      "grad_norm": 1.5,
      "learning_rate": 0.0002742572236005164,
      "loss": 4.8468,
      "mean_token_accuracy": 0.2307700112462044,
      "num_tokens": 112569088.0,
      "step": 64895
    },
    {
      "entropy": 5.827186250686646,
      "epoch": 5.049406730208131,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00027422950816961025,
      "loss": 4.832,
      "mean_token_accuracy": 0.2260448753833771,
      "num_tokens": 112577590.0,
      "step": 64900
    },
    {
      "entropy": 5.982899522781372,
      "epoch": 5.049795759579848,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000274201792750395,
      "loss": 4.9818,
      "mean_token_accuracy": 0.21708086431026458,
      "num_tokens": 112586684.0,
      "step": 64905
    },
    {
      "entropy": 5.999047136306762,
      "epoch": 5.050184788951566,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002741740773432914,
      "loss": 4.9972,
      "mean_token_accuracy": 0.21108605414628984,
      "num_tokens": 112595460.0,
      "step": 64910
    },
    {
      "entropy": 5.884735202789306,
      "epoch": 5.050573818323284,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002741463619487199,
      "loss": 4.8553,
      "mean_token_accuracy": 0.2219989389181137,
      "num_tokens": 112604487.0,
      "step": 64915
    },
    {
      "entropy": 5.948068380355835,
      "epoch": 5.050962847695001,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002741186465671011,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21746286302804946,
      "num_tokens": 112613367.0,
      "step": 64920
    },
    {
      "entropy": 5.907820749282837,
      "epoch": 5.051351877066718,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002740909311988553,
      "loss": 4.8608,
      "mean_token_accuracy": 0.2226157456636429,
      "num_tokens": 112621885.0,
      "step": 64925
    },
    {
      "entropy": 6.007283735275268,
      "epoch": 5.051740906438436,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002740632158444031,
      "loss": 4.9963,
      "mean_token_accuracy": 0.21715058237314225,
      "num_tokens": 112630471.0,
      "step": 64930
    },
    {
      "entropy": 6.069671964645385,
      "epoch": 5.052129935810154,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002740355005041652,
      "loss": 5.0786,
      "mean_token_accuracy": 0.19822930693626403,
      "num_tokens": 112638910.0,
      "step": 64935
    },
    {
      "entropy": 5.951589298248291,
      "epoch": 5.052518965181871,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002740077851785621,
      "loss": 4.9364,
      "mean_token_accuracy": 0.22440241575241088,
      "num_tokens": 112647363.0,
      "step": 64940
    },
    {
      "entropy": 5.907050085067749,
      "epoch": 5.052907994553589,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027398006986801415,
      "loss": 4.8987,
      "mean_token_accuracy": 0.21776199340820312,
      "num_tokens": 112656464.0,
      "step": 64945
    },
    {
      "entropy": 5.945102405548096,
      "epoch": 5.053297023925307,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002739523545729421,
      "loss": 4.8816,
      "mean_token_accuracy": 0.23098332583904266,
      "num_tokens": 112665277.0,
      "step": 64950
    },
    {
      "entropy": 6.006834316253662,
      "epoch": 5.053686053297024,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002739246392937664,
      "loss": 5.0493,
      "mean_token_accuracy": 0.2104157954454422,
      "num_tokens": 112674627.0,
      "step": 64955
    },
    {
      "entropy": 5.88004059791565,
      "epoch": 5.054075082668741,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027389692403090754,
      "loss": 4.8177,
      "mean_token_accuracy": 0.23242831826210023,
      "num_tokens": 112683495.0,
      "step": 64960
    },
    {
      "entropy": 5.903785228729248,
      "epoch": 5.054464112040459,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027386920878478616,
      "loss": 4.8665,
      "mean_token_accuracy": 0.2234046846628189,
      "num_tokens": 112693354.0,
      "step": 64965
    },
    {
      "entropy": 5.961070919036866,
      "epoch": 5.054853141412177,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027384149355582266,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21609951853752135,
      "num_tokens": 112702139.0,
      "step": 64970
    },
    {
      "entropy": 5.985744857788086,
      "epoch": 5.055242170783894,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002738137783444377,
      "loss": 4.926,
      "mean_token_accuracy": 0.21396335810422898,
      "num_tokens": 112710696.0,
      "step": 64975
    },
    {
      "entropy": 5.971740055084228,
      "epoch": 5.055631200155612,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027378606315105175,
      "loss": 5.0054,
      "mean_token_accuracy": 0.20940473824739456,
      "num_tokens": 112719058.0,
      "step": 64980
    },
    {
      "entropy": 6.046074867248535,
      "epoch": 5.056020229527329,
      "grad_norm": 1.625,
      "learning_rate": 0.0002737583479760854,
      "loss": 5.0296,
      "mean_token_accuracy": 0.21401085555553437,
      "num_tokens": 112727719.0,
      "step": 64985
    },
    {
      "entropy": 6.0324279308319095,
      "epoch": 5.056409258899047,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002737306328199591,
      "loss": 5.046,
      "mean_token_accuracy": 0.21264223903417587,
      "num_tokens": 112736631.0,
      "step": 64990
    },
    {
      "entropy": 5.922986936569214,
      "epoch": 5.056798288270764,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027370291768309354,
      "loss": 4.8445,
      "mean_token_accuracy": 0.22336238324642183,
      "num_tokens": 112745609.0,
      "step": 64995
    },
    {
      "entropy": 5.882319736480713,
      "epoch": 5.057187317642482,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000273675202565909,
      "loss": 4.992,
      "mean_token_accuracy": 0.2113862529397011,
      "num_tokens": 112754505.0,
      "step": 65000
    },
    {
      "entropy": 5.981713390350341,
      "epoch": 5.0575763470142,
      "grad_norm": 1.5625,
      "learning_rate": 0.00027364748746882623,
      "loss": 4.9264,
      "mean_token_accuracy": 0.2180582568049431,
      "num_tokens": 112763237.0,
      "step": 65005
    },
    {
      "entropy": 5.990506601333618,
      "epoch": 5.057965376385917,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027361977239226566,
      "loss": 4.9738,
      "mean_token_accuracy": 0.21801112592220306,
      "num_tokens": 112772011.0,
      "step": 65010
    },
    {
      "entropy": 5.974327468872071,
      "epoch": 5.058354405757635,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002735920573366478,
      "loss": 4.9183,
      "mean_token_accuracy": 0.22197403907775878,
      "num_tokens": 112780143.0,
      "step": 65015
    },
    {
      "entropy": 5.934564065933228,
      "epoch": 5.058743435129352,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002735643423023933,
      "loss": 4.9365,
      "mean_token_accuracy": 0.21756203919649125,
      "num_tokens": 112789154.0,
      "step": 65020
    },
    {
      "entropy": 5.927538061141968,
      "epoch": 5.05913246450107,
      "grad_norm": 1.5,
      "learning_rate": 0.0002735366272899226,
      "loss": 4.8996,
      "mean_token_accuracy": 0.2291517212986946,
      "num_tokens": 112797285.0,
      "step": 65025
    },
    {
      "entropy": 5.928336048126221,
      "epoch": 5.059521493872787,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00027350891229965623,
      "loss": 4.9549,
      "mean_token_accuracy": 0.2109627828001976,
      "num_tokens": 112805882.0,
      "step": 65030
    },
    {
      "entropy": 5.870224857330323,
      "epoch": 5.059910523244505,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002734811973320147,
      "loss": 4.8902,
      "mean_token_accuracy": 0.21746993213891982,
      "num_tokens": 112814830.0,
      "step": 65035
    },
    {
      "entropy": 6.0118919849395756,
      "epoch": 5.060299552616223,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00027345348238741864,
      "loss": 4.9844,
      "mean_token_accuracy": 0.21630282700061798,
      "num_tokens": 112823778.0,
      "step": 65040
    },
    {
      "entropy": 5.944325876235962,
      "epoch": 5.06068858198794,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00027342576746628854,
      "loss": 4.9655,
      "mean_token_accuracy": 0.215819251537323,
      "num_tokens": 112832255.0,
      "step": 65045
    },
    {
      "entropy": 5.98446741104126,
      "epoch": 5.061077611359658,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002733980525690449,
      "loss": 5.0198,
      "mean_token_accuracy": 0.21596342772245408,
      "num_tokens": 112840913.0,
      "step": 65050
    },
    {
      "entropy": 5.984060335159302,
      "epoch": 5.061466640731375,
      "grad_norm": 1.625,
      "learning_rate": 0.00027337033769610827,
      "loss": 4.989,
      "mean_token_accuracy": 0.2151486873626709,
      "num_tokens": 112849285.0,
      "step": 65055
    },
    {
      "entropy": 5.96875524520874,
      "epoch": 5.061855670103093,
      "grad_norm": 1.5,
      "learning_rate": 0.0002733426228478991,
      "loss": 4.9667,
      "mean_token_accuracy": 0.21276404857635497,
      "num_tokens": 112858228.0,
      "step": 65060
    },
    {
      "entropy": 5.8846405982971195,
      "epoch": 5.06224469947481,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002733149080248381,
      "loss": 4.813,
      "mean_token_accuracy": 0.22906408160924913,
      "num_tokens": 112866768.0,
      "step": 65065
    },
    {
      "entropy": 5.935096645355225,
      "epoch": 5.062633728846528,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002732871932273456,
      "loss": 4.9408,
      "mean_token_accuracy": 0.22284477353096008,
      "num_tokens": 112875021.0,
      "step": 65070
    },
    {
      "entropy": 5.896332406997681,
      "epoch": 5.063022758218246,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002732594784558423,
      "loss": 4.9177,
      "mean_token_accuracy": 0.2210765540599823,
      "num_tokens": 112884139.0,
      "step": 65075
    },
    {
      "entropy": 5.860652065277099,
      "epoch": 5.063411787589963,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002732317637107486,
      "loss": 4.8275,
      "mean_token_accuracy": 0.22939564734697343,
      "num_tokens": 112891979.0,
      "step": 65080
    },
    {
      "entropy": 5.975470161437988,
      "epoch": 5.063800816961681,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027320404899248507,
      "loss": 4.9662,
      "mean_token_accuracy": 0.21652035117149354,
      "num_tokens": 112901479.0,
      "step": 65085
    },
    {
      "entropy": 5.941552448272705,
      "epoch": 5.064189846333398,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002731763343014722,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22376079857349396,
      "num_tokens": 112909479.0,
      "step": 65090
    },
    {
      "entropy": 6.045760536193848,
      "epoch": 5.064578875705116,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00027314861963813065,
      "loss": 5.0295,
      "mean_token_accuracy": 0.21272787153720857,
      "num_tokens": 112918457.0,
      "step": 65095
    },
    {
      "entropy": 6.001100683212281,
      "epoch": 5.064967905076833,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027312090500288065,
      "loss": 4.9612,
      "mean_token_accuracy": 0.2196800321340561,
      "num_tokens": 112927833.0,
      "step": 65100
    },
    {
      "entropy": 5.9191446781158445,
      "epoch": 5.065356934448551,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002730931903961431,
      "loss": 4.93,
      "mean_token_accuracy": 0.22177924662828447,
      "num_tokens": 112935906.0,
      "step": 65105
    },
    {
      "entropy": 5.965351009368897,
      "epoch": 5.065745963820269,
      "grad_norm": 1.53125,
      "learning_rate": 0.00027306547581833836,
      "loss": 4.9932,
      "mean_token_accuracy": 0.223183673620224,
      "num_tokens": 112944375.0,
      "step": 65110
    },
    {
      "entropy": 5.984156799316406,
      "epoch": 5.066134993191986,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027303776126988687,
      "loss": 4.9434,
      "mean_token_accuracy": 0.21819078475236892,
      "num_tokens": 112953280.0,
      "step": 65115
    },
    {
      "entropy": 5.984451723098755,
      "epoch": 5.066524022563703,
      "grad_norm": 1.375,
      "learning_rate": 0.00027301004675120926,
      "loss": 4.9249,
      "mean_token_accuracy": 0.22579724937677384,
      "num_tokens": 112961752.0,
      "step": 65120
    },
    {
      "entropy": 5.960571813583374,
      "epoch": 5.066913051935421,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002729823322627261,
      "loss": 4.9053,
      "mean_token_accuracy": 0.2195689484477043,
      "num_tokens": 112970558.0,
      "step": 65125
    },
    {
      "entropy": 5.972404718399048,
      "epoch": 5.0673020813071386,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002729546178048577,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21649586260318757,
      "num_tokens": 112979259.0,
      "step": 65130
    },
    {
      "entropy": 5.948299074172974,
      "epoch": 5.067691110678856,
      "grad_norm": 1.5,
      "learning_rate": 0.0002729269033780248,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21851952373981476,
      "num_tokens": 112987807.0,
      "step": 65135
    },
    {
      "entropy": 5.998951435089111,
      "epoch": 5.068080140050574,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027289918898264784,
      "loss": 4.9879,
      "mean_token_accuracy": 0.21094974428415297,
      "num_tokens": 112997102.0,
      "step": 65140
    },
    {
      "entropy": 6.001391077041626,
      "epoch": 5.068469169422292,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002728714746191473,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21674688160419464,
      "num_tokens": 113005036.0,
      "step": 65145
    },
    {
      "entropy": 5.993588924407959,
      "epoch": 5.068858198794009,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027284376028794376,
      "loss": 4.9575,
      "mean_token_accuracy": 0.22314064353704452,
      "num_tokens": 113013415.0,
      "step": 65150
    },
    {
      "entropy": 5.915716314315796,
      "epoch": 5.069247228165726,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00027281604598945774,
      "loss": 4.906,
      "mean_token_accuracy": 0.22116379588842391,
      "num_tokens": 113022239.0,
      "step": 65155
    },
    {
      "entropy": 5.921736478805542,
      "epoch": 5.069636257537444,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002727883317241098,
      "loss": 4.938,
      "mean_token_accuracy": 0.21675106137990952,
      "num_tokens": 113030910.0,
      "step": 65160
    },
    {
      "entropy": 5.935070943832398,
      "epoch": 5.0700252869091615,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00027276061749232034,
      "loss": 4.9374,
      "mean_token_accuracy": 0.2195601910352707,
      "num_tokens": 113039672.0,
      "step": 65165
    },
    {
      "entropy": 6.021441173553467,
      "epoch": 5.070414316280879,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002727329032945099,
      "loss": 4.9993,
      "mean_token_accuracy": 0.2162391796708107,
      "num_tokens": 113047951.0,
      "step": 65170
    },
    {
      "entropy": 5.97096529006958,
      "epoch": 5.070803345652597,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002727051891310991,
      "loss": 5.0113,
      "mean_token_accuracy": 0.2131443127989769,
      "num_tokens": 113056590.0,
      "step": 65175
    },
    {
      "entropy": 6.005364036560058,
      "epoch": 5.071192375024315,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0002726774750025084,
      "loss": 4.9558,
      "mean_token_accuracy": 0.2167626604437828,
      "num_tokens": 113065427.0,
      "step": 65180
    },
    {
      "entropy": 6.029196166992188,
      "epoch": 5.071581404396032,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027264976090915835,
      "loss": 5.0555,
      "mean_token_accuracy": 0.20568238496780394,
      "num_tokens": 113074619.0,
      "step": 65185
    },
    {
      "entropy": 5.986215829849243,
      "epoch": 5.071970433767749,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002726220468514694,
      "loss": 5.0128,
      "mean_token_accuracy": 0.20942171961069106,
      "num_tokens": 113083360.0,
      "step": 65190
    },
    {
      "entropy": 5.934966564178467,
      "epoch": 5.072359463139467,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002725943328298621,
      "loss": 4.8828,
      "mean_token_accuracy": 0.2175072818994522,
      "num_tokens": 113091413.0,
      "step": 65195
    },
    {
      "entropy": 5.958190059661865,
      "epoch": 5.0727484925111845,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000272566618844757,
      "loss": 4.932,
      "mean_token_accuracy": 0.22031228691339494,
      "num_tokens": 113100645.0,
      "step": 65200
    },
    {
      "entropy": 5.975075197219849,
      "epoch": 5.073137521882902,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027253890489657457,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2172503411769867,
      "num_tokens": 113109046.0,
      "step": 65205
    },
    {
      "entropy": 5.991329526901245,
      "epoch": 5.07352655125462,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00027251119098573533,
      "loss": 4.9911,
      "mean_token_accuracy": 0.20559496730566024,
      "num_tokens": 113117046.0,
      "step": 65210
    },
    {
      "entropy": 6.021987104415894,
      "epoch": 5.073915580626338,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027248347711265983,
      "loss": 5.0927,
      "mean_token_accuracy": 0.20659649819135667,
      "num_tokens": 113126854.0,
      "step": 65215
    },
    {
      "entropy": 5.975305223464966,
      "epoch": 5.074304609998054,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002724557632777686,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21689300686120988,
      "num_tokens": 113135471.0,
      "step": 65220
    },
    {
      "entropy": 6.00200834274292,
      "epoch": 5.074693639369772,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00027242804948148197,
      "loss": 4.9628,
      "mean_token_accuracy": 0.21256966143846512,
      "num_tokens": 113143592.0,
      "step": 65225
    },
    {
      "entropy": 5.917579460144043,
      "epoch": 5.07508266874149,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002724003357242207,
      "loss": 4.889,
      "mean_token_accuracy": 0.2196398213505745,
      "num_tokens": 113152415.0,
      "step": 65230
    },
    {
      "entropy": 5.973000574111938,
      "epoch": 5.0754716981132075,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002723726220064051,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21663213074207305,
      "num_tokens": 113161589.0,
      "step": 65235
    },
    {
      "entropy": 5.961618518829345,
      "epoch": 5.075860727484925,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002723449083284559,
      "loss": 4.9633,
      "mean_token_accuracy": 0.21506914049386977,
      "num_tokens": 113169785.0,
      "step": 65240
    },
    {
      "entropy": 6.003759860992432,
      "epoch": 5.076249756856643,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027231719469079344,
      "loss": 5.0433,
      "mean_token_accuracy": 0.21410675197839737,
      "num_tokens": 113178835.0,
      "step": 65245
    },
    {
      "entropy": 5.939591026306152,
      "epoch": 5.076638786228361,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027228948109383826,
      "loss": 4.9436,
      "mean_token_accuracy": 0.22023582756519317,
      "num_tokens": 113187821.0,
      "step": 65250
    },
    {
      "entropy": 5.932314395904541,
      "epoch": 5.077027815600077,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027226176753801095,
      "loss": 4.9853,
      "mean_token_accuracy": 0.21344951242208482,
      "num_tokens": 113196776.0,
      "step": 65255
    },
    {
      "entropy": 5.911448335647583,
      "epoch": 5.077416844971795,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00027223405402373194,
      "loss": 4.9127,
      "mean_token_accuracy": 0.219694322347641,
      "num_tokens": 113205206.0,
      "step": 65260
    },
    {
      "entropy": 5.9763267993927,
      "epoch": 5.077805874343513,
      "grad_norm": 1.5625,
      "learning_rate": 0.00027220634055142174,
      "loss": 4.9564,
      "mean_token_accuracy": 0.22077599614858628,
      "num_tokens": 113213728.0,
      "step": 65265
    },
    {
      "entropy": 5.9543843269348145,
      "epoch": 5.0781949037152305,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002721786271215009,
      "loss": 4.9393,
      "mean_token_accuracy": 0.22649639546871186,
      "num_tokens": 113222135.0,
      "step": 65270
    },
    {
      "entropy": 5.9871947288513185,
      "epoch": 5.078583933086948,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002721509137343899,
      "loss": 4.9308,
      "mean_token_accuracy": 0.21651903241872789,
      "num_tokens": 113231841.0,
      "step": 65275
    },
    {
      "entropy": 5.989714241027832,
      "epoch": 5.078972962458666,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027212320039050924,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21700379252433777,
      "num_tokens": 113240639.0,
      "step": 65280
    },
    {
      "entropy": 5.997336387634277,
      "epoch": 5.079361991830384,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00027209548709027944,
      "loss": 4.9857,
      "mean_token_accuracy": 0.2146141916513443,
      "num_tokens": 113249240.0,
      "step": 65285
    },
    {
      "entropy": 5.9954146385192875,
      "epoch": 5.0797510212021,
      "grad_norm": 1.5,
      "learning_rate": 0.000272067773834121,
      "loss": 4.9859,
      "mean_token_accuracy": 0.21949173957109452,
      "num_tokens": 113258403.0,
      "step": 65290
    },
    {
      "entropy": 5.987963056564331,
      "epoch": 5.080140050573818,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002720400606224544,
      "loss": 4.9177,
      "mean_token_accuracy": 0.22609300911426544,
      "num_tokens": 113267254.0,
      "step": 65295
    },
    {
      "entropy": 5.987097692489624,
      "epoch": 5.080529079945536,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00027201234745570014,
      "loss": 4.9688,
      "mean_token_accuracy": 0.21421261578798295,
      "num_tokens": 113275836.0,
      "step": 65300
    },
    {
      "entropy": 5.962496185302735,
      "epoch": 5.0809181093172535,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002719846343342788,
      "loss": 4.9442,
      "mean_token_accuracy": 0.21429015845060348,
      "num_tokens": 113284261.0,
      "step": 65305
    },
    {
      "entropy": 5.96199254989624,
      "epoch": 5.081307138688971,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002719569212586108,
      "loss": 4.9384,
      "mean_token_accuracy": 0.21760209500789643,
      "num_tokens": 113293444.0,
      "step": 65310
    },
    {
      "entropy": 5.987563896179199,
      "epoch": 5.081696168060689,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002719292082291167,
      "loss": 4.9309,
      "mean_token_accuracy": 0.21371857523918153,
      "num_tokens": 113302566.0,
      "step": 65315
    },
    {
      "entropy": 6.0054858207702635,
      "epoch": 5.082085197432406,
      "grad_norm": 1.53125,
      "learning_rate": 0.000271901495246217,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2190324604511261,
      "num_tokens": 113311668.0,
      "step": 65320
    },
    {
      "entropy": 5.908246326446533,
      "epoch": 5.082474226804123,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002718737823103322,
      "loss": 4.9636,
      "mean_token_accuracy": 0.21549442559480667,
      "num_tokens": 113321450.0,
      "step": 65325
    },
    {
      "entropy": 5.897635841369629,
      "epoch": 5.082863256175841,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027184606942188275,
      "loss": 4.8281,
      "mean_token_accuracy": 0.2301758423447609,
      "num_tokens": 113329707.0,
      "step": 65330
    },
    {
      "entropy": 5.993349075317383,
      "epoch": 5.083252285547559,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027181835658128913,
      "loss": 4.9979,
      "mean_token_accuracy": 0.21561479717493057,
      "num_tokens": 113338483.0,
      "step": 65335
    },
    {
      "entropy": 5.998838615417481,
      "epoch": 5.0836413149192765,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002717906437889719,
      "loss": 4.9609,
      "mean_token_accuracy": 0.21652824580669403,
      "num_tokens": 113346503.0,
      "step": 65340
    },
    {
      "entropy": 5.946887016296387,
      "epoch": 5.084030344290994,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027176293104535154,
      "loss": 4.9228,
      "mean_token_accuracy": 0.21709365248680115,
      "num_tokens": 113354859.0,
      "step": 65345
    },
    {
      "entropy": 5.939001083374023,
      "epoch": 5.084419373662712,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00027173521835084863,
      "loss": 4.9368,
      "mean_token_accuracy": 0.218848879635334,
      "num_tokens": 113363534.0,
      "step": 65350
    },
    {
      "entropy": 5.935678958892822,
      "epoch": 5.084808403034429,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002717075057058835,
      "loss": 4.9234,
      "mean_token_accuracy": 0.21607306897640227,
      "num_tokens": 113372234.0,
      "step": 65355
    },
    {
      "entropy": 5.950419092178345,
      "epoch": 5.085197432406146,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002716797931108768,
      "loss": 5.0201,
      "mean_token_accuracy": 0.20643693655729295,
      "num_tokens": 113381021.0,
      "step": 65360
    },
    {
      "entropy": 6.006762361526489,
      "epoch": 5.085586461777864,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002716520805662489,
      "loss": 4.9845,
      "mean_token_accuracy": 0.21077614426612853,
      "num_tokens": 113388889.0,
      "step": 65365
    },
    {
      "entropy": 5.9278655529022215,
      "epoch": 5.085975491149582,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00027162436807242024,
      "loss": 4.9131,
      "mean_token_accuracy": 0.2195081368088722,
      "num_tokens": 113397080.0,
      "step": 65370
    },
    {
      "entropy": 5.959735298156739,
      "epoch": 5.0863645205212995,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027159665562981154,
      "loss": 4.9644,
      "mean_token_accuracy": 0.2131648227572441,
      "num_tokens": 113405756.0,
      "step": 65375
    },
    {
      "entropy": 5.923728561401367,
      "epoch": 5.086753549893017,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002715689432388433,
      "loss": 4.9487,
      "mean_token_accuracy": 0.22262462824583054,
      "num_tokens": 113413729.0,
      "step": 65380
    },
    {
      "entropy": 5.955289030075074,
      "epoch": 5.087142579264735,
      "grad_norm": 1.5,
      "learning_rate": 0.0002715412308999357,
      "loss": 4.9971,
      "mean_token_accuracy": 0.21626293510198594,
      "num_tokens": 113423649.0,
      "step": 65385
    },
    {
      "entropy": 5.957227945327759,
      "epoch": 5.087531608636452,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00027151351861350953,
      "loss": 4.955,
      "mean_token_accuracy": 0.21354981809854506,
      "num_tokens": 113432579.0,
      "step": 65390
    },
    {
      "entropy": 5.945972347259522,
      "epoch": 5.087920638008169,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002714858063799851,
      "loss": 4.8379,
      "mean_token_accuracy": 0.22348373979330063,
      "num_tokens": 113440158.0,
      "step": 65395
    },
    {
      "entropy": 5.971912527084351,
      "epoch": 5.088309667379887,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027145809419978293,
      "loss": 4.933,
      "mean_token_accuracy": 0.2201162949204445,
      "num_tokens": 113448757.0,
      "step": 65400
    },
    {
      "entropy": 5.926597547531128,
      "epoch": 5.088698696751605,
      "grad_norm": 1.5625,
      "learning_rate": 0.00027143038207332363,
      "loss": 4.9398,
      "mean_token_accuracy": 0.21838116347789766,
      "num_tokens": 113457454.0,
      "step": 65405
    },
    {
      "entropy": 5.9431370258331295,
      "epoch": 5.0890877261233225,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027140267000102755,
      "loss": 4.924,
      "mean_token_accuracy": 0.21589357107877732,
      "num_tokens": 113465783.0,
      "step": 65410
    },
    {
      "entropy": 5.923536014556885,
      "epoch": 5.08947675549504,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027137495798331524,
      "loss": 4.9542,
      "mean_token_accuracy": 0.2177286699414253,
      "num_tokens": 113474376.0,
      "step": 65415
    },
    {
      "entropy": 5.9130407810211185,
      "epoch": 5.089865784866757,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00027134724602060717,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2217514768242836,
      "num_tokens": 113482769.0,
      "step": 65420
    },
    {
      "entropy": 6.0055278778076175,
      "epoch": 5.090254814238475,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027131953411332384,
      "loss": 5.0131,
      "mean_token_accuracy": 0.2143533393740654,
      "num_tokens": 113491579.0,
      "step": 65425
    },
    {
      "entropy": 5.995757055282593,
      "epoch": 5.090643843610192,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027129182226188575,
      "loss": 5.004,
      "mean_token_accuracy": 0.2138711169362068,
      "num_tokens": 113500150.0,
      "step": 65430
    },
    {
      "entropy": 5.995841312408447,
      "epoch": 5.09103287298191,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002712641104667132,
      "loss": 4.9674,
      "mean_token_accuracy": 0.21451892852783203,
      "num_tokens": 113508251.0,
      "step": 65435
    },
    {
      "entropy": 5.96952075958252,
      "epoch": 5.091421902353628,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000271236398728227,
      "loss": 4.94,
      "mean_token_accuracy": 0.21607869863510132,
      "num_tokens": 113517108.0,
      "step": 65440
    },
    {
      "entropy": 5.895830059051514,
      "epoch": 5.0918109317253455,
      "grad_norm": 1.640625,
      "learning_rate": 0.00027120868704684736,
      "loss": 4.8317,
      "mean_token_accuracy": 0.22424757182598115,
      "num_tokens": 113525971.0,
      "step": 65445
    },
    {
      "entropy": 6.0145957469940186,
      "epoch": 5.092199961097063,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002711809754229949,
      "loss": 5.0444,
      "mean_token_accuracy": 0.211050321161747,
      "num_tokens": 113534800.0,
      "step": 65450
    },
    {
      "entropy": 5.933331632614136,
      "epoch": 5.09258899046878,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00027115326385709004,
      "loss": 4.8795,
      "mean_token_accuracy": 0.2192299023270607,
      "num_tokens": 113542871.0,
      "step": 65455
    },
    {
      "entropy": 6.008037042617798,
      "epoch": 5.092978019840498,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002711255523495534,
      "loss": 5.0232,
      "mean_token_accuracy": 0.2143043100833893,
      "num_tokens": 113551317.0,
      "step": 65460
    },
    {
      "entropy": 6.017021608352661,
      "epoch": 5.093367049212215,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002710978409008052,
      "loss": 4.9969,
      "mean_token_accuracy": 0.2175760880112648,
      "num_tokens": 113560636.0,
      "step": 65465
    },
    {
      "entropy": 5.965992164611817,
      "epoch": 5.093756078583933,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00027107012951126607,
      "loss": 4.9065,
      "mean_token_accuracy": 0.22037338614463806,
      "num_tokens": 113569404.0,
      "step": 65470
    },
    {
      "entropy": 5.957309818267822,
      "epoch": 5.094145107955651,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027104241818135655,
      "loss": 4.9343,
      "mean_token_accuracy": 0.22087874859571457,
      "num_tokens": 113577997.0,
      "step": 65475
    },
    {
      "entropy": 5.9449866771697994,
      "epoch": 5.0945341373273685,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027101470691149693,
      "loss": 4.921,
      "mean_token_accuracy": 0.222595152258873,
      "num_tokens": 113587169.0,
      "step": 65480
    },
    {
      "entropy": 5.885840845108032,
      "epoch": 5.094923166699086,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027098699570210787,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2176942154765129,
      "num_tokens": 113595406.0,
      "step": 65485
    },
    {
      "entropy": 5.895366621017456,
      "epoch": 5.095312196070803,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00027095928455360977,
      "loss": 4.9388,
      "mean_token_accuracy": 0.2171615928411484,
      "num_tokens": 113603530.0,
      "step": 65490
    },
    {
      "entropy": 5.958537864685058,
      "epoch": 5.095701225442521,
      "grad_norm": 1.40625,
      "learning_rate": 0.000270931573466423,
      "loss": 4.9539,
      "mean_token_accuracy": 0.21815474331378937,
      "num_tokens": 113612066.0,
      "step": 65495
    },
    {
      "entropy": 5.95005841255188,
      "epoch": 5.096090254814238,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002709038624409681,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22208423167467117,
      "num_tokens": 113619939.0,
      "step": 65500
    },
    {
      "entropy": 5.938938045501709,
      "epoch": 5.096479284185956,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00027087615147766567,
      "loss": 4.9004,
      "mean_token_accuracy": 0.22633223831653596,
      "num_tokens": 113628425.0,
      "step": 65505
    },
    {
      "entropy": 5.925915002822876,
      "epoch": 5.096868313557674,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000270848440576936,
      "loss": 4.975,
      "mean_token_accuracy": 0.21568122506141663,
      "num_tokens": 113637037.0,
      "step": 65510
    },
    {
      "entropy": 5.9763024806976315,
      "epoch": 5.0972573429293915,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027082072973919974,
      "loss": 4.9505,
      "mean_token_accuracy": 0.22626060545444487,
      "num_tokens": 113646586.0,
      "step": 65515
    },
    {
      "entropy": 5.966061925888061,
      "epoch": 5.097646372301108,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00027079301896487716,
      "loss": 4.9494,
      "mean_token_accuracy": 0.2201508939266205,
      "num_tokens": 113655695.0,
      "step": 65520
    },
    {
      "entropy": 6.008488321304322,
      "epoch": 5.098035401672826,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002707653082543889,
      "loss": 4.9645,
      "mean_token_accuracy": 0.2179410144686699,
      "num_tokens": 113663930.0,
      "step": 65525
    },
    {
      "entropy": 5.99328932762146,
      "epoch": 5.098424431044544,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002707375976081553,
      "loss": 4.967,
      "mean_token_accuracy": 0.21146235316991807,
      "num_tokens": 113672628.0,
      "step": 65530
    },
    {
      "entropy": 6.001133108139038,
      "epoch": 5.098813460416261,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00027070988702659687,
      "loss": 5.0007,
      "mean_token_accuracy": 0.21745990067720414,
      "num_tokens": 113681703.0,
      "step": 65535
    },
    {
      "entropy": 5.986928510665893,
      "epoch": 5.099202489787979,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027068217651013415,
      "loss": 4.9761,
      "mean_token_accuracy": 0.21054674535989762,
      "num_tokens": 113690164.0,
      "step": 65540
    },
    {
      "entropy": 5.942771482467651,
      "epoch": 5.099591519159697,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002706544660591875,
      "loss": 4.8901,
      "mean_token_accuracy": 0.22384084910154342,
      "num_tokens": 113698724.0,
      "step": 65545
    },
    {
      "entropy": 5.97576756477356,
      "epoch": 5.0999805485314145,
      "grad_norm": 1.5625,
      "learning_rate": 0.00027062675567417736,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21544487178325653,
      "num_tokens": 113707567.0,
      "step": 65550
    },
    {
      "entropy": 5.985833549499512,
      "epoch": 5.100369577903131,
      "grad_norm": 1.578125,
      "learning_rate": 0.00027059904535552424,
      "loss": 4.9861,
      "mean_token_accuracy": 0.21180417537689208,
      "num_tokens": 113717546.0,
      "step": 65555
    },
    {
      "entropy": 5.956966304779053,
      "epoch": 5.100758607274849,
      "grad_norm": 1.65625,
      "learning_rate": 0.00027057133510364866,
      "loss": 4.8585,
      "mean_token_accuracy": 0.22728909701108932,
      "num_tokens": 113726234.0,
      "step": 65560
    },
    {
      "entropy": 5.8946271419525145,
      "epoch": 5.101147636646567,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002705436249189709,
      "loss": 4.7948,
      "mean_token_accuracy": 0.23061707764863967,
      "num_tokens": 113734659.0,
      "step": 65565
    },
    {
      "entropy": 5.931036520004272,
      "epoch": 5.101536666018284,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027051591480191167,
      "loss": 4.988,
      "mean_token_accuracy": 0.2126322016119957,
      "num_tokens": 113743788.0,
      "step": 65570
    },
    {
      "entropy": 5.915783071517945,
      "epoch": 5.101925695390002,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002704882047528912,
      "loss": 4.9076,
      "mean_token_accuracy": 0.2233823522925377,
      "num_tokens": 113753116.0,
      "step": 65575
    },
    {
      "entropy": 5.950930213928222,
      "epoch": 5.10231472476172,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027046049477233013,
      "loss": 4.9791,
      "mean_token_accuracy": 0.2197417438030243,
      "num_tokens": 113762364.0,
      "step": 65580
    },
    {
      "entropy": 5.883121395111084,
      "epoch": 5.1027037541334375,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002704327848606488,
      "loss": 4.8434,
      "mean_token_accuracy": 0.22654048800468446,
      "num_tokens": 113771550.0,
      "step": 65585
    },
    {
      "entropy": 5.920892572402954,
      "epoch": 5.103092783505154,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002704050750182677,
      "loss": 4.9036,
      "mean_token_accuracy": 0.22428347468376159,
      "num_tokens": 113779897.0,
      "step": 65590
    },
    {
      "entropy": 5.931081533432007,
      "epoch": 5.103481812876872,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027037736524560717,
      "loss": 4.9312,
      "mean_token_accuracy": 0.21831361949443817,
      "num_tokens": 113788608.0,
      "step": 65595
    },
    {
      "entropy": 5.8857725143432615,
      "epoch": 5.10387084224859,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002703496555430879,
      "loss": 4.8735,
      "mean_token_accuracy": 0.21328488886356353,
      "num_tokens": 113796624.0,
      "step": 65600
    },
    {
      "entropy": 5.964398384094238,
      "epoch": 5.104259871620307,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00027032194591113015,
      "loss": 4.9452,
      "mean_token_accuracy": 0.2122003361582756,
      "num_tokens": 113805133.0,
      "step": 65605
    },
    {
      "entropy": 5.941225624084472,
      "epoch": 5.104648900992025,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002702942363501544,
      "loss": 4.9719,
      "mean_token_accuracy": 0.21329574435949325,
      "num_tokens": 113814305.0,
      "step": 65610
    },
    {
      "entropy": 5.909062099456787,
      "epoch": 5.105037930363743,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027026652686058117,
      "loss": 4.8681,
      "mean_token_accuracy": 0.22504508644342422,
      "num_tokens": 113822958.0,
      "step": 65615
    },
    {
      "entropy": 6.011416482925415,
      "epoch": 5.1054269597354605,
      "grad_norm": 1.515625,
      "learning_rate": 0.00027023881744283083,
      "loss": 4.9521,
      "mean_token_accuracy": 0.21228974610567092,
      "num_tokens": 113831665.0,
      "step": 65620
    },
    {
      "entropy": 6.021967458724975,
      "epoch": 5.105815989107177,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002702111080973239,
      "loss": 5.067,
      "mean_token_accuracy": 0.20496160984039308,
      "num_tokens": 113840777.0,
      "step": 65625
    },
    {
      "entropy": 5.981190872192383,
      "epoch": 5.106205018478895,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002701833988244808,
      "loss": 4.9606,
      "mean_token_accuracy": 0.21615783572196962,
      "num_tokens": 113849014.0,
      "step": 65630
    },
    {
      "entropy": 5.896233463287354,
      "epoch": 5.106594047850613,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027015568962472175,
      "loss": 4.8405,
      "mean_token_accuracy": 0.22291838228702546,
      "num_tokens": 113857814.0,
      "step": 65635
    },
    {
      "entropy": 5.965276002883911,
      "epoch": 5.10698307722233,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027012798049846757,
      "loss": 4.9858,
      "mean_token_accuracy": 0.21384083330631257,
      "num_tokens": 113866400.0,
      "step": 65640
    },
    {
      "entropy": 5.945168590545654,
      "epoch": 5.107372106594048,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002701002714461385,
      "loss": 4.9492,
      "mean_token_accuracy": 0.21756862848997116,
      "num_tokens": 113874756.0,
      "step": 65645
    },
    {
      "entropy": 5.9400066375732425,
      "epoch": 5.107761135965766,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002700725624681551,
      "loss": 4.9513,
      "mean_token_accuracy": 0.22223164588212968,
      "num_tokens": 113884172.0,
      "step": 65650
    },
    {
      "entropy": 5.973586368560791,
      "epoch": 5.108150165337483,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00027004485356493763,
      "loss": 4.9753,
      "mean_token_accuracy": 0.21653239876031877,
      "num_tokens": 113892757.0,
      "step": 65655
    },
    {
      "entropy": 5.924812650680542,
      "epoch": 5.1085391947092,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027001714473690665,
      "loss": 4.9026,
      "mean_token_accuracy": 0.22634202986955643,
      "num_tokens": 113901907.0,
      "step": 65660
    },
    {
      "entropy": 5.9740862369537355,
      "epoch": 5.108928224080918,
      "grad_norm": 1.59375,
      "learning_rate": 0.00026998943598448256,
      "loss": 5.0069,
      "mean_token_accuracy": 0.21508377343416213,
      "num_tokens": 113910612.0,
      "step": 65665
    },
    {
      "entropy": 5.918333625793457,
      "epoch": 5.109317253452636,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002699617273080858,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2196083128452301,
      "num_tokens": 113919586.0,
      "step": 65670
    },
    {
      "entropy": 5.940459823608398,
      "epoch": 5.109706282824353,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002699340187081368,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22644262462854386,
      "num_tokens": 113928775.0,
      "step": 65675
    },
    {
      "entropy": 5.939147520065307,
      "epoch": 5.110095312196071,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000269906310185056,
      "loss": 5.0115,
      "mean_token_accuracy": 0.21672501862049104,
      "num_tokens": 113937300.0,
      "step": 65680
    },
    {
      "entropy": 6.055330276489258,
      "epoch": 5.110484341567789,
      "grad_norm": 1.375,
      "learning_rate": 0.00026987860173926387,
      "loss": 5.1075,
      "mean_token_accuracy": 0.20668116509914397,
      "num_tokens": 113946278.0,
      "step": 65685
    },
    {
      "entropy": 6.013170003890991,
      "epoch": 5.1108733709395056,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00026985089337118077,
      "loss": 5.027,
      "mean_token_accuracy": 0.2097335323691368,
      "num_tokens": 113955212.0,
      "step": 65690
    },
    {
      "entropy": 5.9869029998779295,
      "epoch": 5.111262400311223,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026982318508122724,
      "loss": 5.0005,
      "mean_token_accuracy": 0.2259412080049515,
      "num_tokens": 113963938.0,
      "step": 65695
    },
    {
      "entropy": 5.995023632049561,
      "epoch": 5.111651429682941,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002697954768698235,
      "loss": 5.0328,
      "mean_token_accuracy": 0.21452368497848512,
      "num_tokens": 113973002.0,
      "step": 65700
    },
    {
      "entropy": 5.944068670272827,
      "epoch": 5.112040459054659,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026976776873739017,
      "loss": 4.9554,
      "mean_token_accuracy": 0.2166252017021179,
      "num_tokens": 113982955.0,
      "step": 65705
    },
    {
      "entropy": 5.953969287872314,
      "epoch": 5.112429488426376,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002697400606843477,
      "loss": 4.9311,
      "mean_token_accuracy": 0.21803608238697053,
      "num_tokens": 113992221.0,
      "step": 65710
    },
    {
      "entropy": 5.988649082183838,
      "epoch": 5.112818517798094,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026971235271111636,
      "loss": 4.9556,
      "mean_token_accuracy": 0.22221828997135162,
      "num_tokens": 114000820.0,
      "step": 65715
    },
    {
      "entropy": 5.9143448829650875,
      "epoch": 5.113207547169812,
      "grad_norm": 1.375,
      "learning_rate": 0.00026968464481811667,
      "loss": 4.8699,
      "mean_token_accuracy": 0.22805819362401963,
      "num_tokens": 114010240.0,
      "step": 65720
    },
    {
      "entropy": 5.937300729751587,
      "epoch": 5.1135965765415285,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026965693700576904,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22713427394628524,
      "num_tokens": 114018201.0,
      "step": 65725
    },
    {
      "entropy": 5.950121355056763,
      "epoch": 5.113985605913246,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026962922927449394,
      "loss": 4.9647,
      "mean_token_accuracy": 0.22225951105356218,
      "num_tokens": 114026407.0,
      "step": 65730
    },
    {
      "entropy": 5.94598388671875,
      "epoch": 5.114374635284964,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002696015216247117,
      "loss": 4.9515,
      "mean_token_accuracy": 0.21666204631328584,
      "num_tokens": 114035202.0,
      "step": 65735
    },
    {
      "entropy": 5.900272607803345,
      "epoch": 5.114763664656682,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002695738140568427,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21945204436779023,
      "num_tokens": 114043631.0,
      "step": 65740
    },
    {
      "entropy": 5.9470654964447025,
      "epoch": 5.115152694028399,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002695461065713075,
      "loss": 4.8843,
      "mean_token_accuracy": 0.2226795420050621,
      "num_tokens": 114051784.0,
      "step": 65745
    },
    {
      "entropy": 5.939059782028198,
      "epoch": 5.115541723400117,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026951839916852647,
      "loss": 4.9728,
      "mean_token_accuracy": 0.22155822664499283,
      "num_tokens": 114060391.0,
      "step": 65750
    },
    {
      "entropy": 5.879908704757691,
      "epoch": 5.115930752771834,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026949069184892,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21739491522312165,
      "num_tokens": 114068779.0,
      "step": 65755
    },
    {
      "entropy": 5.9142590999603275,
      "epoch": 5.1163197821435515,
      "grad_norm": 1.609375,
      "learning_rate": 0.00026946298461290845,
      "loss": 4.9364,
      "mean_token_accuracy": 0.2163964882493019,
      "num_tokens": 114077087.0,
      "step": 65760
    },
    {
      "entropy": 5.973569393157959,
      "epoch": 5.116708811515269,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002694352774609123,
      "loss": 4.917,
      "mean_token_accuracy": 0.22149318754673003,
      "num_tokens": 114085821.0,
      "step": 65765
    },
    {
      "entropy": 5.933302307128907,
      "epoch": 5.117097840886987,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000269407570393352,
      "loss": 4.9792,
      "mean_token_accuracy": 0.21646054238080978,
      "num_tokens": 114094420.0,
      "step": 65770
    },
    {
      "entropy": 5.822155380249024,
      "epoch": 5.117486870258705,
      "grad_norm": 1.546875,
      "learning_rate": 0.000269379863410648,
      "loss": 4.8224,
      "mean_token_accuracy": 0.2276206612586975,
      "num_tokens": 114103332.0,
      "step": 65775
    },
    {
      "entropy": 5.972478246688842,
      "epoch": 5.117875899630422,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002693521565132205,
      "loss": 5.0057,
      "mean_token_accuracy": 0.2141551584005356,
      "num_tokens": 114111692.0,
      "step": 65780
    },
    {
      "entropy": 6.001748943328858,
      "epoch": 5.11826492900214,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026932444970149006,
      "loss": 4.9323,
      "mean_token_accuracy": 0.21393433660268785,
      "num_tokens": 114120740.0,
      "step": 65785
    },
    {
      "entropy": 5.956196737289429,
      "epoch": 5.118653958373857,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002692967429758771,
      "loss": 4.8795,
      "mean_token_accuracy": 0.2211485281586647,
      "num_tokens": 114129028.0,
      "step": 65790
    },
    {
      "entropy": 6.0029072761535645,
      "epoch": 5.1190429877455745,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026926903633680193,
      "loss": 5.0216,
      "mean_token_accuracy": 0.2101919025182724,
      "num_tokens": 114137076.0,
      "step": 65795
    },
    {
      "entropy": 5.965333652496338,
      "epoch": 5.119432017117292,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026924132978468505,
      "loss": 4.9965,
      "mean_token_accuracy": 0.21408455073833466,
      "num_tokens": 114145948.0,
      "step": 65800
    },
    {
      "entropy": 5.957700490951538,
      "epoch": 5.11982104648901,
      "grad_norm": 1.578125,
      "learning_rate": 0.00026921362331994684,
      "loss": 4.9327,
      "mean_token_accuracy": 0.226883402466774,
      "num_tokens": 114153690.0,
      "step": 65805
    },
    {
      "entropy": 5.888636064529419,
      "epoch": 5.120210075860728,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026918591694300776,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2271373987197876,
      "num_tokens": 114162132.0,
      "step": 65810
    },
    {
      "entropy": 5.895900678634644,
      "epoch": 5.120599105232445,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026915821065428804,
      "loss": 4.8972,
      "mean_token_accuracy": 0.22012446820735931,
      "num_tokens": 114172191.0,
      "step": 65815
    },
    {
      "entropy": 5.95011682510376,
      "epoch": 5.120988134604163,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002691305044542082,
      "loss": 4.8138,
      "mean_token_accuracy": 0.22472718954086304,
      "num_tokens": 114180113.0,
      "step": 65820
    },
    {
      "entropy": 5.961768102645874,
      "epoch": 5.12137716397588,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026910279834318855,
      "loss": 4.8697,
      "mean_token_accuracy": 0.22429970204830169,
      "num_tokens": 114188354.0,
      "step": 65825
    },
    {
      "entropy": 5.9278017520904545,
      "epoch": 5.1217661933475975,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026907509232164964,
      "loss": 4.8544,
      "mean_token_accuracy": 0.2284561201930046,
      "num_tokens": 114197744.0,
      "step": 65830
    },
    {
      "entropy": 5.957499980926514,
      "epoch": 5.122155222719315,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026904738639001163,
      "loss": 4.9235,
      "mean_token_accuracy": 0.22117093354463577,
      "num_tokens": 114206779.0,
      "step": 65835
    },
    {
      "entropy": 5.951617479324341,
      "epoch": 5.122544252091033,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002690196805486952,
      "loss": 5.056,
      "mean_token_accuracy": 0.204644513130188,
      "num_tokens": 114215737.0,
      "step": 65840
    },
    {
      "entropy": 5.918093061447143,
      "epoch": 5.122933281462751,
      "grad_norm": 1.546875,
      "learning_rate": 0.00026899197479812053,
      "loss": 4.935,
      "mean_token_accuracy": 0.222791950404644,
      "num_tokens": 114223686.0,
      "step": 65845
    },
    {
      "entropy": 5.985047245025635,
      "epoch": 5.123322310834468,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002689642691387081,
      "loss": 4.9799,
      "mean_token_accuracy": 0.22100777477025985,
      "num_tokens": 114232562.0,
      "step": 65850
    },
    {
      "entropy": 5.9949455738067625,
      "epoch": 5.123711340206185,
      "grad_norm": 1.5625,
      "learning_rate": 0.00026893656357087833,
      "loss": 4.9554,
      "mean_token_accuracy": 0.2107866257429123,
      "num_tokens": 114240880.0,
      "step": 65855
    },
    {
      "entropy": 5.950732040405273,
      "epoch": 5.124100369577903,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002689088580950515,
      "loss": 4.9151,
      "mean_token_accuracy": 0.22165625989437104,
      "num_tokens": 114249752.0,
      "step": 65860
    },
    {
      "entropy": 5.918896341323853,
      "epoch": 5.1244893989496205,
      "grad_norm": 1.40625,
      "learning_rate": 0.000268881152711648,
      "loss": 4.9838,
      "mean_token_accuracy": 0.2099757432937622,
      "num_tokens": 114258637.0,
      "step": 65865
    },
    {
      "entropy": 6.021253204345703,
      "epoch": 5.124878428321338,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002688534474210884,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21441469341516495,
      "num_tokens": 114266966.0,
      "step": 65870
    },
    {
      "entropy": 5.976773023605347,
      "epoch": 5.125267457693056,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026882574222379283,
      "loss": 4.9865,
      "mean_token_accuracy": 0.20911132395267487,
      "num_tokens": 114276268.0,
      "step": 65875
    },
    {
      "entropy": 5.992915105819702,
      "epoch": 5.125656487064774,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026879803712018187,
      "loss": 5.0585,
      "mean_token_accuracy": 0.21117571592330933,
      "num_tokens": 114284819.0,
      "step": 65880
    },
    {
      "entropy": 5.965303087234497,
      "epoch": 5.126045516436491,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002687703321106758,
      "loss": 4.9386,
      "mean_token_accuracy": 0.22038483917713164,
      "num_tokens": 114294184.0,
      "step": 65885
    },
    {
      "entropy": 6.015755605697632,
      "epoch": 5.126434545808208,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002687426271956951,
      "loss": 5.0141,
      "mean_token_accuracy": 0.2122604012489319,
      "num_tokens": 114302812.0,
      "step": 65890
    },
    {
      "entropy": 5.946567392349243,
      "epoch": 5.126823575179926,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026871492237566,
      "loss": 4.8991,
      "mean_token_accuracy": 0.22335785031318664,
      "num_tokens": 114312283.0,
      "step": 65895
    },
    {
      "entropy": 5.877245092391968,
      "epoch": 5.1272126045516435,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026868721765099086,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2288868635892868,
      "num_tokens": 114320488.0,
      "step": 65900
    },
    {
      "entropy": 5.90310754776001,
      "epoch": 5.127601633923361,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002686595130221082,
      "loss": 4.961,
      "mean_token_accuracy": 0.21879468858242035,
      "num_tokens": 114329097.0,
      "step": 65905
    },
    {
      "entropy": 5.903626728057861,
      "epoch": 5.127990663295079,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026863180848943244,
      "loss": 4.9478,
      "mean_token_accuracy": 0.21968848556280135,
      "num_tokens": 114338146.0,
      "step": 65910
    },
    {
      "entropy": 5.870364618301392,
      "epoch": 5.128379692666797,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002686041040533838,
      "loss": 4.9241,
      "mean_token_accuracy": 0.21870217174291612,
      "num_tokens": 114346563.0,
      "step": 65915
    },
    {
      "entropy": 5.961306858062744,
      "epoch": 5.128768722038514,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002685763997143827,
      "loss": 4.9981,
      "mean_token_accuracy": 0.21138186007738113,
      "num_tokens": 114355067.0,
      "step": 65920
    },
    {
      "entropy": 5.888578176498413,
      "epoch": 5.129157751410231,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026854869547284947,
      "loss": 4.9322,
      "mean_token_accuracy": 0.22094670534133912,
      "num_tokens": 114363983.0,
      "step": 65925
    },
    {
      "entropy": 5.934127473831177,
      "epoch": 5.129546780781949,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026852099132920455,
      "loss": 4.8958,
      "mean_token_accuracy": 0.21835627853870393,
      "num_tokens": 114373515.0,
      "step": 65930
    },
    {
      "entropy": 5.954549789428711,
      "epoch": 5.1299358101536665,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002684932872838683,
      "loss": 4.9016,
      "mean_token_accuracy": 0.22577159702777863,
      "num_tokens": 114381732.0,
      "step": 65935
    },
    {
      "entropy": 5.928292608261108,
      "epoch": 5.130324839525384,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000268465583337261,
      "loss": 4.9491,
      "mean_token_accuracy": 0.20849705189466478,
      "num_tokens": 114390450.0,
      "step": 65940
    },
    {
      "entropy": 6.0004151344299315,
      "epoch": 5.130713868897102,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002684378794898031,
      "loss": 5.0769,
      "mean_token_accuracy": 0.2110137477517128,
      "num_tokens": 114399152.0,
      "step": 65945
    },
    {
      "entropy": 5.929229307174682,
      "epoch": 5.13110289826882,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026841017574191496,
      "loss": 4.9074,
      "mean_token_accuracy": 0.22128274738788606,
      "num_tokens": 114408489.0,
      "step": 65950
    },
    {
      "entropy": 6.003842067718506,
      "epoch": 5.131491927640537,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002683824720940168,
      "loss": 5.0762,
      "mean_token_accuracy": 0.20456138253211975,
      "num_tokens": 114417360.0,
      "step": 65955
    },
    {
      "entropy": 5.975010061264038,
      "epoch": 5.131880957012254,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026835476854652926,
      "loss": 4.9738,
      "mean_token_accuracy": 0.21312254667282104,
      "num_tokens": 114426036.0,
      "step": 65960
    },
    {
      "entropy": 5.924704837799072,
      "epoch": 5.132269986383972,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002683270650998724,
      "loss": 4.8584,
      "mean_token_accuracy": 0.2247190520167351,
      "num_tokens": 114434713.0,
      "step": 65965
    },
    {
      "entropy": 5.910692262649536,
      "epoch": 5.1326590157556895,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002682993617544667,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21432085633277892,
      "num_tokens": 114442846.0,
      "step": 65970
    },
    {
      "entropy": 5.90911340713501,
      "epoch": 5.133048045127407,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002682716585107326,
      "loss": 5.0087,
      "mean_token_accuracy": 0.20875125825405122,
      "num_tokens": 114451427.0,
      "step": 65975
    },
    {
      "entropy": 5.974067640304566,
      "epoch": 5.133437074499125,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026824395536909036,
      "loss": 5.0525,
      "mean_token_accuracy": 0.20319728553295135,
      "num_tokens": 114459962.0,
      "step": 65980
    },
    {
      "entropy": 5.967390012741089,
      "epoch": 5.133826103870843,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002682162523299603,
      "loss": 4.9147,
      "mean_token_accuracy": 0.21885674446821213,
      "num_tokens": 114468197.0,
      "step": 65985
    },
    {
      "entropy": 6.011351013183594,
      "epoch": 5.134215133242559,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026818854939376284,
      "loss": 5.0015,
      "mean_token_accuracy": 0.2033700242638588,
      "num_tokens": 114477274.0,
      "step": 65990
    },
    {
      "entropy": 5.9623270511627195,
      "epoch": 5.134604162614277,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026816084656091835,
      "loss": 4.962,
      "mean_token_accuracy": 0.2173716738820076,
      "num_tokens": 114485264.0,
      "step": 65995
    },
    {
      "entropy": 5.9447057247161865,
      "epoch": 5.134993191985995,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00026813314383184714,
      "loss": 4.9443,
      "mean_token_accuracy": 0.22206167876720428,
      "num_tokens": 114494748.0,
      "step": 66000
    },
    {
      "epoch": 5.134993191985995,
      "eval_entropy": 5.642671900282007,
      "eval_loss": 5.057388782501221,
      "eval_mean_token_accuracy": 0.22136148332086633,
      "eval_num_tokens": 114494748.0,
      "eval_runtime": 21.6337,
      "eval_samples_per_second": 1920.982,
      "eval_steps_per_second": 240.134,
      "step": 66000
    },
    {
      "entropy": 5.968369817733764,
      "epoch": 5.1353822213577125,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026810544120696943,
      "loss": 4.9733,
      "mean_token_accuracy": 0.22180901914834977,
      "num_tokens": 114503235.0,
      "step": 66005
    },
    {
      "entropy": 6.004914379119873,
      "epoch": 5.13577125072943,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026807773868670575,
      "loss": 4.9803,
      "mean_token_accuracy": 0.21408244967460632,
      "num_tokens": 114512352.0,
      "step": 66010
    },
    {
      "entropy": 6.001668405532837,
      "epoch": 5.136160280101148,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002680500362714764,
      "loss": 5.0024,
      "mean_token_accuracy": 0.21757737398147584,
      "num_tokens": 114521205.0,
      "step": 66015
    },
    {
      "entropy": 6.045230770111084,
      "epoch": 5.136549309472866,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00026802233396170164,
      "loss": 5.069,
      "mean_token_accuracy": 0.2074189618229866,
      "num_tokens": 114529817.0,
      "step": 66020
    },
    {
      "entropy": 5.948325109481812,
      "epoch": 5.136938338844582,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002679946317578018,
      "loss": 4.9945,
      "mean_token_accuracy": 0.21114526838064193,
      "num_tokens": 114538394.0,
      "step": 66025
    },
    {
      "entropy": 5.960832023620606,
      "epoch": 5.1373273682163,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002679669296601973,
      "loss": 5.0051,
      "mean_token_accuracy": 0.210135056078434,
      "num_tokens": 114547137.0,
      "step": 66030
    },
    {
      "entropy": 5.9174871921539305,
      "epoch": 5.137716397588018,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026793922766930845,
      "loss": 4.9758,
      "mean_token_accuracy": 0.2211844205856323,
      "num_tokens": 114556633.0,
      "step": 66035
    },
    {
      "entropy": 5.956279850006103,
      "epoch": 5.1381054269597355,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026791152578555563,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21473459154367447,
      "num_tokens": 114565982.0,
      "step": 66040
    },
    {
      "entropy": 5.864475393295288,
      "epoch": 5.138494456331453,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00026788382400935913,
      "loss": 4.8049,
      "mean_token_accuracy": 0.22443415373563766,
      "num_tokens": 114574619.0,
      "step": 66045
    },
    {
      "entropy": 5.9222723007202145,
      "epoch": 5.138883485703171,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002678561223411392,
      "loss": 4.9555,
      "mean_token_accuracy": 0.21397764533758162,
      "num_tokens": 114582691.0,
      "step": 66050
    },
    {
      "entropy": 6.023573160171509,
      "epoch": 5.139272515074889,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002678284207813163,
      "loss": 4.9764,
      "mean_token_accuracy": 0.2210636928677559,
      "num_tokens": 114591551.0,
      "step": 66055
    },
    {
      "entropy": 6.0088457584381105,
      "epoch": 5.139661544446605,
      "grad_norm": 1.640625,
      "learning_rate": 0.0002678007193303107,
      "loss": 5.0126,
      "mean_token_accuracy": 0.21663156002759934,
      "num_tokens": 114599874.0,
      "step": 66060
    },
    {
      "entropy": 5.958954000473023,
      "epoch": 5.140050573818323,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002677730179885427,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21804394274950029,
      "num_tokens": 114608011.0,
      "step": 66065
    },
    {
      "entropy": 5.980318880081176,
      "epoch": 5.140439603190041,
      "grad_norm": 1.5625,
      "learning_rate": 0.00026774531675643267,
      "loss": 4.9909,
      "mean_token_accuracy": 0.22311275750398635,
      "num_tokens": 114615980.0,
      "step": 66070
    },
    {
      "entropy": 5.887480354309082,
      "epoch": 5.1408286325617585,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002677176156344009,
      "loss": 4.8976,
      "mean_token_accuracy": 0.22027385532855986,
      "num_tokens": 114624336.0,
      "step": 66075
    },
    {
      "entropy": 5.975999927520752,
      "epoch": 5.141217661933476,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002676899146228678,
      "loss": 5.0531,
      "mean_token_accuracy": 0.21189430356025696,
      "num_tokens": 114633014.0,
      "step": 66080
    },
    {
      "entropy": 6.034994792938233,
      "epoch": 5.141606691305194,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026766221372225354,
      "loss": 5.0215,
      "mean_token_accuracy": 0.20964909344911575,
      "num_tokens": 114641562.0,
      "step": 66085
    },
    {
      "entropy": 5.998503541946411,
      "epoch": 5.141995720676911,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002676345129329785,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21356947720050812,
      "num_tokens": 114650688.0,
      "step": 66090
    },
    {
      "entropy": 6.000461912155151,
      "epoch": 5.142384750048628,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026760681225546296,
      "loss": 5.0109,
      "mean_token_accuracy": 0.20865440219640732,
      "num_tokens": 114658797.0,
      "step": 66095
    },
    {
      "entropy": 6.005948066711426,
      "epoch": 5.142773779420346,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002675791116901273,
      "loss": 4.9878,
      "mean_token_accuracy": 0.21546838581562042,
      "num_tokens": 114667777.0,
      "step": 66100
    },
    {
      "entropy": 6.003689241409302,
      "epoch": 5.143162808792064,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002675514112373918,
      "loss": 5.034,
      "mean_token_accuracy": 0.21209670305252076,
      "num_tokens": 114676523.0,
      "step": 66105
    },
    {
      "entropy": 6.002839279174805,
      "epoch": 5.1435518381637815,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026752371089767686,
      "loss": 4.9499,
      "mean_token_accuracy": 0.22065724283456803,
      "num_tokens": 114686121.0,
      "step": 66110
    },
    {
      "entropy": 5.91845817565918,
      "epoch": 5.143940867535499,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002674960106714026,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22569520622491837,
      "num_tokens": 114694386.0,
      "step": 66115
    },
    {
      "entropy": 5.931867599487305,
      "epoch": 5.144329896907217,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026746831055898953,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21396079659461975,
      "num_tokens": 114702785.0,
      "step": 66120
    },
    {
      "entropy": 5.902122449874878,
      "epoch": 5.144718926278934,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002674406105608578,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22387531250715256,
      "num_tokens": 114711524.0,
      "step": 66125
    },
    {
      "entropy": 5.9908294677734375,
      "epoch": 5.145107955650651,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002674129106774278,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2178006574511528,
      "num_tokens": 114719663.0,
      "step": 66130
    },
    {
      "entropy": 6.004221105575562,
      "epoch": 5.145496985022369,
      "grad_norm": 1.375,
      "learning_rate": 0.00026738521090911984,
      "loss": 4.9175,
      "mean_token_accuracy": 0.22079660296440123,
      "num_tokens": 114728913.0,
      "step": 66135
    },
    {
      "entropy": 5.920045709609985,
      "epoch": 5.145886014394087,
      "grad_norm": 1.375,
      "learning_rate": 0.0002673575112563541,
      "loss": 4.9179,
      "mean_token_accuracy": 0.22297706454992294,
      "num_tokens": 114738130.0,
      "step": 66140
    },
    {
      "entropy": 5.921075010299683,
      "epoch": 5.1462750437658045,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000267329811719551,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21572670638561248,
      "num_tokens": 114746893.0,
      "step": 66145
    },
    {
      "entropy": 5.969542646408081,
      "epoch": 5.146664073137522,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002673021122991308,
      "loss": 4.9218,
      "mean_token_accuracy": 0.2267557978630066,
      "num_tokens": 114755637.0,
      "step": 66150
    },
    {
      "entropy": 5.973942756652832,
      "epoch": 5.14705310250924,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002672744129955138,
      "loss": 4.8864,
      "mean_token_accuracy": 0.223860564827919,
      "num_tokens": 114764767.0,
      "step": 66155
    },
    {
      "entropy": 5.974945020675659,
      "epoch": 5.147442131880957,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002672467138091204,
      "loss": 4.9486,
      "mean_token_accuracy": 0.21212176531553267,
      "num_tokens": 114773013.0,
      "step": 66160
    },
    {
      "entropy": 5.944044017791748,
      "epoch": 5.147831161252674,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002672190147403705,
      "loss": 4.9637,
      "mean_token_accuracy": 0.21131613105535507,
      "num_tokens": 114782361.0,
      "step": 66165
    },
    {
      "entropy": 5.940251159667969,
      "epoch": 5.148220190624392,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002671913157896849,
      "loss": 4.8872,
      "mean_token_accuracy": 0.22422416657209396,
      "num_tokens": 114791425.0,
      "step": 66170
    },
    {
      "entropy": 5.942361688613891,
      "epoch": 5.14860921999611,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026716361695748353,
      "loss": 4.9168,
      "mean_token_accuracy": 0.21838788837194442,
      "num_tokens": 114800260.0,
      "step": 66175
    },
    {
      "entropy": 5.949578189849854,
      "epoch": 5.1489982493678275,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002671359182441868,
      "loss": 4.9799,
      "mean_token_accuracy": 0.21003224998712539,
      "num_tokens": 114809035.0,
      "step": 66180
    },
    {
      "entropy": 5.97370171546936,
      "epoch": 5.149387278739545,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026710821965021507,
      "loss": 5.0371,
      "mean_token_accuracy": 0.21178975254297255,
      "num_tokens": 114817786.0,
      "step": 66185
    },
    {
      "entropy": 5.954459381103516,
      "epoch": 5.149776308111262,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002670805211759886,
      "loss": 4.9101,
      "mean_token_accuracy": 0.221334408223629,
      "num_tokens": 114826065.0,
      "step": 66190
    },
    {
      "entropy": 5.902453136444092,
      "epoch": 5.15016533748298,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002670528228219275,
      "loss": 4.9046,
      "mean_token_accuracy": 0.22179118990898133,
      "num_tokens": 114834337.0,
      "step": 66195
    },
    {
      "entropy": 5.993678426742553,
      "epoch": 5.150554366854697,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00026702512458845224,
      "loss": 4.9555,
      "mean_token_accuracy": 0.22093888819217683,
      "num_tokens": 114843577.0,
      "step": 66200
    },
    {
      "entropy": 5.965053606033325,
      "epoch": 5.150943396226415,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026699742647598294,
      "loss": 5.0306,
      "mean_token_accuracy": 0.21614261567592621,
      "num_tokens": 114852288.0,
      "step": 66205
    },
    {
      "entropy": 6.00840311050415,
      "epoch": 5.151332425598133,
      "grad_norm": 1.546875,
      "learning_rate": 0.00026696972848494005,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21348073631525039,
      "num_tokens": 114861163.0,
      "step": 66210
    },
    {
      "entropy": 5.90447998046875,
      "epoch": 5.1517214549698505,
      "grad_norm": 1.546875,
      "learning_rate": 0.00026694203061574366,
      "loss": 4.9644,
      "mean_token_accuracy": 0.2179018273949623,
      "num_tokens": 114869665.0,
      "step": 66215
    },
    {
      "entropy": 5.8574096202850345,
      "epoch": 5.152110484341568,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026691433286881415,
      "loss": 4.8202,
      "mean_token_accuracy": 0.22898027747869493,
      "num_tokens": 114878296.0,
      "step": 66220
    },
    {
      "entropy": 5.891721630096436,
      "epoch": 5.152499513713285,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002668866352445718,
      "loss": 4.9154,
      "mean_token_accuracy": 0.22308670729398727,
      "num_tokens": 114886879.0,
      "step": 66225
    },
    {
      "entropy": 6.043492031097412,
      "epoch": 5.152888543085003,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002668589377434368,
      "loss": 5.0402,
      "mean_token_accuracy": 0.21238311529159545,
      "num_tokens": 114894831.0,
      "step": 66230
    },
    {
      "entropy": 5.935453176498413,
      "epoch": 5.15327757245672,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002668312403658295,
      "loss": 4.9086,
      "mean_token_accuracy": 0.21476451754570008,
      "num_tokens": 114903648.0,
      "step": 66235
    },
    {
      "entropy": 5.983807754516602,
      "epoch": 5.153666601828438,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026680354311217006,
      "loss": 4.9493,
      "mean_token_accuracy": 0.22308097183704376,
      "num_tokens": 114912207.0,
      "step": 66240
    },
    {
      "entropy": 5.9431437969207765,
      "epoch": 5.154055631200156,
      "grad_norm": 1.5,
      "learning_rate": 0.00026677584598287893,
      "loss": 4.9012,
      "mean_token_accuracy": 0.22648254483938218,
      "num_tokens": 114921326.0,
      "step": 66245
    },
    {
      "entropy": 5.889452695846558,
      "epoch": 5.1544446605718734,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026674814897837616,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2186141535639763,
      "num_tokens": 114929415.0,
      "step": 66250
    },
    {
      "entropy": 5.967493057250977,
      "epoch": 5.154833689943591,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026672045209908215,
      "loss": 4.9648,
      "mean_token_accuracy": 0.21868237406015395,
      "num_tokens": 114939277.0,
      "step": 66255
    },
    {
      "entropy": 5.990393829345703,
      "epoch": 5.155222719315308,
      "grad_norm": 1.5,
      "learning_rate": 0.00026669275534541703,
      "loss": 4.9645,
      "mean_token_accuracy": 0.2190350756049156,
      "num_tokens": 114948241.0,
      "step": 66260
    },
    {
      "entropy": 5.941273260116577,
      "epoch": 5.155611748687026,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026666505871780115,
      "loss": 4.9255,
      "mean_token_accuracy": 0.21823175996541977,
      "num_tokens": 114956792.0,
      "step": 66265
    },
    {
      "entropy": 5.982630586624145,
      "epoch": 5.156000778058743,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002666373622166548,
      "loss": 5.0553,
      "mean_token_accuracy": 0.20897439867258072,
      "num_tokens": 114966032.0,
      "step": 66270
    },
    {
      "entropy": 5.945557594299316,
      "epoch": 5.156389807430461,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026660966584239815,
      "loss": 4.8896,
      "mean_token_accuracy": 0.21634197980165482,
      "num_tokens": 114974321.0,
      "step": 66275
    },
    {
      "entropy": 6.0597845077514645,
      "epoch": 5.156778836802179,
      "grad_norm": 1.5625,
      "learning_rate": 0.00026658196959545145,
      "loss": 5.0644,
      "mean_token_accuracy": 0.2095256567001343,
      "num_tokens": 114982326.0,
      "step": 66280
    },
    {
      "entropy": 6.006787109375,
      "epoch": 5.157167866173896,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026655427347623494,
      "loss": 5.0487,
      "mean_token_accuracy": 0.21075440496206282,
      "num_tokens": 114992277.0,
      "step": 66285
    },
    {
      "entropy": 5.962014293670654,
      "epoch": 5.157556895545614,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002665265774851689,
      "loss": 4.9257,
      "mean_token_accuracy": 0.22361376881599426,
      "num_tokens": 115000612.0,
      "step": 66290
    },
    {
      "entropy": 5.879933023452759,
      "epoch": 5.157945924917331,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026649888162267357,
      "loss": 4.7969,
      "mean_token_accuracy": 0.22443905770778655,
      "num_tokens": 115009538.0,
      "step": 66295
    },
    {
      "entropy": 5.986502265930175,
      "epoch": 5.158334954289049,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002664711858891691,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21941001564264298,
      "num_tokens": 115017632.0,
      "step": 66300
    },
    {
      "entropy": 6.039589691162109,
      "epoch": 5.158723983660766,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002664434902850759,
      "loss": 5.0791,
      "mean_token_accuracy": 0.21063016057014466,
      "num_tokens": 115027333.0,
      "step": 66305
    },
    {
      "entropy": 5.994335412979126,
      "epoch": 5.159113013032484,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026641579481081405,
      "loss": 4.9736,
      "mean_token_accuracy": 0.21265549957752228,
      "num_tokens": 115035466.0,
      "step": 66310
    },
    {
      "entropy": 5.900611448287964,
      "epoch": 5.159502042404202,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026638809946680393,
      "loss": 4.851,
      "mean_token_accuracy": 0.22301319390535354,
      "num_tokens": 115043998.0,
      "step": 66315
    },
    {
      "entropy": 5.964009428024292,
      "epoch": 5.159891071775919,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026636040425346565,
      "loss": 4.9612,
      "mean_token_accuracy": 0.21696593165397643,
      "num_tokens": 115053445.0,
      "step": 66320
    },
    {
      "entropy": 5.9875139713287355,
      "epoch": 5.160280101147636,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002663327091712195,
      "loss": 4.9783,
      "mean_token_accuracy": 0.2124479591846466,
      "num_tokens": 115061633.0,
      "step": 66325
    },
    {
      "entropy": 6.001373624801635,
      "epoch": 5.160669130519354,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002663050142204856,
      "loss": 4.9744,
      "mean_token_accuracy": 0.22460308969020842,
      "num_tokens": 115070276.0,
      "step": 66330
    },
    {
      "entropy": 5.934246492385864,
      "epoch": 5.161058159891072,
      "grad_norm": 1.546875,
      "learning_rate": 0.00026627731940168445,
      "loss": 4.9431,
      "mean_token_accuracy": 0.2148566871881485,
      "num_tokens": 115078765.0,
      "step": 66335
    },
    {
      "entropy": 5.934921216964722,
      "epoch": 5.161447189262789,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026624962471523595,
      "loss": 4.8893,
      "mean_token_accuracy": 0.22312661558389663,
      "num_tokens": 115087609.0,
      "step": 66340
    },
    {
      "entropy": 5.9245726585388185,
      "epoch": 5.161836218634507,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026622193016156054,
      "loss": 4.9165,
      "mean_token_accuracy": 0.22077118009328842,
      "num_tokens": 115096323.0,
      "step": 66345
    },
    {
      "entropy": 5.99429669380188,
      "epoch": 5.162225248006225,
      "grad_norm": 1.5,
      "learning_rate": 0.0002661942357410784,
      "loss": 5.0314,
      "mean_token_accuracy": 0.2137010246515274,
      "num_tokens": 115104566.0,
      "step": 66350
    },
    {
      "entropy": 5.992999935150147,
      "epoch": 5.162614277377942,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002661665414542097,
      "loss": 4.9832,
      "mean_token_accuracy": 0.21974223703145981,
      "num_tokens": 115112712.0,
      "step": 66355
    },
    {
      "entropy": 5.928037023544311,
      "epoch": 5.163003306749659,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026613884730137454,
      "loss": 4.8543,
      "mean_token_accuracy": 0.22582966238260269,
      "num_tokens": 115121071.0,
      "step": 66360
    },
    {
      "entropy": 5.997103357315064,
      "epoch": 5.163392336121377,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002661111532829933,
      "loss": 5.0372,
      "mean_token_accuracy": 0.2067706525325775,
      "num_tokens": 115130042.0,
      "step": 66365
    },
    {
      "entropy": 5.929024505615234,
      "epoch": 5.163781365493095,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026608345939948625,
      "loss": 4.8943,
      "mean_token_accuracy": 0.2258744165301323,
      "num_tokens": 115138446.0,
      "step": 66370
    },
    {
      "entropy": 5.903096771240234,
      "epoch": 5.164170394864812,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002660557656512734,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21583758294582367,
      "num_tokens": 115147764.0,
      "step": 66375
    },
    {
      "entropy": 5.983061647415161,
      "epoch": 5.16455942423653,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026602807203877525,
      "loss": 4.9855,
      "mean_token_accuracy": 0.2136365666985512,
      "num_tokens": 115156147.0,
      "step": 66380
    },
    {
      "entropy": 5.969338750839233,
      "epoch": 5.164948453608248,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026600037856241174,
      "loss": 4.9208,
      "mean_token_accuracy": 0.22514802366495132,
      "num_tokens": 115164319.0,
      "step": 66385
    },
    {
      "entropy": 5.956699800491333,
      "epoch": 5.1653374829799645,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026597268522260317,
      "loss": 4.9856,
      "mean_token_accuracy": 0.20743007957935333,
      "num_tokens": 115172985.0,
      "step": 66390
    },
    {
      "entropy": 5.916711664199829,
      "epoch": 5.165726512351682,
      "grad_norm": 1.5,
      "learning_rate": 0.00026594499201976966,
      "loss": 4.9324,
      "mean_token_accuracy": 0.21323742419481279,
      "num_tokens": 115181551.0,
      "step": 66395
    },
    {
      "entropy": 5.979283571243286,
      "epoch": 5.1661155417234,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026591729895433157,
      "loss": 4.9441,
      "mean_token_accuracy": 0.21716931015253066,
      "num_tokens": 115190170.0,
      "step": 66400
    },
    {
      "entropy": 5.942608165740967,
      "epoch": 5.166504571095118,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026588960602670896,
      "loss": 4.9132,
      "mean_token_accuracy": 0.22490473389625548,
      "num_tokens": 115198565.0,
      "step": 66405
    },
    {
      "entropy": 5.984029722213745,
      "epoch": 5.166893600466835,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026586191323732213,
      "loss": 5.0166,
      "mean_token_accuracy": 0.2181270092725754,
      "num_tokens": 115206741.0,
      "step": 66410
    },
    {
      "entropy": 5.8804222583770756,
      "epoch": 5.167282629838553,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002658342205865912,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22230149656534196,
      "num_tokens": 115215301.0,
      "step": 66415
    },
    {
      "entropy": 5.918810129165649,
      "epoch": 5.167671659210271,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0002658065280749363,
      "loss": 4.8568,
      "mean_token_accuracy": 0.22524704337120055,
      "num_tokens": 115223749.0,
      "step": 66420
    },
    {
      "entropy": 5.864546155929565,
      "epoch": 5.1680606885819875,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026577883570277786,
      "loss": 4.8139,
      "mean_token_accuracy": 0.22768890857696533,
      "num_tokens": 115233161.0,
      "step": 66425
    },
    {
      "entropy": 5.936381530761719,
      "epoch": 5.168449717953705,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00026575114347053574,
      "loss": 4.8872,
      "mean_token_accuracy": 0.22647102326154708,
      "num_tokens": 115241063.0,
      "step": 66430
    },
    {
      "entropy": 5.910263156890869,
      "epoch": 5.168838747325423,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002657234513786304,
      "loss": 4.8631,
      "mean_token_accuracy": 0.21983306854963303,
      "num_tokens": 115249387.0,
      "step": 66435
    },
    {
      "entropy": 5.970818948745728,
      "epoch": 5.169227776697141,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002656957594274819,
      "loss": 5.0335,
      "mean_token_accuracy": 0.20826046615839006,
      "num_tokens": 115258222.0,
      "step": 66440
    },
    {
      "entropy": 5.949391412734985,
      "epoch": 5.169616806068858,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026566806761751043,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22223554104566573,
      "num_tokens": 115266807.0,
      "step": 66445
    },
    {
      "entropy": 5.890896129608154,
      "epoch": 5.170005835440576,
      "grad_norm": 1.5,
      "learning_rate": 0.0002656403759491362,
      "loss": 4.7957,
      "mean_token_accuracy": 0.228914475440979,
      "num_tokens": 115274914.0,
      "step": 66450
    },
    {
      "entropy": 5.903735256195068,
      "epoch": 5.170394864812294,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002656126844227794,
      "loss": 4.9259,
      "mean_token_accuracy": 0.2184283271431923,
      "num_tokens": 115283836.0,
      "step": 66455
    },
    {
      "entropy": 5.958111572265625,
      "epoch": 5.1707838941840105,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026558499303886015,
      "loss": 4.9936,
      "mean_token_accuracy": 0.21513524204492568,
      "num_tokens": 115292684.0,
      "step": 66460
    },
    {
      "entropy": 5.957755994796753,
      "epoch": 5.171172923555728,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026555730179779864,
      "loss": 5.0027,
      "mean_token_accuracy": 0.21319006234407425,
      "num_tokens": 115301219.0,
      "step": 66465
    },
    {
      "entropy": 6.019247007369995,
      "epoch": 5.171561952927446,
      "grad_norm": 1.546875,
      "learning_rate": 0.000265529610700015,
      "loss": 4.932,
      "mean_token_accuracy": 0.22019873261451722,
      "num_tokens": 115310250.0,
      "step": 66470
    },
    {
      "entropy": 5.988768672943115,
      "epoch": 5.171950982299164,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002655019197459295,
      "loss": 4.9117,
      "mean_token_accuracy": 0.23307146430015563,
      "num_tokens": 115318970.0,
      "step": 66475
    },
    {
      "entropy": 5.962932443618774,
      "epoch": 5.172340011670881,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026547422893596223,
      "loss": 5.0213,
      "mean_token_accuracy": 0.21874122321605682,
      "num_tokens": 115327380.0,
      "step": 66480
    },
    {
      "entropy": 5.93153223991394,
      "epoch": 5.172729041042599,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026544653827053336,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21946187764406205,
      "num_tokens": 115335285.0,
      "step": 66485
    },
    {
      "entropy": 6.002415895462036,
      "epoch": 5.173118070414317,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002654188477500631,
      "loss": 5.0273,
      "mean_token_accuracy": 0.21890933364629744,
      "num_tokens": 115344562.0,
      "step": 66490
    },
    {
      "entropy": 6.005094003677368,
      "epoch": 5.1735070997860335,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002653911573749715,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21747490018606186,
      "num_tokens": 115352555.0,
      "step": 66495
    },
    {
      "entropy": 5.9982678413391115,
      "epoch": 5.173896129157751,
      "grad_norm": 1.75,
      "learning_rate": 0.00026536346714567887,
      "loss": 4.9391,
      "mean_token_accuracy": 0.2261232078075409,
      "num_tokens": 115361043.0,
      "step": 66500
    },
    {
      "entropy": 5.971865797042847,
      "epoch": 5.174285158529469,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026533577706260524,
      "loss": 4.9765,
      "mean_token_accuracy": 0.2178919404745102,
      "num_tokens": 115369736.0,
      "step": 66505
    },
    {
      "entropy": 5.909951639175415,
      "epoch": 5.174674187901187,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002653080871261709,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22167007029056549,
      "num_tokens": 115378520.0,
      "step": 66510
    },
    {
      "entropy": 5.97301173210144,
      "epoch": 5.175063217272904,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002652803973367958,
      "loss": 4.9931,
      "mean_token_accuracy": 0.2126220941543579,
      "num_tokens": 115387307.0,
      "step": 66515
    },
    {
      "entropy": 5.907970905303955,
      "epoch": 5.175452246644622,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026525270769490024,
      "loss": 4.9157,
      "mean_token_accuracy": 0.2151072606444359,
      "num_tokens": 115396422.0,
      "step": 66520
    },
    {
      "entropy": 5.90191683769226,
      "epoch": 5.175841276016339,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026522501820090433,
      "loss": 4.9444,
      "mean_token_accuracy": 0.22937846332788467,
      "num_tokens": 115405779.0,
      "step": 66525
    },
    {
      "entropy": 6.033365058898926,
      "epoch": 5.1762303053880565,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00026519732885522826,
      "loss": 5.0013,
      "mean_token_accuracy": 0.21341147720813752,
      "num_tokens": 115414300.0,
      "step": 66530
    },
    {
      "entropy": 5.9769867897033695,
      "epoch": 5.176619334759774,
      "grad_norm": 1.546875,
      "learning_rate": 0.00026516963965829204,
      "loss": 4.8795,
      "mean_token_accuracy": 0.22244765609502792,
      "num_tokens": 115422499.0,
      "step": 66535
    },
    {
      "entropy": 5.994920921325684,
      "epoch": 5.177008364131492,
      "grad_norm": 1.40625,
      "learning_rate": 0.000265141950610516,
      "loss": 5.03,
      "mean_token_accuracy": 0.21086086481809616,
      "num_tokens": 115431478.0,
      "step": 66540
    },
    {
      "entropy": 5.874392890930176,
      "epoch": 5.17739739350321,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002651142617123201,
      "loss": 4.8658,
      "mean_token_accuracy": 0.22605574429035186,
      "num_tokens": 115439963.0,
      "step": 66545
    },
    {
      "entropy": 5.915239953994751,
      "epoch": 5.177786422874927,
      "grad_norm": 1.625,
      "learning_rate": 0.0002650865729641246,
      "loss": 4.9294,
      "mean_token_accuracy": 0.21355801075696945,
      "num_tokens": 115448452.0,
      "step": 66550
    },
    {
      "entropy": 5.96868405342102,
      "epoch": 5.178175452246645,
      "grad_norm": 1.5,
      "learning_rate": 0.0002650588843663495,
      "loss": 5.0023,
      "mean_token_accuracy": 0.2224422886967659,
      "num_tokens": 115457506.0,
      "step": 66555
    },
    {
      "entropy": 5.931577968597412,
      "epoch": 5.178564481618362,
      "grad_norm": 1.4375,
      "learning_rate": 0.000265031195919415,
      "loss": 4.9147,
      "mean_token_accuracy": 0.2216121658682823,
      "num_tokens": 115465736.0,
      "step": 66560
    },
    {
      "entropy": 5.938823175430298,
      "epoch": 5.1789535109900795,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026500350762374115,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21767596900463104,
      "num_tokens": 115474699.0,
      "step": 66565
    },
    {
      "entropy": 5.9521890640258786,
      "epoch": 5.179342540361797,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026497581947974836,
      "loss": 4.9249,
      "mean_token_accuracy": 0.2129475235939026,
      "num_tokens": 115483321.0,
      "step": 66570
    },
    {
      "entropy": 6.012880182266235,
      "epoch": 5.179731569733515,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026494813148785645,
      "loss": 4.9825,
      "mean_token_accuracy": 0.213930843770504,
      "num_tokens": 115492082.0,
      "step": 66575
    },
    {
      "entropy": 5.938001394271851,
      "epoch": 5.180120599105233,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00026492044364848566,
      "loss": 4.8916,
      "mean_token_accuracy": 0.22605274617671967,
      "num_tokens": 115500084.0,
      "step": 66580
    },
    {
      "entropy": 5.878219747543335,
      "epoch": 5.18050962847695,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002648927559620561,
      "loss": 4.8823,
      "mean_token_accuracy": 0.22185494750738144,
      "num_tokens": 115508289.0,
      "step": 66585
    },
    {
      "entropy": 5.9562476634979244,
      "epoch": 5.180898657848668,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002648650684289879,
      "loss": 5.0076,
      "mean_token_accuracy": 0.21302755922079086,
      "num_tokens": 115517227.0,
      "step": 66590
    },
    {
      "entropy": 5.9779986381530765,
      "epoch": 5.181287687220385,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002648373810497011,
      "loss": 5.0034,
      "mean_token_accuracy": 0.21729738563299178,
      "num_tokens": 115526084.0,
      "step": 66595
    },
    {
      "entropy": 5.943472671508789,
      "epoch": 5.1816767165921025,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002648096938246159,
      "loss": 4.9284,
      "mean_token_accuracy": 0.21740686893463135,
      "num_tokens": 115534595.0,
      "step": 66600
    },
    {
      "entropy": 5.934382677078247,
      "epoch": 5.18206574596382,
      "grad_norm": 1.625,
      "learning_rate": 0.0002647820067541524,
      "loss": 4.9009,
      "mean_token_accuracy": 0.22420788407325745,
      "num_tokens": 115543005.0,
      "step": 66605
    },
    {
      "entropy": 5.9620753765106205,
      "epoch": 5.182454775335538,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026475431983873066,
      "loss": 4.9362,
      "mean_token_accuracy": 0.2201591983437538,
      "num_tokens": 115551992.0,
      "step": 66610
    },
    {
      "entropy": 6.005690717697144,
      "epoch": 5.182843804707256,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002647266330787708,
      "loss": 4.9631,
      "mean_token_accuracy": 0.2171798378229141,
      "num_tokens": 115560769.0,
      "step": 66615
    },
    {
      "entropy": 5.949775075912475,
      "epoch": 5.183232834078973,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026469894647469295,
      "loss": 4.9726,
      "mean_token_accuracy": 0.20849032551050187,
      "num_tokens": 115569984.0,
      "step": 66620
    },
    {
      "entropy": 5.980661106109619,
      "epoch": 5.183621863450691,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0002646712600269172,
      "loss": 5.0167,
      "mean_token_accuracy": 0.21600945591926574,
      "num_tokens": 115578757.0,
      "step": 66625
    },
    {
      "entropy": 5.983853578567505,
      "epoch": 5.184010892822408,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002646435737358635,
      "loss": 4.9634,
      "mean_token_accuracy": 0.22297757714986802,
      "num_tokens": 115587234.0,
      "step": 66630
    },
    {
      "entropy": 5.9093116283416744,
      "epoch": 5.1843999221941255,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002646158876019522,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22673479169607164,
      "num_tokens": 115596045.0,
      "step": 66635
    },
    {
      "entropy": 5.9347052574157715,
      "epoch": 5.184788951565843,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002645882016256033,
      "loss": 4.8892,
      "mean_token_accuracy": 0.2210402637720108,
      "num_tokens": 115604579.0,
      "step": 66640
    },
    {
      "entropy": 5.991714572906494,
      "epoch": 5.185177980937561,
      "grad_norm": 1.5,
      "learning_rate": 0.0002645605158072368,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21870239675045014,
      "num_tokens": 115613084.0,
      "step": 66645
    },
    {
      "entropy": 6.024325180053711,
      "epoch": 5.185567010309279,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000264532830147273,
      "loss": 4.98,
      "mean_token_accuracy": 0.22079526036977767,
      "num_tokens": 115622166.0,
      "step": 66650
    },
    {
      "entropy": 5.897892475128174,
      "epoch": 5.185956039680996,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026450514464613177,
      "loss": 4.8619,
      "mean_token_accuracy": 0.2215588942170143,
      "num_tokens": 115630196.0,
      "step": 66655
    },
    {
      "entropy": 5.948202133178711,
      "epoch": 5.186345069052713,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002644774593042332,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21815174221992492,
      "num_tokens": 115638116.0,
      "step": 66660
    },
    {
      "entropy": 5.8935283660888675,
      "epoch": 5.186734098424431,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002644497741219975,
      "loss": 4.9087,
      "mean_token_accuracy": 0.22061051577329635,
      "num_tokens": 115647172.0,
      "step": 66665
    },
    {
      "entropy": 6.000073766708374,
      "epoch": 5.1871231277961485,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002644220890998447,
      "loss": 5.0106,
      "mean_token_accuracy": 0.21943570524454117,
      "num_tokens": 115657129.0,
      "step": 66670
    },
    {
      "entropy": 5.980986881256103,
      "epoch": 5.187512157167866,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026439440423819483,
      "loss": 4.8837,
      "mean_token_accuracy": 0.22461167275905608,
      "num_tokens": 115665933.0,
      "step": 66675
    },
    {
      "entropy": 6.0121807098388675,
      "epoch": 5.187901186539584,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000264366719537468,
      "loss": 4.9767,
      "mean_token_accuracy": 0.2183927997946739,
      "num_tokens": 115674433.0,
      "step": 66680
    },
    {
      "entropy": 5.976059913635254,
      "epoch": 5.188290215911302,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002643390349980843,
      "loss": 5.0432,
      "mean_token_accuracy": 0.20859669297933578,
      "num_tokens": 115683251.0,
      "step": 66685
    },
    {
      "entropy": 5.90155725479126,
      "epoch": 5.188679245283019,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002643113506204637,
      "loss": 4.9292,
      "mean_token_accuracy": 0.2196235477924347,
      "num_tokens": 115692257.0,
      "step": 66690
    },
    {
      "entropy": 5.943556261062622,
      "epoch": 5.189068274654736,
      "grad_norm": 1.625,
      "learning_rate": 0.0002642836664050263,
      "loss": 4.907,
      "mean_token_accuracy": 0.22124728113412856,
      "num_tokens": 115700624.0,
      "step": 66695
    },
    {
      "entropy": 5.924200105667114,
      "epoch": 5.189457304026454,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026425598235219235,
      "loss": 4.917,
      "mean_token_accuracy": 0.22173608988523483,
      "num_tokens": 115709456.0,
      "step": 66700
    },
    {
      "entropy": 5.947666215896606,
      "epoch": 5.1898463333981715,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026422829846238174,
      "loss": 4.9194,
      "mean_token_accuracy": 0.21792084723711014,
      "num_tokens": 115718771.0,
      "step": 66705
    },
    {
      "entropy": 6.025786542892456,
      "epoch": 5.190235362769889,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026420061473601445,
      "loss": 5.034,
      "mean_token_accuracy": 0.20923073291778566,
      "num_tokens": 115727826.0,
      "step": 66710
    },
    {
      "entropy": 6.003018808364868,
      "epoch": 5.190624392141607,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026417293117351077,
      "loss": 5.0249,
      "mean_token_accuracy": 0.21205544471740723,
      "num_tokens": 115736125.0,
      "step": 66715
    },
    {
      "entropy": 6.007673025131226,
      "epoch": 5.191013421513325,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026414524777529055,
      "loss": 4.9824,
      "mean_token_accuracy": 0.21497223675251007,
      "num_tokens": 115744038.0,
      "step": 66720
    },
    {
      "entropy": 5.9365119457244875,
      "epoch": 5.191402450885041,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002641175645417739,
      "loss": 4.9322,
      "mean_token_accuracy": 0.21846683472394943,
      "num_tokens": 115753036.0,
      "step": 66725
    },
    {
      "entropy": 5.916239786148071,
      "epoch": 5.191791480256759,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000264089881473381,
      "loss": 4.8867,
      "mean_token_accuracy": 0.22359254211187363,
      "num_tokens": 115762145.0,
      "step": 66730
    },
    {
      "entropy": 5.972447347640991,
      "epoch": 5.192180509628477,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002640621985705318,
      "loss": 5.0164,
      "mean_token_accuracy": 0.21508548706769942,
      "num_tokens": 115770799.0,
      "step": 66735
    },
    {
      "entropy": 5.975220441818237,
      "epoch": 5.1925695390001945,
      "grad_norm": 1.578125,
      "learning_rate": 0.00026403451583364616,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21522020399570466,
      "num_tokens": 115778652.0,
      "step": 66740
    },
    {
      "entropy": 5.938592386245728,
      "epoch": 5.192958568371912,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002640068332631444,
      "loss": 4.8952,
      "mean_token_accuracy": 0.2193417564034462,
      "num_tokens": 115787262.0,
      "step": 66745
    },
    {
      "entropy": 6.007817554473877,
      "epoch": 5.19334759774363,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002639791508594464,
      "loss": 5.0097,
      "mean_token_accuracy": 0.21527755111455918,
      "num_tokens": 115795622.0,
      "step": 66750
    },
    {
      "entropy": 5.9245460510253904,
      "epoch": 5.193736627115348,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026395146862297225,
      "loss": 4.894,
      "mean_token_accuracy": 0.22461221218109131,
      "num_tokens": 115804107.0,
      "step": 66755
    },
    {
      "entropy": 5.955900716781616,
      "epoch": 5.194125656487064,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000263923786554142,
      "loss": 4.9413,
      "mean_token_accuracy": 0.2210921064019203,
      "num_tokens": 115812442.0,
      "step": 66760
    },
    {
      "entropy": 5.989454317092895,
      "epoch": 5.194514685858782,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002638961046533756,
      "loss": 4.9828,
      "mean_token_accuracy": 0.21563769429922103,
      "num_tokens": 115820296.0,
      "step": 66765
    },
    {
      "entropy": 5.963762092590332,
      "epoch": 5.1949037152305,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002638684229210932,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21823097616434098,
      "num_tokens": 115829280.0,
      "step": 66770
    },
    {
      "entropy": 5.942563152313232,
      "epoch": 5.1952927446022175,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026384074135771474,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21190258115530014,
      "num_tokens": 115838452.0,
      "step": 66775
    },
    {
      "entropy": 5.960718011856079,
      "epoch": 5.195681773973935,
      "grad_norm": 1.5,
      "learning_rate": 0.0002638130599636603,
      "loss": 4.9403,
      "mean_token_accuracy": 0.21989195495843888,
      "num_tokens": 115847382.0,
      "step": 66780
    },
    {
      "entropy": 5.957814931869507,
      "epoch": 5.196070803345653,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026378537873934983,
      "loss": 4.9742,
      "mean_token_accuracy": 0.21336373835802078,
      "num_tokens": 115856818.0,
      "step": 66785
    },
    {
      "entropy": 5.921224164962768,
      "epoch": 5.1964598327173706,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026375769768520344,
      "loss": 4.9269,
      "mean_token_accuracy": 0.2165870651602745,
      "num_tokens": 115865086.0,
      "step": 66790
    },
    {
      "entropy": 6.015837144851685,
      "epoch": 5.196848862089087,
      "grad_norm": 1.609375,
      "learning_rate": 0.000263730016801641,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20953391045331954,
      "num_tokens": 115874091.0,
      "step": 66795
    },
    {
      "entropy": 5.971311092376709,
      "epoch": 5.197237891460805,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002637023360890827,
      "loss": 4.9594,
      "mean_token_accuracy": 0.22122769206762313,
      "num_tokens": 115883227.0,
      "step": 66800
    },
    {
      "entropy": 6.007909154891967,
      "epoch": 5.197626920832523,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026367465554794843,
      "loss": 4.9997,
      "mean_token_accuracy": 0.21437253803014755,
      "num_tokens": 115891808.0,
      "step": 66805
    },
    {
      "entropy": 5.941662454605103,
      "epoch": 5.1980159502042405,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026364697517865834,
      "loss": 4.9003,
      "mean_token_accuracy": 0.22316804826259612,
      "num_tokens": 115899724.0,
      "step": 66810
    },
    {
      "entropy": 5.9394158840179445,
      "epoch": 5.198404979575958,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002636192949816322,
      "loss": 4.9434,
      "mean_token_accuracy": 0.22258237153291702,
      "num_tokens": 115909027.0,
      "step": 66815
    },
    {
      "entropy": 6.005552196502686,
      "epoch": 5.198794008947676,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002635916149572902,
      "loss": 5.006,
      "mean_token_accuracy": 0.21511903405189514,
      "num_tokens": 115916893.0,
      "step": 66820
    },
    {
      "entropy": 5.9230433940887455,
      "epoch": 5.1991830383193935,
      "grad_norm": 1.625,
      "learning_rate": 0.00026356393510605224,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22207140177488327,
      "num_tokens": 115925168.0,
      "step": 66825
    },
    {
      "entropy": 5.926455736160278,
      "epoch": 5.19957206769111,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002635362554283383,
      "loss": 4.9773,
      "mean_token_accuracy": 0.2124258503317833,
      "num_tokens": 115933897.0,
      "step": 66830
    },
    {
      "entropy": 5.941997146606445,
      "epoch": 5.199961097062828,
      "grad_norm": 1.5,
      "learning_rate": 0.0002635085759245686,
      "loss": 4.9243,
      "mean_token_accuracy": 0.2218412473797798,
      "num_tokens": 115942575.0,
      "step": 66835
    },
    {
      "entropy": 5.999353075027466,
      "epoch": 5.200350126434546,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002634808965951629,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22421287894248962,
      "num_tokens": 115951298.0,
      "step": 66840
    },
    {
      "entropy": 5.908012056350708,
      "epoch": 5.2007391558062634,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002634532174405413,
      "loss": 4.9055,
      "mean_token_accuracy": 0.21508170515298844,
      "num_tokens": 115959706.0,
      "step": 66845
    },
    {
      "entropy": 5.912952518463134,
      "epoch": 5.201128185177981,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026342553846112364,
      "loss": 4.9437,
      "mean_token_accuracy": 0.2237588256597519,
      "num_tokens": 115968040.0,
      "step": 66850
    },
    {
      "entropy": 6.011100244522095,
      "epoch": 5.201517214549699,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002633978596573301,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20873406976461412,
      "num_tokens": 115976101.0,
      "step": 66855
    },
    {
      "entropy": 6.003397655487061,
      "epoch": 5.201906243921416,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026337018102958056,
      "loss": 4.9812,
      "mean_token_accuracy": 0.21380960047245026,
      "num_tokens": 115984652.0,
      "step": 66860
    },
    {
      "entropy": 5.9457159519195555,
      "epoch": 5.202295273293133,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000263342502578295,
      "loss": 4.8863,
      "mean_token_accuracy": 0.226133131980896,
      "num_tokens": 115992423.0,
      "step": 66865
    },
    {
      "entropy": 5.931352519989014,
      "epoch": 5.202684302664851,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026331482430389343,
      "loss": 5.0071,
      "mean_token_accuracy": 0.21697636395692826,
      "num_tokens": 116001992.0,
      "step": 66870
    },
    {
      "entropy": 5.934729671478271,
      "epoch": 5.203073332036569,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026328714620679576,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22169479727745056,
      "num_tokens": 116010434.0,
      "step": 66875
    },
    {
      "entropy": 5.938559484481812,
      "epoch": 5.203462361408286,
      "grad_norm": 1.5625,
      "learning_rate": 0.00026325946828742203,
      "loss": 4.9291,
      "mean_token_accuracy": 0.21455109715461732,
      "num_tokens": 116019460.0,
      "step": 66880
    },
    {
      "entropy": 5.947332859039307,
      "epoch": 5.203851390780004,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026323179054619215,
      "loss": 5.0217,
      "mean_token_accuracy": 0.2153773933649063,
      "num_tokens": 116028324.0,
      "step": 66885
    },
    {
      "entropy": 5.9788295269012455,
      "epoch": 5.204240420151722,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002632041129835262,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2111601337790489,
      "num_tokens": 116037384.0,
      "step": 66890
    },
    {
      "entropy": 6.005895519256592,
      "epoch": 5.204629449523439,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00026317643559984396,
      "loss": 5.0515,
      "mean_token_accuracy": 0.2141922652721405,
      "num_tokens": 116046103.0,
      "step": 66895
    },
    {
      "entropy": 6.047944164276123,
      "epoch": 5.205018478895156,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002631487583955655,
      "loss": 4.9906,
      "mean_token_accuracy": 0.21233060359954833,
      "num_tokens": 116055504.0,
      "step": 66900
    },
    {
      "entropy": 5.995011663436889,
      "epoch": 5.205407508266874,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00026312108137111086,
      "loss": 4.9011,
      "mean_token_accuracy": 0.21788831055164337,
      "num_tokens": 116063481.0,
      "step": 66905
    },
    {
      "entropy": 5.888275861740112,
      "epoch": 5.205796537638592,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002630934045268998,
      "loss": 4.8428,
      "mean_token_accuracy": 0.22547418922185897,
      "num_tokens": 116072068.0,
      "step": 66910
    },
    {
      "entropy": 5.9567742347717285,
      "epoch": 5.206185567010309,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002630657278633525,
      "loss": 4.9315,
      "mean_token_accuracy": 0.21005211472511293,
      "num_tokens": 116079605.0,
      "step": 66915
    },
    {
      "entropy": 5.874494981765747,
      "epoch": 5.206574596382027,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002630380513808887,
      "loss": 4.8552,
      "mean_token_accuracy": 0.23065195530653,
      "num_tokens": 116088615.0,
      "step": 66920
    },
    {
      "entropy": 5.975090885162354,
      "epoch": 5.206963625753745,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002630103750799284,
      "loss": 5.0048,
      "mean_token_accuracy": 0.2065598711371422,
      "num_tokens": 116097010.0,
      "step": 66925
    },
    {
      "entropy": 5.955994606018066,
      "epoch": 5.207352655125462,
      "grad_norm": 1.59375,
      "learning_rate": 0.00026298269896089164,
      "loss": 4.9539,
      "mean_token_accuracy": 0.21543947756290435,
      "num_tokens": 116106069.0,
      "step": 66930
    },
    {
      "entropy": 5.949639892578125,
      "epoch": 5.207741684497179,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026295502302419826,
      "loss": 4.8892,
      "mean_token_accuracy": 0.2266635403037071,
      "num_tokens": 116114482.0,
      "step": 66935
    },
    {
      "entropy": 5.982015895843506,
      "epoch": 5.208130713868897,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002629273472702683,
      "loss": 4.999,
      "mean_token_accuracy": 0.21833478063344955,
      "num_tokens": 116123529.0,
      "step": 66940
    },
    {
      "entropy": 5.901161193847656,
      "epoch": 5.208519743240615,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002628996716995215,
      "loss": 4.9,
      "mean_token_accuracy": 0.2245624005794525,
      "num_tokens": 116132118.0,
      "step": 66945
    },
    {
      "entropy": 5.947268676757813,
      "epoch": 5.208908772612332,
      "grad_norm": 1.578125,
      "learning_rate": 0.000262871996312378,
      "loss": 5.0108,
      "mean_token_accuracy": 0.21327919363975525,
      "num_tokens": 116140468.0,
      "step": 66950
    },
    {
      "entropy": 6.007161235809326,
      "epoch": 5.20929780198405,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026284432110925765,
      "loss": 5.0185,
      "mean_token_accuracy": 0.21165337711572646,
      "num_tokens": 116148719.0,
      "step": 66955
    },
    {
      "entropy": 6.026771354675293,
      "epoch": 5.209686831355767,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026281664609058037,
      "loss": 4.9881,
      "mean_token_accuracy": 0.2111056089401245,
      "num_tokens": 116157738.0,
      "step": 66960
    },
    {
      "entropy": 5.8964488983154295,
      "epoch": 5.210075860727485,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002627889712567661,
      "loss": 4.906,
      "mean_token_accuracy": 0.21750743091106414,
      "num_tokens": 116166860.0,
      "step": 66965
    },
    {
      "entropy": 5.956106805801392,
      "epoch": 5.210464890099202,
      "grad_norm": 1.734375,
      "learning_rate": 0.00026276129660823475,
      "loss": 4.9311,
      "mean_token_accuracy": 0.2220573380589485,
      "num_tokens": 116174772.0,
      "step": 66970
    },
    {
      "entropy": 5.974246835708618,
      "epoch": 5.21085391947092,
      "grad_norm": 1.5,
      "learning_rate": 0.00026273362214540623,
      "loss": 5.0049,
      "mean_token_accuracy": 0.2151297375559807,
      "num_tokens": 116184007.0,
      "step": 66975
    },
    {
      "entropy": 5.9906055450439455,
      "epoch": 5.211242948842638,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002627059478687005,
      "loss": 4.9829,
      "mean_token_accuracy": 0.21531809270381927,
      "num_tokens": 116193049.0,
      "step": 66980
    },
    {
      "entropy": 6.015655994415283,
      "epoch": 5.211631978214355,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002626782737785374,
      "loss": 5.0276,
      "mean_token_accuracy": 0.21609100699424744,
      "num_tokens": 116201474.0,
      "step": 66985
    },
    {
      "entropy": 6.037283182144165,
      "epoch": 5.212021007586073,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00026265059987533697,
      "loss": 5.0254,
      "mean_token_accuracy": 0.21452055424451827,
      "num_tokens": 116210253.0,
      "step": 66990
    },
    {
      "entropy": 6.016984462738037,
      "epoch": 5.21241003695779,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026262292615951893,
      "loss": 5.0623,
      "mean_token_accuracy": 0.2129225954413414,
      "num_tokens": 116218784.0,
      "step": 66995
    },
    {
      "entropy": 5.939670562744141,
      "epoch": 5.212799066329508,
      "grad_norm": 1.5,
      "learning_rate": 0.0002625952526315034,
      "loss": 4.8953,
      "mean_token_accuracy": 0.22594502717256545,
      "num_tokens": 116227078.0,
      "step": 67000
    },
    {
      "entropy": 5.97763147354126,
      "epoch": 5.213188095701225,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002625675792917101,
      "loss": 4.8759,
      "mean_token_accuracy": 0.2284654900431633,
      "num_tokens": 116235840.0,
      "step": 67005
    },
    {
      "entropy": 5.961347007751465,
      "epoch": 5.213577125072943,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000262539906140559,
      "loss": 4.9745,
      "mean_token_accuracy": 0.21392288506031037,
      "num_tokens": 116244686.0,
      "step": 67010
    },
    {
      "entropy": 5.864618492126465,
      "epoch": 5.213966154444661,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026251223317847003,
      "loss": 4.891,
      "mean_token_accuracy": 0.22194795906543732,
      "num_tokens": 116252836.0,
      "step": 67015
    },
    {
      "entropy": 5.873418521881104,
      "epoch": 5.214355183816378,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000262484560405863,
      "loss": 4.856,
      "mean_token_accuracy": 0.22879645377397537,
      "num_tokens": 116261619.0,
      "step": 67020
    },
    {
      "entropy": 5.9820723056793215,
      "epoch": 5.214744213188096,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026245688782315787,
      "loss": 4.9586,
      "mean_token_accuracy": 0.20923326015472413,
      "num_tokens": 116269960.0,
      "step": 67025
    },
    {
      "entropy": 6.012358331680298,
      "epoch": 5.215133242559813,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026242921543077435,
      "loss": 4.9005,
      "mean_token_accuracy": 0.22624083310365678,
      "num_tokens": 116278286.0,
      "step": 67030
    },
    {
      "entropy": 5.96365647315979,
      "epoch": 5.215522271931531,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002624015432291326,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22628503143787385,
      "num_tokens": 116286798.0,
      "step": 67035
    },
    {
      "entropy": 5.926325464248658,
      "epoch": 5.215911301303248,
      "grad_norm": 1.625,
      "learning_rate": 0.0002623738712186525,
      "loss": 4.9519,
      "mean_token_accuracy": 0.22156002521514892,
      "num_tokens": 116295279.0,
      "step": 67040
    },
    {
      "entropy": 5.920442390441894,
      "epoch": 5.216300330674966,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026234619939975365,
      "loss": 4.9375,
      "mean_token_accuracy": 0.22073664516210556,
      "num_tokens": 116303788.0,
      "step": 67045
    },
    {
      "entropy": 5.946451711654663,
      "epoch": 5.216689360046684,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026231852777285606,
      "loss": 4.8955,
      "mean_token_accuracy": 0.22114986032247544,
      "num_tokens": 116312023.0,
      "step": 67050
    },
    {
      "entropy": 5.934589195251465,
      "epoch": 5.217078389418401,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026229085633837975,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21604447811841965,
      "num_tokens": 116320755.0,
      "step": 67055
    },
    {
      "entropy": 5.980952358245849,
      "epoch": 5.217467418790118,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002622631850967444,
      "loss": 5.0162,
      "mean_token_accuracy": 0.215207239985466,
      "num_tokens": 116329259.0,
      "step": 67060
    },
    {
      "entropy": 5.899078989028931,
      "epoch": 5.217856448161836,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026223551404837,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22085072696208954,
      "num_tokens": 116337573.0,
      "step": 67065
    },
    {
      "entropy": 5.946717405319214,
      "epoch": 5.218245477533554,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026220784319367626,
      "loss": 4.9289,
      "mean_token_accuracy": 0.2237103119492531,
      "num_tokens": 116346249.0,
      "step": 67070
    },
    {
      "entropy": 5.991045951843262,
      "epoch": 5.218634506905271,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002621801725330832,
      "loss": 4.9914,
      "mean_token_accuracy": 0.21662231981754304,
      "num_tokens": 116354670.0,
      "step": 67075
    },
    {
      "entropy": 6.003362941741943,
      "epoch": 5.219023536276989,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002621525020670106,
      "loss": 5.0002,
      "mean_token_accuracy": 0.2165423408150673,
      "num_tokens": 116363578.0,
      "step": 67080
    },
    {
      "entropy": 5.994909048080444,
      "epoch": 5.219412565648707,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002621248317958784,
      "loss": 4.9388,
      "mean_token_accuracy": 0.22562653124332427,
      "num_tokens": 116372639.0,
      "step": 67085
    },
    {
      "entropy": 6.001397514343262,
      "epoch": 5.219801595020424,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026209716172010636,
      "loss": 4.9514,
      "mean_token_accuracy": 0.2220144286751747,
      "num_tokens": 116380808.0,
      "step": 67090
    },
    {
      "entropy": 5.956270027160644,
      "epoch": 5.220190624392141,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002620694918401142,
      "loss": 4.9686,
      "mean_token_accuracy": 0.2127075105905533,
      "num_tokens": 116389456.0,
      "step": 67095
    },
    {
      "entropy": 6.0104651927948,
      "epoch": 5.220579653763859,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000262041822156322,
      "loss": 5.0204,
      "mean_token_accuracy": 0.2180896982550621,
      "num_tokens": 116397634.0,
      "step": 67100
    },
    {
      "entropy": 5.979139137268066,
      "epoch": 5.220968683135577,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002620141526691496,
      "loss": 4.9998,
      "mean_token_accuracy": 0.21423858106136323,
      "num_tokens": 116407068.0,
      "step": 67105
    },
    {
      "entropy": 6.021339654922485,
      "epoch": 5.221357712507294,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026198648337901674,
      "loss": 5.0054,
      "mean_token_accuracy": 0.22253150641918182,
      "num_tokens": 116415601.0,
      "step": 67110
    },
    {
      "entropy": 5.904582977294922,
      "epoch": 5.221746741879012,
      "grad_norm": 1.671875,
      "learning_rate": 0.00026195881428634334,
      "loss": 4.8033,
      "mean_token_accuracy": 0.23814280182123185,
      "num_tokens": 116424095.0,
      "step": 67115
    },
    {
      "entropy": 5.884514188766479,
      "epoch": 5.22213577125073,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026193114539154903,
      "loss": 4.8265,
      "mean_token_accuracy": 0.2234618678689003,
      "num_tokens": 116432581.0,
      "step": 67120
    },
    {
      "entropy": 5.992266035079956,
      "epoch": 5.222524800622447,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002619034766950539,
      "loss": 4.9407,
      "mean_token_accuracy": 0.21185623109340668,
      "num_tokens": 116440936.0,
      "step": 67125
    },
    {
      "entropy": 5.998810338973999,
      "epoch": 5.222913829994164,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002618758081972776,
      "loss": 4.9942,
      "mean_token_accuracy": 0.21485570520162584,
      "num_tokens": 116449273.0,
      "step": 67130
    },
    {
      "entropy": 5.968427944183349,
      "epoch": 5.223302859365882,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002618481398986401,
      "loss": 4.9885,
      "mean_token_accuracy": 0.21848909556865692,
      "num_tokens": 116458073.0,
      "step": 67135
    },
    {
      "entropy": 5.927435493469238,
      "epoch": 5.2236918887376,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026182047179956107,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22533820569515228,
      "num_tokens": 116466086.0,
      "step": 67140
    },
    {
      "entropy": 5.953805351257325,
      "epoch": 5.224080918109317,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026179280390046047,
      "loss": 4.9111,
      "mean_token_accuracy": 0.2251986861228943,
      "num_tokens": 116474496.0,
      "step": 67145
    },
    {
      "entropy": 5.9867737770080565,
      "epoch": 5.224469947481035,
      "grad_norm": 1.59375,
      "learning_rate": 0.00026176513620175795,
      "loss": 5.029,
      "mean_token_accuracy": 0.2086343139410019,
      "num_tokens": 116483606.0,
      "step": 67150
    },
    {
      "entropy": 6.024207019805909,
      "epoch": 5.224858976852753,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002617374687038735,
      "loss": 5.041,
      "mean_token_accuracy": 0.21349333822727204,
      "num_tokens": 116493033.0,
      "step": 67155
    },
    {
      "entropy": 5.94017391204834,
      "epoch": 5.22524800622447,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026170980140722677,
      "loss": 4.8594,
      "mean_token_accuracy": 0.229592065513134,
      "num_tokens": 116501054.0,
      "step": 67160
    },
    {
      "entropy": 6.002553558349609,
      "epoch": 5.225637035596187,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026168213431223777,
      "loss": 4.9834,
      "mean_token_accuracy": 0.20949667394161225,
      "num_tokens": 116509681.0,
      "step": 67165
    },
    {
      "entropy": 5.993343544006348,
      "epoch": 5.226026064967905,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00026165446741932606,
      "loss": 4.979,
      "mean_token_accuracy": 0.21300880908966063,
      "num_tokens": 116518490.0,
      "step": 67170
    },
    {
      "entropy": 5.985824632644653,
      "epoch": 5.226415094339623,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00026162680072891166,
      "loss": 5.0,
      "mean_token_accuracy": 0.2130703240633011,
      "num_tokens": 116526598.0,
      "step": 67175
    },
    {
      "entropy": 6.045921373367309,
      "epoch": 5.22680412371134,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026159913424141426,
      "loss": 5.0965,
      "mean_token_accuracy": 0.20459864884614945,
      "num_tokens": 116535650.0,
      "step": 67180
    },
    {
      "entropy": 5.897486448287964,
      "epoch": 5.227193153083058,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002615714679572536,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22449268102645875,
      "num_tokens": 116544569.0,
      "step": 67185
    },
    {
      "entropy": 5.971323442459107,
      "epoch": 5.227582182454776,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026154380187684963,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2131018415093422,
      "num_tokens": 116552351.0,
      "step": 67190
    },
    {
      "entropy": 5.900402307510376,
      "epoch": 5.2279712118264925,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000261516136000622,
      "loss": 4.8786,
      "mean_token_accuracy": 0.2262655422091484,
      "num_tokens": 116561761.0,
      "step": 67195
    },
    {
      "entropy": 6.011790180206299,
      "epoch": 5.22836024119821,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002614884703289906,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21118742376565933,
      "num_tokens": 116570573.0,
      "step": 67200
    },
    {
      "entropy": 5.918563413619995,
      "epoch": 5.228749270569928,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026146080486237513,
      "loss": 4.774,
      "mean_token_accuracy": 0.23690352886915206,
      "num_tokens": 116580840.0,
      "step": 67205
    },
    {
      "entropy": 5.945854806900025,
      "epoch": 5.229138299941646,
      "grad_norm": 1.5625,
      "learning_rate": 0.00026143313960119534,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21707655787467955,
      "num_tokens": 116589657.0,
      "step": 67210
    },
    {
      "entropy": 5.968327331542969,
      "epoch": 5.229527329313363,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002614054745458711,
      "loss": 4.9454,
      "mean_token_accuracy": 0.21669345647096633,
      "num_tokens": 116597796.0,
      "step": 67215
    },
    {
      "entropy": 5.872777938842773,
      "epoch": 5.229916358685081,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026137780969682215,
      "loss": 4.8186,
      "mean_token_accuracy": 0.22715088129043579,
      "num_tokens": 116606750.0,
      "step": 67220
    },
    {
      "entropy": 5.893414402008057,
      "epoch": 5.230305388056799,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002613501450544683,
      "loss": 4.889,
      "mean_token_accuracy": 0.21555591076612474,
      "num_tokens": 116614552.0,
      "step": 67225
    },
    {
      "entropy": 5.861094903945923,
      "epoch": 5.2306944174285155,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002613224806192291,
      "loss": 4.7745,
      "mean_token_accuracy": 0.23180656433105468,
      "num_tokens": 116622669.0,
      "step": 67230
    },
    {
      "entropy": 5.946749258041382,
      "epoch": 5.231083446800233,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002612948163915247,
      "loss": 4.9572,
      "mean_token_accuracy": 0.21943146288394927,
      "num_tokens": 116631828.0,
      "step": 67235
    },
    {
      "entropy": 5.918754577636719,
      "epoch": 5.231472476171951,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002612671523717745,
      "loss": 4.8552,
      "mean_token_accuracy": 0.21917740851640702,
      "num_tokens": 116640482.0,
      "step": 67240
    },
    {
      "entropy": 6.014663791656494,
      "epoch": 5.231861505543669,
      "grad_norm": 1.5,
      "learning_rate": 0.0002612394885603985,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21510806679725647,
      "num_tokens": 116649029.0,
      "step": 67245
    },
    {
      "entropy": 5.906226015090942,
      "epoch": 5.232250534915386,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002612118249578163,
      "loss": 4.9222,
      "mean_token_accuracy": 0.22507931143045426,
      "num_tokens": 116657698.0,
      "step": 67250
    },
    {
      "entropy": 5.910842227935791,
      "epoch": 5.232639564287104,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002611841615644477,
      "loss": 4.9564,
      "mean_token_accuracy": 0.21994110345840454,
      "num_tokens": 116665890.0,
      "step": 67255
    },
    {
      "entropy": 5.935106229782105,
      "epoch": 5.233028593658822,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00026115649838071253,
      "loss": 4.91,
      "mean_token_accuracy": 0.2199905350804329,
      "num_tokens": 116674075.0,
      "step": 67260
    },
    {
      "entropy": 6.024548768997192,
      "epoch": 5.2334176230305385,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002611288354070304,
      "loss": 5.0898,
      "mean_token_accuracy": 0.20910734236240386,
      "num_tokens": 116683119.0,
      "step": 67265
    },
    {
      "entropy": 5.94524655342102,
      "epoch": 5.233806652402256,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002611011726438212,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2212102919816971,
      "num_tokens": 116690931.0,
      "step": 67270
    },
    {
      "entropy": 5.852141571044922,
      "epoch": 5.234195681773974,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026107351009150453,
      "loss": 4.8284,
      "mean_token_accuracy": 0.2245680034160614,
      "num_tokens": 116699415.0,
      "step": 67275
    },
    {
      "entropy": 5.892734718322754,
      "epoch": 5.234584711145692,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002610458477505002,
      "loss": 4.917,
      "mean_token_accuracy": 0.2279587984085083,
      "num_tokens": 116707807.0,
      "step": 67280
    },
    {
      "entropy": 5.921682262420655,
      "epoch": 5.234973740517409,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002610181856212279,
      "loss": 4.85,
      "mean_token_accuracy": 0.2278549477458,
      "num_tokens": 116716941.0,
      "step": 67285
    },
    {
      "entropy": 5.976714754104615,
      "epoch": 5.235362769889127,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002609905237041074,
      "loss": 4.9689,
      "mean_token_accuracy": 0.21142105162143707,
      "num_tokens": 116725398.0,
      "step": 67290
    },
    {
      "entropy": 6.031628084182739,
      "epoch": 5.235751799260844,
      "grad_norm": 1.5,
      "learning_rate": 0.0002609628619995582,
      "loss": 5.0367,
      "mean_token_accuracy": 0.21499659717082978,
      "num_tokens": 116733652.0,
      "step": 67295
    },
    {
      "entropy": 5.958323812484741,
      "epoch": 5.2361408286325615,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026093520050800046,
      "loss": 4.9144,
      "mean_token_accuracy": 0.21862298995256424,
      "num_tokens": 116741822.0,
      "step": 67300
    },
    {
      "entropy": 5.993869209289551,
      "epoch": 5.236529858004279,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002609075392298536,
      "loss": 5.0001,
      "mean_token_accuracy": 0.21936078518629074,
      "num_tokens": 116750338.0,
      "step": 67305
    },
    {
      "entropy": 5.913096809387207,
      "epoch": 5.236918887375997,
      "grad_norm": 1.5,
      "learning_rate": 0.0002608798781655374,
      "loss": 4.8893,
      "mean_token_accuracy": 0.222600582242012,
      "num_tokens": 116758587.0,
      "step": 67310
    },
    {
      "entropy": 5.976931142807007,
      "epoch": 5.237307916747715,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002608522173154716,
      "loss": 5.0059,
      "mean_token_accuracy": 0.2131909266114235,
      "num_tokens": 116767255.0,
      "step": 67315
    },
    {
      "entropy": 5.943133735656739,
      "epoch": 5.237696946119432,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002608245566800759,
      "loss": 4.9571,
      "mean_token_accuracy": 0.21705740094184875,
      "num_tokens": 116775174.0,
      "step": 67320
    },
    {
      "entropy": 5.9356146335601805,
      "epoch": 5.23808597549115,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026079689625976995,
      "loss": 4.9574,
      "mean_token_accuracy": 0.21710429191589356,
      "num_tokens": 116783237.0,
      "step": 67325
    },
    {
      "entropy": 5.954853534698486,
      "epoch": 5.238475004862867,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026076923605497347,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21751187443733216,
      "num_tokens": 116793143.0,
      "step": 67330
    },
    {
      "entropy": 5.976344776153565,
      "epoch": 5.2388640342345845,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00026074157606610626,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2237047255039215,
      "num_tokens": 116801130.0,
      "step": 67335
    },
    {
      "entropy": 5.9367701530456545,
      "epoch": 5.239253063606302,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026071391629358786,
      "loss": 4.9263,
      "mean_token_accuracy": 0.2184154912829399,
      "num_tokens": 116809641.0,
      "step": 67340
    },
    {
      "entropy": 5.95395565032959,
      "epoch": 5.23964209297802,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026068625673783807,
      "loss": 4.937,
      "mean_token_accuracy": 0.2190174415707588,
      "num_tokens": 116817569.0,
      "step": 67345
    },
    {
      "entropy": 5.929007244110108,
      "epoch": 5.240031122349738,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026065859739927657,
      "loss": 4.9016,
      "mean_token_accuracy": 0.22254942804574968,
      "num_tokens": 116825676.0,
      "step": 67350
    },
    {
      "entropy": 6.02333698272705,
      "epoch": 5.240420151721455,
      "grad_norm": 1.4375,
      "learning_rate": 0.000260630938278323,
      "loss": 5.0952,
      "mean_token_accuracy": 0.20371092855930328,
      "num_tokens": 116835046.0,
      "step": 67355
    },
    {
      "entropy": 6.011357641220092,
      "epoch": 5.240809181093173,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026060327937539704,
      "loss": 4.9962,
      "mean_token_accuracy": 0.2141246572136879,
      "num_tokens": 116844037.0,
      "step": 67360
    },
    {
      "entropy": 5.956971740722656,
      "epoch": 5.24119821046489,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026057562069091845,
      "loss": 4.9588,
      "mean_token_accuracy": 0.21505456566810607,
      "num_tokens": 116852680.0,
      "step": 67365
    },
    {
      "entropy": 6.005250358581543,
      "epoch": 5.2415872398366075,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026054796222530684,
      "loss": 4.9897,
      "mean_token_accuracy": 0.21494847238063813,
      "num_tokens": 116862012.0,
      "step": 67370
    },
    {
      "entropy": 6.019654035568237,
      "epoch": 5.241976269208325,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002605203039789819,
      "loss": 4.9569,
      "mean_token_accuracy": 0.22150174528360367,
      "num_tokens": 116870418.0,
      "step": 67375
    },
    {
      "entropy": 6.025184154510498,
      "epoch": 5.242365298580043,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002604926459523633,
      "loss": 5.0162,
      "mean_token_accuracy": 0.2078032523393631,
      "num_tokens": 116879377.0,
      "step": 67380
    },
    {
      "entropy": 6.0381426334381105,
      "epoch": 5.2427543279517606,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026046498814587076,
      "loss": 4.9666,
      "mean_token_accuracy": 0.21931336522102357,
      "num_tokens": 116888045.0,
      "step": 67385
    },
    {
      "entropy": 5.973710775375366,
      "epoch": 5.243143357323478,
      "grad_norm": 1.546875,
      "learning_rate": 0.00026043733055992386,
      "loss": 4.959,
      "mean_token_accuracy": 0.2144218549132347,
      "num_tokens": 116896304.0,
      "step": 67390
    },
    {
      "entropy": 5.956200218200683,
      "epoch": 5.243532386695195,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026040967319494225,
      "loss": 5.0165,
      "mean_token_accuracy": 0.21816113442182541,
      "num_tokens": 116905208.0,
      "step": 67395
    },
    {
      "entropy": 6.012505912780762,
      "epoch": 5.243921416066913,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026038201605134567,
      "loss": 4.9949,
      "mean_token_accuracy": 0.21771894991397858,
      "num_tokens": 116913550.0,
      "step": 67400
    },
    {
      "entropy": 5.933548212051392,
      "epoch": 5.2443104454386305,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026035435912955375,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21573529690504073,
      "num_tokens": 116922014.0,
      "step": 67405
    },
    {
      "entropy": 6.025525617599487,
      "epoch": 5.244699474810348,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002603267024299861,
      "loss": 4.9839,
      "mean_token_accuracy": 0.21877204477787018,
      "num_tokens": 116930402.0,
      "step": 67410
    },
    {
      "entropy": 5.962979888916015,
      "epoch": 5.245088504182066,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002602990459530623,
      "loss": 4.9414,
      "mean_token_accuracy": 0.21893858313560485,
      "num_tokens": 116938549.0,
      "step": 67415
    },
    {
      "entropy": 5.999041795730591,
      "epoch": 5.2454775335537835,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002602713896992022,
      "loss": 5.02,
      "mean_token_accuracy": 0.21472976952791215,
      "num_tokens": 116947772.0,
      "step": 67420
    },
    {
      "entropy": 5.898231363296508,
      "epoch": 5.245866562925501,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002602437336688252,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21743259578943253,
      "num_tokens": 116956242.0,
      "step": 67425
    },
    {
      "entropy": 5.9734001636505125,
      "epoch": 5.246255592297218,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002602160778623511,
      "loss": 4.8621,
      "mean_token_accuracy": 0.22080378979444504,
      "num_tokens": 116964448.0,
      "step": 67430
    },
    {
      "entropy": 6.015415191650391,
      "epoch": 5.246644621668936,
      "grad_norm": 1.5625,
      "learning_rate": 0.00026018842228019953,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21415061503648758,
      "num_tokens": 116972794.0,
      "step": 67435
    },
    {
      "entropy": 5.978651285171509,
      "epoch": 5.2470336510406534,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026016076692279003,
      "loss": 5.0457,
      "mean_token_accuracy": 0.21299476325511932,
      "num_tokens": 116981518.0,
      "step": 67440
    },
    {
      "entropy": 5.971816682815552,
      "epoch": 5.247422680412371,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002601331117905423,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21835253089666368,
      "num_tokens": 116989408.0,
      "step": 67445
    },
    {
      "entropy": 5.955803489685058,
      "epoch": 5.247811709784089,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026010545688387595,
      "loss": 4.9983,
      "mean_token_accuracy": 0.2091404154896736,
      "num_tokens": 116998253.0,
      "step": 67450
    },
    {
      "entropy": 6.001044225692749,
      "epoch": 5.2482007391558065,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026007780220321054,
      "loss": 5.0082,
      "mean_token_accuracy": 0.21240423023700714,
      "num_tokens": 117006580.0,
      "step": 67455
    },
    {
      "entropy": 5.954071187973023,
      "epoch": 5.248589768527524,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026005014774896574,
      "loss": 4.9183,
      "mean_token_accuracy": 0.22001922875642776,
      "num_tokens": 117015388.0,
      "step": 67460
    },
    {
      "entropy": 6.05723385810852,
      "epoch": 5.248978797899241,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00026002249352156124,
      "loss": 5.067,
      "mean_token_accuracy": 0.20637932568788528,
      "num_tokens": 117024047.0,
      "step": 67465
    },
    {
      "entropy": 6.0372590065002445,
      "epoch": 5.249367827270959,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002599948395214165,
      "loss": 5.0407,
      "mean_token_accuracy": 0.21453590244054793,
      "num_tokens": 117031858.0,
      "step": 67470
    },
    {
      "entropy": 5.951749324798584,
      "epoch": 5.249756856642676,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025996718574895116,
      "loss": 4.9096,
      "mean_token_accuracy": 0.2232792615890503,
      "num_tokens": 117041336.0,
      "step": 67475
    },
    {
      "entropy": 5.9806647300720215,
      "epoch": 5.250145886014394,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025993953220458487,
      "loss": 4.9775,
      "mean_token_accuracy": 0.20810811072587967,
      "num_tokens": 117049929.0,
      "step": 67480
    },
    {
      "entropy": 5.904570245742798,
      "epoch": 5.250534915386112,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002599118788887372,
      "loss": 4.8441,
      "mean_token_accuracy": 0.23001528084278106,
      "num_tokens": 117058099.0,
      "step": 67485
    },
    {
      "entropy": 6.006984090805053,
      "epoch": 5.2509239447578295,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025988422580182775,
      "loss": 5.0007,
      "mean_token_accuracy": 0.21607559472322463,
      "num_tokens": 117066778.0,
      "step": 67490
    },
    {
      "entropy": 5.933389472961426,
      "epoch": 5.251312974129547,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002598565729442761,
      "loss": 4.9245,
      "mean_token_accuracy": 0.22233477979898453,
      "num_tokens": 117075459.0,
      "step": 67495
    },
    {
      "entropy": 5.943446254730224,
      "epoch": 5.251702003501264,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025982892031650186,
      "loss": 5.0251,
      "mean_token_accuracy": 0.21826497912406922,
      "num_tokens": 117084793.0,
      "step": 67500
    },
    {
      "entropy": 5.965963363647461,
      "epoch": 5.252091032872982,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002598012679189246,
      "loss": 4.9416,
      "mean_token_accuracy": 0.2156515747308731,
      "num_tokens": 117093002.0,
      "step": 67505
    },
    {
      "entropy": 5.977499532699585,
      "epoch": 5.252480062244699,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025977361575196394,
      "loss": 4.9409,
      "mean_token_accuracy": 0.22108418643474578,
      "num_tokens": 117101879.0,
      "step": 67510
    },
    {
      "entropy": 5.982732629776001,
      "epoch": 5.252869091616417,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025974596381603943,
      "loss": 4.9747,
      "mean_token_accuracy": 0.2210507497191429,
      "num_tokens": 117110437.0,
      "step": 67515
    },
    {
      "entropy": 6.003619766235351,
      "epoch": 5.253258120988135,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002597183121115707,
      "loss": 5.0733,
      "mean_token_accuracy": 0.21963540464639664,
      "num_tokens": 117119306.0,
      "step": 67520
    },
    {
      "entropy": 5.9856935977935795,
      "epoch": 5.2536471503598525,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002596906606389771,
      "loss": 4.9099,
      "mean_token_accuracy": 0.2174041301012039,
      "num_tokens": 117128071.0,
      "step": 67525
    },
    {
      "entropy": 5.944335985183716,
      "epoch": 5.254036179731569,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025966300939867843,
      "loss": 4.8246,
      "mean_token_accuracy": 0.2332903802394867,
      "num_tokens": 117136026.0,
      "step": 67530
    },
    {
      "entropy": 5.907544660568237,
      "epoch": 5.254425209103287,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002596353583910942,
      "loss": 4.8987,
      "mean_token_accuracy": 0.2193308711051941,
      "num_tokens": 117144693.0,
      "step": 67535
    },
    {
      "entropy": 5.998728179931641,
      "epoch": 5.254814238475005,
      "grad_norm": 1.5,
      "learning_rate": 0.00025960770761664396,
      "loss": 5.0923,
      "mean_token_accuracy": 0.2034485310316086,
      "num_tokens": 117154076.0,
      "step": 67540
    },
    {
      "entropy": 5.996291875839233,
      "epoch": 5.255203267846722,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025958005707574726,
      "loss": 4.9755,
      "mean_token_accuracy": 0.2131444275379181,
      "num_tokens": 117163581.0,
      "step": 67545
    },
    {
      "entropy": 5.993049383163452,
      "epoch": 5.25559229721844,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002595524067688237,
      "loss": 4.8384,
      "mean_token_accuracy": 0.22515734583139418,
      "num_tokens": 117171734.0,
      "step": 67550
    },
    {
      "entropy": 5.982550048828125,
      "epoch": 5.255981326590158,
      "grad_norm": 1.5,
      "learning_rate": 0.00025952475669629267,
      "loss": 5.0011,
      "mean_token_accuracy": 0.21663331538438796,
      "num_tokens": 117180399.0,
      "step": 67555
    },
    {
      "entropy": 5.970515346527099,
      "epoch": 5.2563703559618755,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025949710685857383,
      "loss": 5.0173,
      "mean_token_accuracy": 0.21684306859970093,
      "num_tokens": 117189135.0,
      "step": 67560
    },
    {
      "entropy": 5.951988458633423,
      "epoch": 5.256759385333592,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025946945725608675,
      "loss": 4.8672,
      "mean_token_accuracy": 0.22158548086881638,
      "num_tokens": 117198717.0,
      "step": 67565
    },
    {
      "entropy": 5.927959680557251,
      "epoch": 5.25714841470531,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002594418078892509,
      "loss": 4.9272,
      "mean_token_accuracy": 0.22490579634904861,
      "num_tokens": 117207887.0,
      "step": 67570
    },
    {
      "entropy": 5.9595386505126955,
      "epoch": 5.257537444077028,
      "grad_norm": 1.390625,
      "learning_rate": 0.000259414158758486,
      "loss": 4.9246,
      "mean_token_accuracy": 0.2206644594669342,
      "num_tokens": 117216843.0,
      "step": 67575
    },
    {
      "entropy": 5.971211290359497,
      "epoch": 5.257926473448745,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025938650986421125,
      "loss": 5.0054,
      "mean_token_accuracy": 0.20718593150377274,
      "num_tokens": 117225565.0,
      "step": 67580
    },
    {
      "entropy": 6.014598608016968,
      "epoch": 5.258315502820463,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002593588612068465,
      "loss": 5.0533,
      "mean_token_accuracy": 0.21116822063922883,
      "num_tokens": 117234084.0,
      "step": 67585
    },
    {
      "entropy": 5.980208253860473,
      "epoch": 5.258704532192181,
      "grad_norm": 1.53125,
      "learning_rate": 0.000259331212786811,
      "loss": 4.9742,
      "mean_token_accuracy": 0.21508805453777313,
      "num_tokens": 117243222.0,
      "step": 67590
    },
    {
      "entropy": 5.924001884460449,
      "epoch": 5.259093561563898,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025930356460452454,
      "loss": 4.8988,
      "mean_token_accuracy": 0.22040313482284546,
      "num_tokens": 117251334.0,
      "step": 67595
    },
    {
      "entropy": 5.997925853729248,
      "epoch": 5.259482590935615,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002592759166604064,
      "loss": 5.0242,
      "mean_token_accuracy": 0.20812930762767792,
      "num_tokens": 117260393.0,
      "step": 67600
    },
    {
      "entropy": 5.951492214202881,
      "epoch": 5.259871620307333,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025924826895487623,
      "loss": 4.9099,
      "mean_token_accuracy": 0.21920438557863237,
      "num_tokens": 117269271.0,
      "step": 67605
    },
    {
      "entropy": 6.015502262115478,
      "epoch": 5.260260649679051,
      "grad_norm": 1.5,
      "learning_rate": 0.00025922062148835357,
      "loss": 5.0163,
      "mean_token_accuracy": 0.21689097136259078,
      "num_tokens": 117277888.0,
      "step": 67610
    },
    {
      "entropy": 5.973323392868042,
      "epoch": 5.260649679050768,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002591929742612578,
      "loss": 4.9748,
      "mean_token_accuracy": 0.2149810701608658,
      "num_tokens": 117286251.0,
      "step": 67615
    },
    {
      "entropy": 5.963575172424316,
      "epoch": 5.261038708422486,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002591653272740085,
      "loss": 4.9535,
      "mean_token_accuracy": 0.2143026664853096,
      "num_tokens": 117295280.0,
      "step": 67620
    },
    {
      "entropy": 6.0192780017852785,
      "epoch": 5.261427737794204,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025913768052702514,
      "loss": 4.9794,
      "mean_token_accuracy": 0.2192216098308563,
      "num_tokens": 117304008.0,
      "step": 67625
    },
    {
      "entropy": 5.979583978652954,
      "epoch": 5.261816767165921,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002591100340207272,
      "loss": 4.9639,
      "mean_token_accuracy": 0.21068847924470901,
      "num_tokens": 117313052.0,
      "step": 67630
    },
    {
      "entropy": 5.978602600097656,
      "epoch": 5.262205796537638,
      "grad_norm": 1.5,
      "learning_rate": 0.0002590823877555343,
      "loss": 4.9695,
      "mean_token_accuracy": 0.2220069244503975,
      "num_tokens": 117321807.0,
      "step": 67635
    },
    {
      "entropy": 5.9941566467285154,
      "epoch": 5.262594825909356,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002590547417318657,
      "loss": 4.9741,
      "mean_token_accuracy": 0.2164134278893471,
      "num_tokens": 117330646.0,
      "step": 67640
    },
    {
      "entropy": 5.9821264266967775,
      "epoch": 5.262983855281074,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002590270959501411,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2209234282374382,
      "num_tokens": 117339585.0,
      "step": 67645
    },
    {
      "entropy": 5.936215734481811,
      "epoch": 5.263372884652791,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025899945041077995,
      "loss": 4.9165,
      "mean_token_accuracy": 0.2252254456281662,
      "num_tokens": 117348036.0,
      "step": 67650
    },
    {
      "entropy": 5.981120204925537,
      "epoch": 5.263761914024509,
      "grad_norm": 1.5,
      "learning_rate": 0.00025897180511420164,
      "loss": 4.9656,
      "mean_token_accuracy": 0.21870957612991332,
      "num_tokens": 117356397.0,
      "step": 67655
    },
    {
      "entropy": 5.909137630462647,
      "epoch": 5.264150943396227,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025894416006082564,
      "loss": 4.9706,
      "mean_token_accuracy": 0.21453395932912828,
      "num_tokens": 117364832.0,
      "step": 67660
    },
    {
      "entropy": 6.024837303161621,
      "epoch": 5.264539972767944,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025891651525107145,
      "loss": 5.0307,
      "mean_token_accuracy": 0.21221089661121367,
      "num_tokens": 117373987.0,
      "step": 67665
    },
    {
      "entropy": 5.924259185791016,
      "epoch": 5.264929002139661,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025888887068535853,
      "loss": 4.9363,
      "mean_token_accuracy": 0.2253112718462944,
      "num_tokens": 117382279.0,
      "step": 67670
    },
    {
      "entropy": 5.88146653175354,
      "epoch": 5.265318031511379,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025886122636410643,
      "loss": 4.8252,
      "mean_token_accuracy": 0.22863128185272216,
      "num_tokens": 117390682.0,
      "step": 67675
    },
    {
      "entropy": 5.972973823547363,
      "epoch": 5.265707060883097,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025883358228773445,
      "loss": 4.9737,
      "mean_token_accuracy": 0.21298437863588332,
      "num_tokens": 117399604.0,
      "step": 67680
    },
    {
      "entropy": 5.974107265472412,
      "epoch": 5.266096090254814,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002588059384566622,
      "loss": 4.9382,
      "mean_token_accuracy": 0.21664115637540818,
      "num_tokens": 117408483.0,
      "step": 67685
    },
    {
      "entropy": 5.982531023025513,
      "epoch": 5.266485119626532,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025877829487130894,
      "loss": 5.0071,
      "mean_token_accuracy": 0.22118644118309022,
      "num_tokens": 117417838.0,
      "step": 67690
    },
    {
      "entropy": 5.911885738372803,
      "epoch": 5.26687414899825,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002587506515320943,
      "loss": 4.9648,
      "mean_token_accuracy": 0.21247802674770355,
      "num_tokens": 117426486.0,
      "step": 67695
    },
    {
      "entropy": 6.034180974960327,
      "epoch": 5.267263178369967,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002587230084394377,
      "loss": 4.982,
      "mean_token_accuracy": 0.22045515924692155,
      "num_tokens": 117434839.0,
      "step": 67700
    },
    {
      "entropy": 5.940950727462768,
      "epoch": 5.267652207741684,
      "grad_norm": 1.5,
      "learning_rate": 0.0002586953655937585,
      "loss": 4.8914,
      "mean_token_accuracy": 0.21908270418643952,
      "num_tokens": 117443184.0,
      "step": 67705
    },
    {
      "entropy": 5.961883974075318,
      "epoch": 5.268041237113402,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002586677229954761,
      "loss": 4.8605,
      "mean_token_accuracy": 0.22108512967824936,
      "num_tokens": 117451183.0,
      "step": 67710
    },
    {
      "entropy": 5.968191194534302,
      "epoch": 5.26843026648512,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002586400806450102,
      "loss": 4.9178,
      "mean_token_accuracy": 0.22062088549137115,
      "num_tokens": 117459804.0,
      "step": 67715
    },
    {
      "entropy": 5.928032255172729,
      "epoch": 5.268819295856837,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025861243854277986,
      "loss": 4.8907,
      "mean_token_accuracy": 0.22233970165252687,
      "num_tokens": 117469365.0,
      "step": 67720
    },
    {
      "entropy": 5.946619653701783,
      "epoch": 5.269208325228555,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002585847966892048,
      "loss": 4.9211,
      "mean_token_accuracy": 0.2221261218190193,
      "num_tokens": 117477590.0,
      "step": 67725
    },
    {
      "entropy": 6.030374670028687,
      "epoch": 5.269597354600272,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002585571550847042,
      "loss": 4.994,
      "mean_token_accuracy": 0.21638210415840148,
      "num_tokens": 117487559.0,
      "step": 67730
    },
    {
      "entropy": 5.953281879425049,
      "epoch": 5.26998638397199,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002585295137296978,
      "loss": 4.942,
      "mean_token_accuracy": 0.21133701950311662,
      "num_tokens": 117496242.0,
      "step": 67735
    },
    {
      "entropy": 6.015468406677246,
      "epoch": 5.270375413343707,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002585018726246047,
      "loss": 4.9763,
      "mean_token_accuracy": 0.21944270133972169,
      "num_tokens": 117504281.0,
      "step": 67740
    },
    {
      "entropy": 6.0001167297363285,
      "epoch": 5.270764442715425,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002584742317698444,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21966755539178848,
      "num_tokens": 117512969.0,
      "step": 67745
    },
    {
      "entropy": 6.020474290847778,
      "epoch": 5.271153472087143,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025844659116583635,
      "loss": 5.0873,
      "mean_token_accuracy": 0.20550852715969087,
      "num_tokens": 117522189.0,
      "step": 67750
    },
    {
      "entropy": 5.962597846984863,
      "epoch": 5.27154250145886,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002584189508129999,
      "loss": 4.9809,
      "mean_token_accuracy": 0.21557166278362275,
      "num_tokens": 117530994.0,
      "step": 67755
    },
    {
      "entropy": 5.887809085845947,
      "epoch": 5.271931530830578,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002583913107117544,
      "loss": 4.7914,
      "mean_token_accuracy": 0.22135401368141175,
      "num_tokens": 117539967.0,
      "step": 67760
    },
    {
      "entropy": 5.936080694198608,
      "epoch": 5.272320560202295,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002583636708625195,
      "loss": 4.9337,
      "mean_token_accuracy": 0.21798508018255233,
      "num_tokens": 117549220.0,
      "step": 67765
    },
    {
      "entropy": 5.9961730480194095,
      "epoch": 5.272709589574013,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002583360312657143,
      "loss": 4.9364,
      "mean_token_accuracy": 0.22163780629634858,
      "num_tokens": 117557039.0,
      "step": 67770
    },
    {
      "entropy": 5.976439046859741,
      "epoch": 5.27309861894573,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025830839192175835,
      "loss": 4.889,
      "mean_token_accuracy": 0.228982974588871,
      "num_tokens": 117564828.0,
      "step": 67775
    },
    {
      "entropy": 5.9513756275177006,
      "epoch": 5.273487648317448,
      "grad_norm": 1.546875,
      "learning_rate": 0.000258280752831071,
      "loss": 4.9524,
      "mean_token_accuracy": 0.2213790625333786,
      "num_tokens": 117573860.0,
      "step": 67780
    },
    {
      "entropy": 5.9165136337280275,
      "epoch": 5.273876677689166,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025825311399407157,
      "loss": 4.9534,
      "mean_token_accuracy": 0.2195112094283104,
      "num_tokens": 117582536.0,
      "step": 67785
    },
    {
      "entropy": 5.99713888168335,
      "epoch": 5.274265707060883,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002582254754111795,
      "loss": 4.9689,
      "mean_token_accuracy": 0.21397227942943572,
      "num_tokens": 117590986.0,
      "step": 67790
    },
    {
      "entropy": 5.961357498168946,
      "epoch": 5.274654736432601,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002581978370828142,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21776940077543258,
      "num_tokens": 117599637.0,
      "step": 67795
    },
    {
      "entropy": 5.983619546890258,
      "epoch": 5.275043765804318,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002581701990093949,
      "loss": 4.9712,
      "mean_token_accuracy": 0.21711639165878296,
      "num_tokens": 117608367.0,
      "step": 67800
    },
    {
      "entropy": 6.005378198623657,
      "epoch": 5.275432795176036,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002581425611913411,
      "loss": 5.0244,
      "mean_token_accuracy": 0.2101608246564865,
      "num_tokens": 117617787.0,
      "step": 67805
    },
    {
      "entropy": 6.033416843414306,
      "epoch": 5.275821824547753,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002581149236290721,
      "loss": 5.0229,
      "mean_token_accuracy": 0.21322278380393983,
      "num_tokens": 117625925.0,
      "step": 67810
    },
    {
      "entropy": 5.912703609466552,
      "epoch": 5.276210853919471,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002580872863230072,
      "loss": 4.9021,
      "mean_token_accuracy": 0.21993038505315782,
      "num_tokens": 117634970.0,
      "step": 67815
    },
    {
      "entropy": 5.904629468917847,
      "epoch": 5.276599883291189,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025805964927356584,
      "loss": 4.8427,
      "mean_token_accuracy": 0.22811115980148317,
      "num_tokens": 117644177.0,
      "step": 67820
    },
    {
      "entropy": 5.9897730350494385,
      "epoch": 5.276988912662906,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00025803201248116725,
      "loss": 4.9331,
      "mean_token_accuracy": 0.2187851458787918,
      "num_tokens": 117651901.0,
      "step": 67825
    },
    {
      "entropy": 6.023100519180298,
      "epoch": 5.277377942034624,
      "grad_norm": 1.5,
      "learning_rate": 0.00025800437594623095,
      "loss": 5.0683,
      "mean_token_accuracy": 0.2137395054101944,
      "num_tokens": 117661464.0,
      "step": 67830
    },
    {
      "entropy": 6.041058111190796,
      "epoch": 5.277766971406341,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002579767396691762,
      "loss": 5.0698,
      "mean_token_accuracy": 0.20243177711963653,
      "num_tokens": 117670260.0,
      "step": 67835
    },
    {
      "entropy": 5.9597327709198,
      "epoch": 5.278156000778059,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025794910365042233,
      "loss": 4.9781,
      "mean_token_accuracy": 0.21887548863887787,
      "num_tokens": 117679432.0,
      "step": 67840
    },
    {
      "entropy": 5.975893640518189,
      "epoch": 5.278545030149776,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025792146789038875,
      "loss": 4.9305,
      "mean_token_accuracy": 0.21914642602205275,
      "num_tokens": 117687668.0,
      "step": 67845
    },
    {
      "entropy": 5.979355573654175,
      "epoch": 5.278934059521494,
      "grad_norm": 1.5,
      "learning_rate": 0.0002578938323894946,
      "loss": 4.9431,
      "mean_token_accuracy": 0.2167784407734871,
      "num_tokens": 117696950.0,
      "step": 67850
    },
    {
      "entropy": 5.969687986373901,
      "epoch": 5.279323088893212,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025786619714815933,
      "loss": 4.958,
      "mean_token_accuracy": 0.2214703619480133,
      "num_tokens": 117706003.0,
      "step": 67855
    },
    {
      "entropy": 5.878231811523437,
      "epoch": 5.279712118264929,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002578385621668023,
      "loss": 4.9065,
      "mean_token_accuracy": 0.22559356838464736,
      "num_tokens": 117715142.0,
      "step": 67860
    },
    {
      "entropy": 5.925505590438843,
      "epoch": 5.280101147636646,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002578109274458427,
      "loss": 5.0236,
      "mean_token_accuracy": 0.20252797305583953,
      "num_tokens": 117724500.0,
      "step": 67865
    },
    {
      "entropy": 5.98210186958313,
      "epoch": 5.280490177008364,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025778329298569997,
      "loss": 4.9728,
      "mean_token_accuracy": 0.21505208164453507,
      "num_tokens": 117733056.0,
      "step": 67870
    },
    {
      "entropy": 6.0959666728973385,
      "epoch": 5.280879206380082,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002577556587867934,
      "loss": 5.114,
      "mean_token_accuracy": 0.2012871116399765,
      "num_tokens": 117741854.0,
      "step": 67875
    },
    {
      "entropy": 5.953777217864991,
      "epoch": 5.281268235751799,
      "grad_norm": 1.5625,
      "learning_rate": 0.00025772802484954217,
      "loss": 4.9346,
      "mean_token_accuracy": 0.22369454950094222,
      "num_tokens": 117751104.0,
      "step": 67880
    },
    {
      "entropy": 5.927759790420533,
      "epoch": 5.281657265123517,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025770039117436566,
      "loss": 4.8596,
      "mean_token_accuracy": 0.21584402471780778,
      "num_tokens": 117758529.0,
      "step": 67885
    },
    {
      "entropy": 6.077860927581787,
      "epoch": 5.282046294495235,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002576727577616832,
      "loss": 5.073,
      "mean_token_accuracy": 0.21445686370134354,
      "num_tokens": 117768011.0,
      "step": 67890
    },
    {
      "entropy": 5.977238130569458,
      "epoch": 5.282435323866952,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025764512461191413,
      "loss": 4.9224,
      "mean_token_accuracy": 0.21748718172311782,
      "num_tokens": 117776461.0,
      "step": 67895
    },
    {
      "entropy": 6.0271893501281735,
      "epoch": 5.282824353238669,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002576174917254776,
      "loss": 5.0406,
      "mean_token_accuracy": 0.21793659180402755,
      "num_tokens": 117785634.0,
      "step": 67900
    },
    {
      "entropy": 5.940367364883423,
      "epoch": 5.283213382610387,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025758985910279304,
      "loss": 4.9148,
      "mean_token_accuracy": 0.22272120118141175,
      "num_tokens": 117794287.0,
      "step": 67905
    },
    {
      "entropy": 5.9818182468414305,
      "epoch": 5.283602411982105,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002575622267442796,
      "loss": 4.9645,
      "mean_token_accuracy": 0.22372244745492936,
      "num_tokens": 117802664.0,
      "step": 67910
    },
    {
      "entropy": 5.955509662628174,
      "epoch": 5.283991441353822,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002575345946503566,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2203936383128166,
      "num_tokens": 117811620.0,
      "step": 67915
    },
    {
      "entropy": 5.955630826950073,
      "epoch": 5.28438047072554,
      "grad_norm": 1.53125,
      "learning_rate": 0.00025750696282144335,
      "loss": 4.9616,
      "mean_token_accuracy": 0.2234398052096367,
      "num_tokens": 117819925.0,
      "step": 67920
    },
    {
      "entropy": 6.028868961334228,
      "epoch": 5.284769500097258,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002574793312579592,
      "loss": 5.0959,
      "mean_token_accuracy": 0.20464053750038147,
      "num_tokens": 117829734.0,
      "step": 67925
    },
    {
      "entropy": 5.969915533065796,
      "epoch": 5.2851585294689745,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025745169996032317,
      "loss": 4.8887,
      "mean_token_accuracy": 0.22824964821338653,
      "num_tokens": 117837751.0,
      "step": 67930
    },
    {
      "entropy": 5.988775396347046,
      "epoch": 5.285547558840692,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002574240689289547,
      "loss": 4.9376,
      "mean_token_accuracy": 0.21857507079839705,
      "num_tokens": 117846661.0,
      "step": 67935
    },
    {
      "entropy": 5.8921019554138185,
      "epoch": 5.28593658821241,
      "grad_norm": 1.5,
      "learning_rate": 0.00025739643816427305,
      "loss": 4.9003,
      "mean_token_accuracy": 0.22008283138275148,
      "num_tokens": 117855802.0,
      "step": 67940
    },
    {
      "entropy": 5.98436222076416,
      "epoch": 5.286325617584128,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002573688076666974,
      "loss": 4.9289,
      "mean_token_accuracy": 0.2196797251701355,
      "num_tokens": 117864150.0,
      "step": 67945
    },
    {
      "entropy": 5.946332740783691,
      "epoch": 5.286714646955845,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025734117743664704,
      "loss": 4.969,
      "mean_token_accuracy": 0.21360151022672652,
      "num_tokens": 117873048.0,
      "step": 67950
    },
    {
      "entropy": 5.972882223129273,
      "epoch": 5.287103676327563,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002573135474745412,
      "loss": 4.9244,
      "mean_token_accuracy": 0.218148772418499,
      "num_tokens": 117881607.0,
      "step": 67955
    },
    {
      "entropy": 5.906508207321167,
      "epoch": 5.287492705699281,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025728591778079906,
      "loss": 4.8494,
      "mean_token_accuracy": 0.23455032259225844,
      "num_tokens": 117890627.0,
      "step": 67960
    },
    {
      "entropy": 5.9497801780700685,
      "epoch": 5.2878817350709975,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025725828835584,
      "loss": 4.9612,
      "mean_token_accuracy": 0.22152750939130783,
      "num_tokens": 117900213.0,
      "step": 67965
    },
    {
      "entropy": 6.033314752578735,
      "epoch": 5.288270764442715,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025723065920008314,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21802776604890822,
      "num_tokens": 117909215.0,
      "step": 67970
    },
    {
      "entropy": 5.9280993938446045,
      "epoch": 5.288659793814433,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002572030303139477,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21690268963575363,
      "num_tokens": 117918372.0,
      "step": 67975
    },
    {
      "entropy": 5.977475547790528,
      "epoch": 5.2890488231861505,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025717540169785304,
      "loss": 4.9767,
      "mean_token_accuracy": 0.21845737844705582,
      "num_tokens": 117928135.0,
      "step": 67980
    },
    {
      "entropy": 5.955758666992187,
      "epoch": 5.289437852557868,
      "grad_norm": 1.5,
      "learning_rate": 0.0002571477733522183,
      "loss": 4.9561,
      "mean_token_accuracy": 0.2193334624171257,
      "num_tokens": 117937047.0,
      "step": 67985
    },
    {
      "entropy": 5.980472660064697,
      "epoch": 5.289826881929586,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025712014527746264,
      "loss": 4.9589,
      "mean_token_accuracy": 0.22593217939138413,
      "num_tokens": 117945387.0,
      "step": 67990
    },
    {
      "entropy": 5.935102701187134,
      "epoch": 5.290215911301304,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002570925174740053,
      "loss": 4.9159,
      "mean_token_accuracy": 0.2308706074953079,
      "num_tokens": 117953845.0,
      "step": 67995
    },
    {
      "entropy": 5.95358452796936,
      "epoch": 5.2906049406730205,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002570648899422655,
      "loss": 4.9419,
      "mean_token_accuracy": 0.2197519749403,
      "num_tokens": 117962009.0,
      "step": 68000
    },
    {
      "entropy": 5.917940092086792,
      "epoch": 5.290993970044738,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002570372626826625,
      "loss": 5.0294,
      "mean_token_accuracy": 0.216458360850811,
      "num_tokens": 117971314.0,
      "step": 68005
    },
    {
      "entropy": 5.9393883228302,
      "epoch": 5.291382999416456,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002570096356956155,
      "loss": 4.9566,
      "mean_token_accuracy": 0.2139214336872101,
      "num_tokens": 117979074.0,
      "step": 68010
    },
    {
      "entropy": 5.950162029266357,
      "epoch": 5.2917720287881735,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00025698200898154354,
      "loss": 4.8842,
      "mean_token_accuracy": 0.225216306746006,
      "num_tokens": 117987806.0,
      "step": 68015
    },
    {
      "entropy": 5.956601572036743,
      "epoch": 5.292161058159891,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00025695438254086595,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21760119795799254,
      "num_tokens": 117997392.0,
      "step": 68020
    },
    {
      "entropy": 5.928466129302978,
      "epoch": 5.292550087531609,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002569267563740018,
      "loss": 4.907,
      "mean_token_accuracy": 0.23111872673034667,
      "num_tokens": 118006117.0,
      "step": 68025
    },
    {
      "entropy": 5.98553352355957,
      "epoch": 5.292939116903327,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002568991304813704,
      "loss": 4.9663,
      "mean_token_accuracy": 0.2185293436050415,
      "num_tokens": 118014550.0,
      "step": 68030
    },
    {
      "entropy": 5.987444067001343,
      "epoch": 5.2933281462750434,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025687150486339094,
      "loss": 4.9628,
      "mean_token_accuracy": 0.22105581909418107,
      "num_tokens": 118023389.0,
      "step": 68035
    },
    {
      "entropy": 6.007596063613891,
      "epoch": 5.293717175646761,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025684387952048247,
      "loss": 4.972,
      "mean_token_accuracy": 0.2186387985944748,
      "num_tokens": 118031949.0,
      "step": 68040
    },
    {
      "entropy": 5.985912704467774,
      "epoch": 5.294106205018479,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025681625445306435,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21374718844890594,
      "num_tokens": 118040788.0,
      "step": 68045
    },
    {
      "entropy": 5.890295886993409,
      "epoch": 5.2944952343901965,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002567886296615555,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2262079581618309,
      "num_tokens": 118049945.0,
      "step": 68050
    },
    {
      "entropy": 6.016670751571655,
      "epoch": 5.294884263761914,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025676100514637523,
      "loss": 5.008,
      "mean_token_accuracy": 0.2097589373588562,
      "num_tokens": 118059205.0,
      "step": 68055
    },
    {
      "entropy": 5.9184588432312015,
      "epoch": 5.295273293133632,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025673338090794276,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2226523905992508,
      "num_tokens": 118067444.0,
      "step": 68060
    },
    {
      "entropy": 5.921497631072998,
      "epoch": 5.295662322505349,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002567057569466771,
      "loss": 4.8873,
      "mean_token_accuracy": 0.22092754393815994,
      "num_tokens": 118075269.0,
      "step": 68065
    },
    {
      "entropy": 5.992755174636841,
      "epoch": 5.296051351877066,
      "grad_norm": 1.578125,
      "learning_rate": 0.00025667813326299746,
      "loss": 5.0033,
      "mean_token_accuracy": 0.21625787168741226,
      "num_tokens": 118083424.0,
      "step": 68070
    },
    {
      "entropy": 5.982354688644409,
      "epoch": 5.296440381248784,
      "grad_norm": 1.546875,
      "learning_rate": 0.00025665050985732295,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21906355321407317,
      "num_tokens": 118091505.0,
      "step": 68075
    },
    {
      "entropy": 5.979870986938477,
      "epoch": 5.296829410620502,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025662288673007286,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22124415040016174,
      "num_tokens": 118100111.0,
      "step": 68080
    },
    {
      "entropy": 5.960256814956665,
      "epoch": 5.2972184399922195,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025659526388166617,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21691467016935348,
      "num_tokens": 118109174.0,
      "step": 68085
    },
    {
      "entropy": 5.955220937728882,
      "epoch": 5.297607469363937,
      "grad_norm": 1.578125,
      "learning_rate": 0.00025656764131252194,
      "loss": 4.9253,
      "mean_token_accuracy": 0.21405142694711685,
      "num_tokens": 118117611.0,
      "step": 68090
    },
    {
      "entropy": 6.006940126419067,
      "epoch": 5.297996498735655,
      "grad_norm": 1.625,
      "learning_rate": 0.0002565400190230595,
      "loss": 4.98,
      "mean_token_accuracy": 0.21333175748586655,
      "num_tokens": 118126585.0,
      "step": 68095
    },
    {
      "entropy": 5.958860731124878,
      "epoch": 5.298385528107372,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002565123970136978,
      "loss": 4.8851,
      "mean_token_accuracy": 0.21939504593610765,
      "num_tokens": 118135430.0,
      "step": 68100
    },
    {
      "entropy": 5.982917213439942,
      "epoch": 5.298774557479089,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002564847752848562,
      "loss": 4.9586,
      "mean_token_accuracy": 0.2155200347304344,
      "num_tokens": 118144332.0,
      "step": 68105
    },
    {
      "entropy": 5.982881307601929,
      "epoch": 5.299163586850807,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025645715383695357,
      "loss": 4.9705,
      "mean_token_accuracy": 0.21116551607847214,
      "num_tokens": 118153588.0,
      "step": 68110
    },
    {
      "entropy": 5.98503942489624,
      "epoch": 5.299552616222525,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002564295326704092,
      "loss": 4.9316,
      "mean_token_accuracy": 0.21525246649980545,
      "num_tokens": 118161286.0,
      "step": 68115
    },
    {
      "entropy": 5.94559006690979,
      "epoch": 5.2999416455942425,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002564019117856421,
      "loss": 4.9233,
      "mean_token_accuracy": 0.22242445647716522,
      "num_tokens": 118169392.0,
      "step": 68120
    },
    {
      "entropy": 5.9724445819854735,
      "epoch": 5.30033067496596,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00025637429118307133,
      "loss": 4.9784,
      "mean_token_accuracy": 0.21505019217729568,
      "num_tokens": 118178178.0,
      "step": 68125
    },
    {
      "entropy": 5.9332911491394045,
      "epoch": 5.300719704337677,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002563466708631161,
      "loss": 4.9254,
      "mean_token_accuracy": 0.22648773789405824,
      "num_tokens": 118187492.0,
      "step": 68130
    },
    {
      "entropy": 5.850452327728272,
      "epoch": 5.301108733709395,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002563190508261954,
      "loss": 4.7431,
      "mean_token_accuracy": 0.23320295065641403,
      "num_tokens": 118197123.0,
      "step": 68135
    },
    {
      "entropy": 5.9589324474334715,
      "epoch": 5.301497763081112,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025629143107272837,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20879955291748048,
      "num_tokens": 118206600.0,
      "step": 68140
    },
    {
      "entropy": 5.9717308521270756,
      "epoch": 5.30188679245283,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002562638116031341,
      "loss": 5.0459,
      "mean_token_accuracy": 0.22101704478263856,
      "num_tokens": 118215250.0,
      "step": 68145
    },
    {
      "entropy": 5.975613689422607,
      "epoch": 5.302275821824548,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002562361924178317,
      "loss": 4.9306,
      "mean_token_accuracy": 0.21861519962549208,
      "num_tokens": 118223739.0,
      "step": 68150
    },
    {
      "entropy": 5.971756792068481,
      "epoch": 5.3026648511962655,
      "grad_norm": 1.609375,
      "learning_rate": 0.00025620857351724017,
      "loss": 4.9969,
      "mean_token_accuracy": 0.2177880585193634,
      "num_tokens": 118232446.0,
      "step": 68155
    },
    {
      "entropy": 5.951284408569336,
      "epoch": 5.303053880567983,
      "grad_norm": 1.578125,
      "learning_rate": 0.00025618095490177864,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21249286234378814,
      "num_tokens": 118241057.0,
      "step": 68160
    },
    {
      "entropy": 6.0204792499542235,
      "epoch": 5.303442909939701,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025615333657186615,
      "loss": 4.9703,
      "mean_token_accuracy": 0.21929802149534225,
      "num_tokens": 118249173.0,
      "step": 68165
    },
    {
      "entropy": 5.925481271743775,
      "epoch": 5.303831939311418,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002561257185279218,
      "loss": 4.917,
      "mean_token_accuracy": 0.22514718025922775,
      "num_tokens": 118257976.0,
      "step": 68170
    },
    {
      "entropy": 5.991113662719727,
      "epoch": 5.304220968683135,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025609810077036467,
      "loss": 4.9904,
      "mean_token_accuracy": 0.21183126419782639,
      "num_tokens": 118266836.0,
      "step": 68175
    },
    {
      "entropy": 5.938324594497681,
      "epoch": 5.304609998054853,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002560704832996137,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2308041587471962,
      "num_tokens": 118275442.0,
      "step": 68180
    },
    {
      "entropy": 6.046685028076172,
      "epoch": 5.304999027426571,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002560428661160881,
      "loss": 5.0754,
      "mean_token_accuracy": 0.20719914883375168,
      "num_tokens": 118284106.0,
      "step": 68185
    },
    {
      "entropy": 5.921615505218506,
      "epoch": 5.3053880567982885,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00025601524922020676,
      "loss": 4.843,
      "mean_token_accuracy": 0.2375665009021759,
      "num_tokens": 118292541.0,
      "step": 68190
    },
    {
      "entropy": 5.964902257919311,
      "epoch": 5.305777086170006,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002559876326123889,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21858649998903273,
      "num_tokens": 118301168.0,
      "step": 68195
    },
    {
      "entropy": 5.951911878585816,
      "epoch": 5.306166115541723,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002559600162930534,
      "loss": 4.9684,
      "mean_token_accuracy": 0.2081761747598648,
      "num_tokens": 118309652.0,
      "step": 68200
    },
    {
      "entropy": 5.943492221832275,
      "epoch": 5.306555144913441,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002559324002626193,
      "loss": 4.8758,
      "mean_token_accuracy": 0.22620240598917007,
      "num_tokens": 118318985.0,
      "step": 68205
    },
    {
      "entropy": 5.958231639862061,
      "epoch": 5.306944174285158,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002559047845215057,
      "loss": 4.9272,
      "mean_token_accuracy": 0.22273888438940048,
      "num_tokens": 118327637.0,
      "step": 68210
    },
    {
      "entropy": 6.005891990661621,
      "epoch": 5.307333203656876,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002558771690701316,
      "loss": 4.9555,
      "mean_token_accuracy": 0.21176522672176362,
      "num_tokens": 118336584.0,
      "step": 68215
    },
    {
      "entropy": 5.993188714981079,
      "epoch": 5.307722233028594,
      "grad_norm": 1.515625,
      "learning_rate": 0.000255849553908916,
      "loss": 5.0426,
      "mean_token_accuracy": 0.22552598416805267,
      "num_tokens": 118345211.0,
      "step": 68220
    },
    {
      "entropy": 5.980638742446899,
      "epoch": 5.3081112624003115,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025582193903827795,
      "loss": 4.9514,
      "mean_token_accuracy": 0.2156098872423172,
      "num_tokens": 118353894.0,
      "step": 68225
    },
    {
      "entropy": 5.970308065414429,
      "epoch": 5.308500291772029,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002557943244586365,
      "loss": 4.9545,
      "mean_token_accuracy": 0.2234280914068222,
      "num_tokens": 118362683.0,
      "step": 68230
    },
    {
      "entropy": 5.937115097045899,
      "epoch": 5.308889321143746,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025576671017041053,
      "loss": 5.0311,
      "mean_token_accuracy": 0.22063904255628586,
      "num_tokens": 118371860.0,
      "step": 68235
    },
    {
      "entropy": 5.953917074203491,
      "epoch": 5.309278350515464,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002557390961740192,
      "loss": 4.9982,
      "mean_token_accuracy": 0.21407688558101653,
      "num_tokens": 118380404.0,
      "step": 68240
    },
    {
      "entropy": 6.024641227722168,
      "epoch": 5.309667379887181,
      "grad_norm": 1.5,
      "learning_rate": 0.00025571148246988137,
      "loss": 5.0293,
      "mean_token_accuracy": 0.21674028486013414,
      "num_tokens": 118388467.0,
      "step": 68245
    },
    {
      "entropy": 6.040879106521606,
      "epoch": 5.310056409258899,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025568386905841614,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2168242484331131,
      "num_tokens": 118396975.0,
      "step": 68250
    },
    {
      "entropy": 5.991826057434082,
      "epoch": 5.310445438630617,
      "grad_norm": 1.5,
      "learning_rate": 0.0002556562559400424,
      "loss": 4.9542,
      "mean_token_accuracy": 0.2128160759806633,
      "num_tokens": 118405339.0,
      "step": 68255
    },
    {
      "entropy": 5.961790084838867,
      "epoch": 5.3108344680023345,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002556286431151792,
      "loss": 4.9437,
      "mean_token_accuracy": 0.22477167546749116,
      "num_tokens": 118413328.0,
      "step": 68260
    },
    {
      "entropy": 5.993412780761719,
      "epoch": 5.311223497374051,
      "grad_norm": 1.625,
      "learning_rate": 0.0002556010305842456,
      "loss": 5.0271,
      "mean_token_accuracy": 0.2083316907286644,
      "num_tokens": 118421337.0,
      "step": 68265
    },
    {
      "entropy": 6.012970876693726,
      "epoch": 5.311612526745769,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002555734183476603,
      "loss": 5.0028,
      "mean_token_accuracy": 0.21002837568521499,
      "num_tokens": 118429692.0,
      "step": 68270
    },
    {
      "entropy": 5.974552822113037,
      "epoch": 5.312001556117487,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002555458064058427,
      "loss": 4.9687,
      "mean_token_accuracy": 0.21571324467658998,
      "num_tokens": 118438904.0,
      "step": 68275
    },
    {
      "entropy": 6.106216669082642,
      "epoch": 5.312390585489204,
      "grad_norm": 1.5,
      "learning_rate": 0.00025551819475921135,
      "loss": 5.0399,
      "mean_token_accuracy": 0.2099968209862709,
      "num_tokens": 118447697.0,
      "step": 68280
    },
    {
      "entropy": 6.034918403625488,
      "epoch": 5.312779614860922,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002554905834081854,
      "loss": 4.9794,
      "mean_token_accuracy": 0.22075135856866837,
      "num_tokens": 118456163.0,
      "step": 68285
    },
    {
      "entropy": 5.998078775405884,
      "epoch": 5.31316864423264,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002554629723531838,
      "loss": 4.9722,
      "mean_token_accuracy": 0.21584932506084442,
      "num_tokens": 118465152.0,
      "step": 68290
    },
    {
      "entropy": 5.969123029708863,
      "epoch": 5.3135576736043575,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002554353615946255,
      "loss": 4.9829,
      "mean_token_accuracy": 0.22040852159261703,
      "num_tokens": 118474580.0,
      "step": 68295
    },
    {
      "entropy": 6.009294319152832,
      "epoch": 5.313946702976074,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025540775113292943,
      "loss": 4.9668,
      "mean_token_accuracy": 0.21590973138809205,
      "num_tokens": 118484216.0,
      "step": 68300
    },
    {
      "entropy": 5.981473350524903,
      "epoch": 5.314335732347792,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002553801409685146,
      "loss": 4.965,
      "mean_token_accuracy": 0.21797425746917726,
      "num_tokens": 118493889.0,
      "step": 68305
    },
    {
      "entropy": 6.050318050384521,
      "epoch": 5.31472476171951,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00025535253110179994,
      "loss": 5.0414,
      "mean_token_accuracy": 0.2085738003253937,
      "num_tokens": 118502134.0,
      "step": 68310
    },
    {
      "entropy": 5.916487503051758,
      "epoch": 5.315113791091227,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002553249215332043,
      "loss": 4.8786,
      "mean_token_accuracy": 0.22368923872709273,
      "num_tokens": 118510478.0,
      "step": 68315
    },
    {
      "entropy": 5.959349298477173,
      "epoch": 5.315502820462945,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00025529731226314663,
      "loss": 5.0048,
      "mean_token_accuracy": 0.22175308018922807,
      "num_tokens": 118518492.0,
      "step": 68320
    },
    {
      "entropy": 5.987621402740478,
      "epoch": 5.315891849834663,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002552697032920459,
      "loss": 4.9791,
      "mean_token_accuracy": 0.2137409344315529,
      "num_tokens": 118527033.0,
      "step": 68325
    },
    {
      "entropy": 6.014165019989013,
      "epoch": 5.3162808792063805,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025524209462032105,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21551700532436371,
      "num_tokens": 118534850.0,
      "step": 68330
    },
    {
      "entropy": 5.948311710357666,
      "epoch": 5.316669908578097,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002552144862483909,
      "loss": 4.9215,
      "mean_token_accuracy": 0.21659501343965532,
      "num_tokens": 118543470.0,
      "step": 68335
    },
    {
      "entropy": 5.898744010925293,
      "epoch": 5.317058937949815,
      "grad_norm": 1.375,
      "learning_rate": 0.0002551868781766745,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21884360611438752,
      "num_tokens": 118553052.0,
      "step": 68340
    },
    {
      "entropy": 5.887262535095215,
      "epoch": 5.317447967321533,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002551592704055907,
      "loss": 4.8407,
      "mean_token_accuracy": 0.22473496496677398,
      "num_tokens": 118561466.0,
      "step": 68345
    },
    {
      "entropy": 5.9857958316802975,
      "epoch": 5.31783699669325,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025513166293555836,
      "loss": 4.9155,
      "mean_token_accuracy": 0.22444014698266984,
      "num_tokens": 118569693.0,
      "step": 68350
    },
    {
      "entropy": 5.983163213729858,
      "epoch": 5.318226026064968,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025510405576699636,
      "loss": 4.9726,
      "mean_token_accuracy": 0.206914921104908,
      "num_tokens": 118578217.0,
      "step": 68355
    },
    {
      "entropy": 5.913167285919189,
      "epoch": 5.318615055436686,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002550764489003237,
      "loss": 4.82,
      "mean_token_accuracy": 0.22570654302835463,
      "num_tokens": 118586438.0,
      "step": 68360
    },
    {
      "entropy": 6.013030624389648,
      "epoch": 5.3190040848084035,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002550488423359592,
      "loss": 5.0326,
      "mean_token_accuracy": 0.21578581333160402,
      "num_tokens": 118594665.0,
      "step": 68365
    },
    {
      "entropy": 5.933290910720825,
      "epoch": 5.31939311418012,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025502123607432174,
      "loss": 4.9248,
      "mean_token_accuracy": 0.22106171399354935,
      "num_tokens": 118602899.0,
      "step": 68370
    },
    {
      "entropy": 5.946288394927978,
      "epoch": 5.319782143551838,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00025499363011583024,
      "loss": 4.92,
      "mean_token_accuracy": 0.2160491868853569,
      "num_tokens": 118611319.0,
      "step": 68375
    },
    {
      "entropy": 5.980380535125732,
      "epoch": 5.320171172923556,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002549660244609036,
      "loss": 4.9197,
      "mean_token_accuracy": 0.21199082732200622,
      "num_tokens": 118620312.0,
      "step": 68380
    },
    {
      "entropy": 5.947138166427612,
      "epoch": 5.320560202295273,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002549384191099607,
      "loss": 4.8199,
      "mean_token_accuracy": 0.22625802308321,
      "num_tokens": 118629243.0,
      "step": 68385
    },
    {
      "entropy": 5.9385583877563475,
      "epoch": 5.320949231666991,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00025491081406342025,
      "loss": 4.9144,
      "mean_token_accuracy": 0.23004505932331085,
      "num_tokens": 118637914.0,
      "step": 68390
    },
    {
      "entropy": 5.9176939010620115,
      "epoch": 5.321338261038709,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025488320932170127,
      "loss": 4.859,
      "mean_token_accuracy": 0.2311266303062439,
      "num_tokens": 118646638.0,
      "step": 68395
    },
    {
      "entropy": 5.982489109039307,
      "epoch": 5.321727290410426,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002548556048852226,
      "loss": 4.9968,
      "mean_token_accuracy": 0.21720242202281953,
      "num_tokens": 118655548.0,
      "step": 68400
    },
    {
      "entropy": 5.995305442810059,
      "epoch": 5.322116319782143,
      "grad_norm": 1.609375,
      "learning_rate": 0.00025482800075440296,
      "loss": 4.96,
      "mean_token_accuracy": 0.22541037648916246,
      "num_tokens": 118664120.0,
      "step": 68405
    },
    {
      "entropy": 6.02892918586731,
      "epoch": 5.322505349153861,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002548003969296614,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20813868045806885,
      "num_tokens": 118672479.0,
      "step": 68410
    },
    {
      "entropy": 5.993100118637085,
      "epoch": 5.322894378525579,
      "grad_norm": 1.5,
      "learning_rate": 0.00025477279341141665,
      "loss": 4.9462,
      "mean_token_accuracy": 0.21778721660375594,
      "num_tokens": 118681413.0,
      "step": 68415
    },
    {
      "entropy": 5.911940383911133,
      "epoch": 5.323283407897296,
      "grad_norm": 1.625,
      "learning_rate": 0.0002547451902000875,
      "loss": 4.902,
      "mean_token_accuracy": 0.21896504163742064,
      "num_tokens": 118690059.0,
      "step": 68420
    },
    {
      "entropy": 5.88900671005249,
      "epoch": 5.323672437269014,
      "grad_norm": 1.6875,
      "learning_rate": 0.00025471758729609284,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2289841964840889,
      "num_tokens": 118698304.0,
      "step": 68425
    },
    {
      "entropy": 5.903467750549316,
      "epoch": 5.324061466640732,
      "grad_norm": 1.5,
      "learning_rate": 0.00025468998469985155,
      "loss": 4.9246,
      "mean_token_accuracy": 0.22387812584638594,
      "num_tokens": 118706680.0,
      "step": 68430
    },
    {
      "entropy": 6.032144355773926,
      "epoch": 5.324450496012449,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002546623824117824,
      "loss": 4.9573,
      "mean_token_accuracy": 0.2247374400496483,
      "num_tokens": 118714981.0,
      "step": 68435
    },
    {
      "entropy": 5.9985755443572994,
      "epoch": 5.324839525384166,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002546347804323043,
      "loss": 4.9437,
      "mean_token_accuracy": 0.22131745368242264,
      "num_tokens": 118723043.0,
      "step": 68440
    },
    {
      "entropy": 5.965850782394409,
      "epoch": 5.325228554755884,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002546071787618359,
      "loss": 5.0168,
      "mean_token_accuracy": 0.21590138226747513,
      "num_tokens": 118731049.0,
      "step": 68445
    },
    {
      "entropy": 5.961363506317139,
      "epoch": 5.325617584127602,
      "grad_norm": 1.703125,
      "learning_rate": 0.0002545795774007962,
      "loss": 4.9596,
      "mean_token_accuracy": 0.2234008714556694,
      "num_tokens": 118739809.0,
      "step": 68450
    },
    {
      "entropy": 5.969890117645264,
      "epoch": 5.326006613499319,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002545519763496038,
      "loss": 4.9428,
      "mean_token_accuracy": 0.2242473766207695,
      "num_tokens": 118748400.0,
      "step": 68455
    },
    {
      "entropy": 5.944365453720093,
      "epoch": 5.326395642871037,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025452437560867774,
      "loss": 4.8911,
      "mean_token_accuracy": 0.22004154622554778,
      "num_tokens": 118757225.0,
      "step": 68460
    },
    {
      "entropy": 5.957920122146606,
      "epoch": 5.326784672242754,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002544967751784367,
      "loss": 4.9189,
      "mean_token_accuracy": 0.21668036580085753,
      "num_tokens": 118765493.0,
      "step": 68465
    },
    {
      "entropy": 6.005521821975708,
      "epoch": 5.327173701614472,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025446917505929937,
      "loss": 5.0024,
      "mean_token_accuracy": 0.21753767281770706,
      "num_tokens": 118774822.0,
      "step": 68470
    },
    {
      "entropy": 6.021592807769776,
      "epoch": 5.327562730986189,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002544415752516846,
      "loss": 4.9972,
      "mean_token_accuracy": 0.21167413592338563,
      "num_tokens": 118783091.0,
      "step": 68475
    },
    {
      "entropy": 6.002358293533325,
      "epoch": 5.327951760357907,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00025441397575601125,
      "loss": 5.0597,
      "mean_token_accuracy": 0.2069375604391098,
      "num_tokens": 118792658.0,
      "step": 68480
    },
    {
      "entropy": 6.0177875518798825,
      "epoch": 5.328340789729625,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025438637657269805,
      "loss": 5.0303,
      "mean_token_accuracy": 0.21282542794942855,
      "num_tokens": 118801281.0,
      "step": 68485
    },
    {
      "entropy": 6.019017696380615,
      "epoch": 5.328729819101342,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002543587777021636,
      "loss": 4.9905,
      "mean_token_accuracy": 0.2160816252231598,
      "num_tokens": 118809551.0,
      "step": 68490
    },
    {
      "entropy": 5.943583059310913,
      "epoch": 5.32911884847306,
      "grad_norm": 1.625,
      "learning_rate": 0.0002543311791448271,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21791905611753465,
      "num_tokens": 118818284.0,
      "step": 68495
    },
    {
      "entropy": 6.0152685165405275,
      "epoch": 5.329507877844778,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025430358090110705,
      "loss": 5.0123,
      "mean_token_accuracy": 0.21641748994588852,
      "num_tokens": 118827255.0,
      "step": 68500
    },
    {
      "entropy": 5.953259229660034,
      "epoch": 5.329896907216495,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00025427598297142213,
      "loss": 4.8883,
      "mean_token_accuracy": 0.2179053395986557,
      "num_tokens": 118835761.0,
      "step": 68505
    },
    {
      "entropy": 5.948803329467774,
      "epoch": 5.330285936588212,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002542483853561912,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22528119534254074,
      "num_tokens": 118844107.0,
      "step": 68510
    },
    {
      "entropy": 5.9101897239685055,
      "epoch": 5.33067496595993,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000254220788055833,
      "loss": 4.9372,
      "mean_token_accuracy": 0.21991024166345596,
      "num_tokens": 118853392.0,
      "step": 68515
    },
    {
      "entropy": 5.938868665695191,
      "epoch": 5.331063995331648,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002541931910707662,
      "loss": 4.9074,
      "mean_token_accuracy": 0.2271184042096138,
      "num_tokens": 118861764.0,
      "step": 68520
    },
    {
      "entropy": 6.043102216720581,
      "epoch": 5.331453024703365,
      "grad_norm": 1.5,
      "learning_rate": 0.0002541655944014097,
      "loss": 4.9838,
      "mean_token_accuracy": 0.21622209399938583,
      "num_tokens": 118870389.0,
      "step": 68525
    },
    {
      "entropy": 5.958861303329468,
      "epoch": 5.331842054075083,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002541379980481821,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21867304146289826,
      "num_tokens": 118878623.0,
      "step": 68530
    },
    {
      "entropy": 5.953266954421997,
      "epoch": 5.3322310834468,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002541104020115022,
      "loss": 4.9556,
      "mean_token_accuracy": 0.20903703719377517,
      "num_tokens": 118886590.0,
      "step": 68535
    },
    {
      "entropy": 5.988226699829101,
      "epoch": 5.3326201128185176,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025408280629178864,
      "loss": 5.0029,
      "mean_token_accuracy": 0.21069106906652452,
      "num_tokens": 118895898.0,
      "step": 68540
    },
    {
      "entropy": 6.001074028015137,
      "epoch": 5.333009142190235,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002540552108894603,
      "loss": 4.9355,
      "mean_token_accuracy": 0.2175791159272194,
      "num_tokens": 118904002.0,
      "step": 68545
    },
    {
      "entropy": 5.937737321853637,
      "epoch": 5.333398171561953,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002540276158049358,
      "loss": 4.8776,
      "mean_token_accuracy": 0.2239292711019516,
      "num_tokens": 118912788.0,
      "step": 68550
    },
    {
      "entropy": 5.944146537780762,
      "epoch": 5.333787200933671,
      "grad_norm": 1.546875,
      "learning_rate": 0.00025400002103863367,
      "loss": 4.9869,
      "mean_token_accuracy": 0.21585267633199692,
      "num_tokens": 118921812.0,
      "step": 68555
    },
    {
      "entropy": 5.981778955459594,
      "epoch": 5.334176230305388,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025397242659097293,
      "loss": 4.8479,
      "mean_token_accuracy": 0.2262992739677429,
      "num_tokens": 118930419.0,
      "step": 68560
    },
    {
      "entropy": 5.90911078453064,
      "epoch": 5.334565259677106,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002539448324623722,
      "loss": 4.9074,
      "mean_token_accuracy": 0.22482523918151856,
      "num_tokens": 118939468.0,
      "step": 68565
    },
    {
      "entropy": 5.95219407081604,
      "epoch": 5.334954289048823,
      "grad_norm": 1.53125,
      "learning_rate": 0.00025391723865325,
      "loss": 4.9171,
      "mean_token_accuracy": 0.21627605259418486,
      "num_tokens": 118948722.0,
      "step": 68570
    },
    {
      "entropy": 6.0599876880645756,
      "epoch": 5.3353433184205405,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00025388964516402527,
      "loss": 5.0354,
      "mean_token_accuracy": 0.20854562819004058,
      "num_tokens": 118956737.0,
      "step": 68575
    },
    {
      "entropy": 6.008223581314087,
      "epoch": 5.335732347792258,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002538620519951165,
      "loss": 4.9984,
      "mean_token_accuracy": 0.21353024393320083,
      "num_tokens": 118965474.0,
      "step": 68580
    },
    {
      "entropy": 5.914157199859619,
      "epoch": 5.336121377163976,
      "grad_norm": 1.5,
      "learning_rate": 0.00025383445914694245,
      "loss": 4.8955,
      "mean_token_accuracy": 0.2215004876255989,
      "num_tokens": 118974576.0,
      "step": 68585
    },
    {
      "entropy": 5.90371265411377,
      "epoch": 5.336510406535694,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002538068666199218,
      "loss": 4.8914,
      "mean_token_accuracy": 0.21959423571825026,
      "num_tokens": 118982967.0,
      "step": 68590
    },
    {
      "entropy": 6.0358363628387455,
      "epoch": 5.336899435907411,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002537792744144732,
      "loss": 5.0072,
      "mean_token_accuracy": 0.21556587815284728,
      "num_tokens": 118991356.0,
      "step": 68595
    },
    {
      "entropy": 5.966252040863037,
      "epoch": 5.337288465279128,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002537516825310154,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22226946353912352,
      "num_tokens": 118999709.0,
      "step": 68600
    },
    {
      "entropy": 5.9338422298431395,
      "epoch": 5.337677494650846,
      "grad_norm": 1.5,
      "learning_rate": 0.0002537240909699669,
      "loss": 4.9373,
      "mean_token_accuracy": 0.212609900534153,
      "num_tokens": 119008238.0,
      "step": 68605
    },
    {
      "entropy": 5.8643840789794925,
      "epoch": 5.3380665240225635,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002536964997317465,
      "loss": 4.8493,
      "mean_token_accuracy": 0.2245175674557686,
      "num_tokens": 119017211.0,
      "step": 68610
    },
    {
      "entropy": 5.98246397972107,
      "epoch": 5.338455553394281,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025366890881677284,
      "loss": 5.0023,
      "mean_token_accuracy": 0.2108254462480545,
      "num_tokens": 119027006.0,
      "step": 68615
    },
    {
      "entropy": 5.903141260147095,
      "epoch": 5.338844582765999,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002536413182254645,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21526656448841094,
      "num_tokens": 119035700.0,
      "step": 68620
    },
    {
      "entropy": 6.0115643501281735,
      "epoch": 5.339233612137717,
      "grad_norm": 1.65625,
      "learning_rate": 0.00025361372795824015,
      "loss": 4.969,
      "mean_token_accuracy": 0.2190668135881424,
      "num_tokens": 119043703.0,
      "step": 68625
    },
    {
      "entropy": 5.962553262710571,
      "epoch": 5.339622641509434,
      "grad_norm": 1.5,
      "learning_rate": 0.0002535861380155185,
      "loss": 4.9883,
      "mean_token_accuracy": 0.21768307238817214,
      "num_tokens": 119052404.0,
      "step": 68630
    },
    {
      "entropy": 5.941882514953614,
      "epoch": 5.340011670881151,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025355854839771805,
      "loss": 4.9274,
      "mean_token_accuracy": 0.22336046695709227,
      "num_tokens": 119061377.0,
      "step": 68635
    },
    {
      "entropy": 6.031041860580444,
      "epoch": 5.340400700252869,
      "grad_norm": 1.578125,
      "learning_rate": 0.00025353095910525753,
      "loss": 4.9964,
      "mean_token_accuracy": 0.2178083524107933,
      "num_tokens": 119069206.0,
      "step": 68640
    },
    {
      "entropy": 5.961220407485962,
      "epoch": 5.3407897296245865,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002535033701385555,
      "loss": 4.8179,
      "mean_token_accuracy": 0.22317448407411575,
      "num_tokens": 119077519.0,
      "step": 68645
    },
    {
      "entropy": 5.862696552276612,
      "epoch": 5.341178758996304,
      "grad_norm": 1.5,
      "learning_rate": 0.0002534757814980307,
      "loss": 4.865,
      "mean_token_accuracy": 0.2191338285803795,
      "num_tokens": 119085899.0,
      "step": 68650
    },
    {
      "entropy": 5.961934614181518,
      "epoch": 5.341567788368022,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025344819318410167,
      "loss": 4.9946,
      "mean_token_accuracy": 0.22243429273366927,
      "num_tokens": 119095289.0,
      "step": 68655
    },
    {
      "entropy": 6.075289011001587,
      "epoch": 5.34195681773974,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025342060519718693,
      "loss": 5.0121,
      "mean_token_accuracy": 0.21083366125822067,
      "num_tokens": 119104541.0,
      "step": 68660
    },
    {
      "entropy": 6.024787187576294,
      "epoch": 5.342345847111457,
      "grad_norm": 1.5625,
      "learning_rate": 0.00025339301753770515,
      "loss": 4.9396,
      "mean_token_accuracy": 0.22392162978649138,
      "num_tokens": 119113231.0,
      "step": 68665
    },
    {
      "entropy": 5.962022113800049,
      "epoch": 5.342734876483174,
      "grad_norm": 1.359375,
      "learning_rate": 0.000253365430206075,
      "loss": 5.0162,
      "mean_token_accuracy": 0.21551752239465713,
      "num_tokens": 119123319.0,
      "step": 68670
    },
    {
      "entropy": 5.910531759262085,
      "epoch": 5.343123905854892,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002533378432027149,
      "loss": 4.8891,
      "mean_token_accuracy": 0.2244861215353012,
      "num_tokens": 119132540.0,
      "step": 68675
    },
    {
      "entropy": 6.056602859497071,
      "epoch": 5.3435129352266095,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002533102565280437,
      "loss": 5.061,
      "mean_token_accuracy": 0.21344958245754242,
      "num_tokens": 119141124.0,
      "step": 68680
    },
    {
      "entropy": 5.943327856063843,
      "epoch": 5.343901964598327,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002532826701824797,
      "loss": 4.8952,
      "mean_token_accuracy": 0.22335027754306794,
      "num_tokens": 119149516.0,
      "step": 68685
    },
    {
      "entropy": 5.980960512161255,
      "epoch": 5.344290993970045,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002532550841664416,
      "loss": 4.9985,
      "mean_token_accuracy": 0.21369147151708603,
      "num_tokens": 119157864.0,
      "step": 68690
    },
    {
      "entropy": 5.9514669418334964,
      "epoch": 5.344680023341763,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002532274984803481,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21361672431230544,
      "num_tokens": 119166707.0,
      "step": 68695
    },
    {
      "entropy": 5.935279464721679,
      "epoch": 5.34506905271348,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002531999131246176,
      "loss": 4.9127,
      "mean_token_accuracy": 0.2217701941728592,
      "num_tokens": 119175219.0,
      "step": 68700
    },
    {
      "entropy": 6.015015602111816,
      "epoch": 5.345458082085197,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002531723280996687,
      "loss": 5.0146,
      "mean_token_accuracy": 0.2138627991080284,
      "num_tokens": 119185770.0,
      "step": 68705
    },
    {
      "entropy": 5.9967986106872555,
      "epoch": 5.345847111456915,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025314474340592,
      "loss": 4.9032,
      "mean_token_accuracy": 0.21791941672563553,
      "num_tokens": 119194290.0,
      "step": 68710
    },
    {
      "entropy": 5.938464593887329,
      "epoch": 5.3462361408286325,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025311715904379004,
      "loss": 4.9331,
      "mean_token_accuracy": 0.21329679638147353,
      "num_tokens": 119203060.0,
      "step": 68715
    },
    {
      "entropy": 5.913239669799805,
      "epoch": 5.34662517020035,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025308957501369737,
      "loss": 4.9054,
      "mean_token_accuracy": 0.22053345739841462,
      "num_tokens": 119211562.0,
      "step": 68720
    },
    {
      "entropy": 6.03177924156189,
      "epoch": 5.347014199572068,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025306199131606055,
      "loss": 5.0095,
      "mean_token_accuracy": 0.21362678110599517,
      "num_tokens": 119220598.0,
      "step": 68725
    },
    {
      "entropy": 5.893276357650757,
      "epoch": 5.347403228943786,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002530344079512981,
      "loss": 4.8632,
      "mean_token_accuracy": 0.2317168653011322,
      "num_tokens": 119229297.0,
      "step": 68730
    },
    {
      "entropy": 5.872857284545899,
      "epoch": 5.347792258315502,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025300682491982853,
      "loss": 4.9013,
      "mean_token_accuracy": 0.21773624420166016,
      "num_tokens": 119237595.0,
      "step": 68735
    },
    {
      "entropy": 5.911192989349365,
      "epoch": 5.34818128768722,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002529792422220704,
      "loss": 4.9433,
      "mean_token_accuracy": 0.21790813207626342,
      "num_tokens": 119246319.0,
      "step": 68740
    },
    {
      "entropy": 5.922870016098022,
      "epoch": 5.348570317058938,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025295165985844215,
      "loss": 4.8708,
      "mean_token_accuracy": 0.2242673695087433,
      "num_tokens": 119254635.0,
      "step": 68745
    },
    {
      "entropy": 5.994980335235596,
      "epoch": 5.3489593464306555,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002529240778293624,
      "loss": 5.0166,
      "mean_token_accuracy": 0.21458190381526948,
      "num_tokens": 119262681.0,
      "step": 68750
    },
    {
      "entropy": 5.951914262771607,
      "epoch": 5.349348375802373,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025289649613524954,
      "loss": 4.9498,
      "mean_token_accuracy": 0.22685148566961288,
      "num_tokens": 119271276.0,
      "step": 68755
    },
    {
      "entropy": 5.943306112289429,
      "epoch": 5.349737405174091,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002528689147765223,
      "loss": 4.9846,
      "mean_token_accuracy": 0.21579465121030808,
      "num_tokens": 119280788.0,
      "step": 68760
    },
    {
      "entropy": 5.982010269165039,
      "epoch": 5.350126434545809,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025284133375359904,
      "loss": 5.0165,
      "mean_token_accuracy": 0.21098989099264145,
      "num_tokens": 119289627.0,
      "step": 68765
    },
    {
      "entropy": 5.921328067779541,
      "epoch": 5.350515463917525,
      "grad_norm": 1.546875,
      "learning_rate": 0.00025281375306689824,
      "loss": 4.8808,
      "mean_token_accuracy": 0.22521564811468125,
      "num_tokens": 119298038.0,
      "step": 68770
    },
    {
      "entropy": 5.950640058517456,
      "epoch": 5.350904493289243,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025278617271683846,
      "loss": 4.917,
      "mean_token_accuracy": 0.22160522192716597,
      "num_tokens": 119306371.0,
      "step": 68775
    },
    {
      "entropy": 5.956131553649902,
      "epoch": 5.351293522660961,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002527585927038381,
      "loss": 4.9283,
      "mean_token_accuracy": 0.22353136092424392,
      "num_tokens": 119315032.0,
      "step": 68780
    },
    {
      "entropy": 5.923272752761841,
      "epoch": 5.3516825520326785,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025273101302831575,
      "loss": 4.9689,
      "mean_token_accuracy": 0.22264983952045442,
      "num_tokens": 119323657.0,
      "step": 68785
    },
    {
      "entropy": 5.89572434425354,
      "epoch": 5.352071581404396,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025270343369068976,
      "loss": 4.8189,
      "mean_token_accuracy": 0.22656875997781753,
      "num_tokens": 119332461.0,
      "step": 68790
    },
    {
      "entropy": 5.9316164493560795,
      "epoch": 5.352460610776114,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002526758546913787,
      "loss": 4.9257,
      "mean_token_accuracy": 0.21910176426172256,
      "num_tokens": 119341213.0,
      "step": 68795
    },
    {
      "entropy": 5.902061414718628,
      "epoch": 5.352849640147831,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002526482760308011,
      "loss": 4.9052,
      "mean_token_accuracy": 0.22096736878156661,
      "num_tokens": 119349547.0,
      "step": 68800
    },
    {
      "entropy": 5.9578498840332035,
      "epoch": 5.353238669519548,
      "grad_norm": 1.59375,
      "learning_rate": 0.00025262069770937525,
      "loss": 4.9621,
      "mean_token_accuracy": 0.2123180150985718,
      "num_tokens": 119358854.0,
      "step": 68805
    },
    {
      "entropy": 5.978305673599243,
      "epoch": 5.353627698891266,
      "grad_norm": 1.5,
      "learning_rate": 0.0002525931197275197,
      "loss": 4.9178,
      "mean_token_accuracy": 0.21753152459859848,
      "num_tokens": 119368352.0,
      "step": 68810
    },
    {
      "entropy": 5.995351791381836,
      "epoch": 5.354016728262984,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025256554208565296,
      "loss": 4.9632,
      "mean_token_accuracy": 0.21783646643161775,
      "num_tokens": 119376818.0,
      "step": 68815
    },
    {
      "entropy": 5.967716693878174,
      "epoch": 5.3544057576347015,
      "grad_norm": 1.53125,
      "learning_rate": 0.00025253796478419323,
      "loss": 4.9243,
      "mean_token_accuracy": 0.21736811250448226,
      "num_tokens": 119386080.0,
      "step": 68820
    },
    {
      "entropy": 5.9799854278564455,
      "epoch": 5.354794787006419,
      "grad_norm": 1.640625,
      "learning_rate": 0.0002525103878235593,
      "loss": 4.946,
      "mean_token_accuracy": 0.2186930298805237,
      "num_tokens": 119394672.0,
      "step": 68825
    },
    {
      "entropy": 5.9518153190612795,
      "epoch": 5.355183816378137,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002524828112041694,
      "loss": 4.9276,
      "mean_token_accuracy": 0.22101527452468872,
      "num_tokens": 119402952.0,
      "step": 68830
    },
    {
      "entropy": 5.923309564590454,
      "epoch": 5.355572845749854,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00025245523492644197,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2108168348670006,
      "num_tokens": 119411833.0,
      "step": 68835
    },
    {
      "entropy": 5.863316869735717,
      "epoch": 5.355961875121571,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002524276589907955,
      "loss": 4.7678,
      "mean_token_accuracy": 0.2340243935585022,
      "num_tokens": 119420671.0,
      "step": 68840
    },
    {
      "entropy": 5.980744647979736,
      "epoch": 5.356350904493289,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00025240008339764844,
      "loss": 5.0261,
      "mean_token_accuracy": 0.21652208566665648,
      "num_tokens": 119430072.0,
      "step": 68845
    },
    {
      "entropy": 6.012799406051636,
      "epoch": 5.356739933865007,
      "grad_norm": 1.5625,
      "learning_rate": 0.000252372508147419,
      "loss": 4.9887,
      "mean_token_accuracy": 0.21683751940727233,
      "num_tokens": 119438882.0,
      "step": 68850
    },
    {
      "entropy": 6.049327325820923,
      "epoch": 5.3571289632367245,
      "grad_norm": 1.5,
      "learning_rate": 0.0002523449332405258,
      "loss": 5.0186,
      "mean_token_accuracy": 0.21043152064085008,
      "num_tokens": 119447454.0,
      "step": 68855
    },
    {
      "entropy": 6.023149633407593,
      "epoch": 5.357517992608442,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002523173586773872,
      "loss": 4.99,
      "mean_token_accuracy": 0.2194528415799141,
      "num_tokens": 119455393.0,
      "step": 68860
    },
    {
      "entropy": 5.895042753219604,
      "epoch": 5.35790702198016,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002522897844584216,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2255195766687393,
      "num_tokens": 119463769.0,
      "step": 68865
    },
    {
      "entropy": 5.91559362411499,
      "epoch": 5.358296051351877,
      "grad_norm": 1.53125,
      "learning_rate": 0.00025226221058404733,
      "loss": 4.9978,
      "mean_token_accuracy": 0.21756433099508285,
      "num_tokens": 119473026.0,
      "step": 68870
    },
    {
      "entropy": 5.951250457763672,
      "epoch": 5.358685080723594,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002522346370546828,
      "loss": 4.894,
      "mean_token_accuracy": 0.22102364897727966,
      "num_tokens": 119480965.0,
      "step": 68875
    },
    {
      "entropy": 5.952284908294677,
      "epoch": 5.359074110095312,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002522070638707465,
      "loss": 4.8487,
      "mean_token_accuracy": 0.2278127208352089,
      "num_tokens": 119489273.0,
      "step": 68880
    },
    {
      "entropy": 5.860654592514038,
      "epoch": 5.35946313946703,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025217949103265665,
      "loss": 4.8856,
      "mean_token_accuracy": 0.2216194674372673,
      "num_tokens": 119497914.0,
      "step": 68885
    },
    {
      "entropy": 5.956837701797485,
      "epoch": 5.3598521688387475,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00025215191854083166,
      "loss": 5.0194,
      "mean_token_accuracy": 0.21451205611228943,
      "num_tokens": 119506963.0,
      "step": 68890
    },
    {
      "entropy": 5.969401216506958,
      "epoch": 5.360241198210465,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025212434639568996,
      "loss": 5.0005,
      "mean_token_accuracy": 0.20681636929512023,
      "num_tokens": 119515787.0,
      "step": 68895
    },
    {
      "entropy": 6.029029846191406,
      "epoch": 5.360630227582183,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025209677459765,
      "loss": 5.0569,
      "mean_token_accuracy": 0.20825241208076478,
      "num_tokens": 119523562.0,
      "step": 68900
    },
    {
      "entropy": 5.988119888305664,
      "epoch": 5.3610192569539,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002520692031471299,
      "loss": 5.0264,
      "mean_token_accuracy": 0.21438842117786408,
      "num_tokens": 119532526.0,
      "step": 68905
    },
    {
      "entropy": 5.924050760269165,
      "epoch": 5.361408286325617,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025204163204454826,
      "loss": 4.9478,
      "mean_token_accuracy": 0.21393176764249802,
      "num_tokens": 119540844.0,
      "step": 68910
    },
    {
      "entropy": 6.0246318817138675,
      "epoch": 5.361797315697335,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002520140612903233,
      "loss": 5.0154,
      "mean_token_accuracy": 0.21892350167036057,
      "num_tokens": 119550185.0,
      "step": 68915
    },
    {
      "entropy": 5.993782663345337,
      "epoch": 5.362186345069053,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002519864908848733,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21636709272861482,
      "num_tokens": 119559493.0,
      "step": 68920
    },
    {
      "entropy": 6.001300716400147,
      "epoch": 5.3625753744407705,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025195892082861674,
      "loss": 4.9085,
      "mean_token_accuracy": 0.22554118931293488,
      "num_tokens": 119567876.0,
      "step": 68925
    },
    {
      "entropy": 6.000425863265991,
      "epoch": 5.362964403812488,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002519313511219719,
      "loss": 4.9492,
      "mean_token_accuracy": 0.2203022673726082,
      "num_tokens": 119576463.0,
      "step": 68930
    },
    {
      "entropy": 5.993487453460693,
      "epoch": 5.363353433184205,
      "grad_norm": 1.546875,
      "learning_rate": 0.00025190378176535703,
      "loss": 4.9074,
      "mean_token_accuracy": 0.22356517612934113,
      "num_tokens": 119584947.0,
      "step": 68935
    },
    {
      "entropy": 5.936607694625854,
      "epoch": 5.363742462555923,
      "grad_norm": 1.5,
      "learning_rate": 0.0002518762127591905,
      "loss": 4.8393,
      "mean_token_accuracy": 0.2291971743106842,
      "num_tokens": 119593005.0,
      "step": 68940
    },
    {
      "entropy": 5.870314455032348,
      "epoch": 5.36413149192764,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002518486441038907,
      "loss": 4.9061,
      "mean_token_accuracy": 0.21924936771392822,
      "num_tokens": 119601849.0,
      "step": 68945
    },
    {
      "entropy": 5.920074892044068,
      "epoch": 5.364520521299358,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002518210757998758,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2212374359369278,
      "num_tokens": 119610065.0,
      "step": 68950
    },
    {
      "entropy": 5.937868976593018,
      "epoch": 5.364909550671076,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002517935078475642,
      "loss": 4.8637,
      "mean_token_accuracy": 0.22682141065597533,
      "num_tokens": 119618770.0,
      "step": 68955
    },
    {
      "entropy": 6.048112106323242,
      "epoch": 5.3652985800427935,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002517659402473742,
      "loss": 5.0452,
      "mean_token_accuracy": 0.2091663584113121,
      "num_tokens": 119627768.0,
      "step": 68960
    },
    {
      "entropy": 6.023472642898559,
      "epoch": 5.365687609414511,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002517383729997241,
      "loss": 5.0737,
      "mean_token_accuracy": 0.2108045145869255,
      "num_tokens": 119636260.0,
      "step": 68965
    },
    {
      "entropy": 6.016500473022461,
      "epoch": 5.366076638786228,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002517108061050321,
      "loss": 5.054,
      "mean_token_accuracy": 0.21104868203401567,
      "num_tokens": 119645451.0,
      "step": 68970
    },
    {
      "entropy": 5.973011016845703,
      "epoch": 5.366465668157946,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025168323956371657,
      "loss": 4.9712,
      "mean_token_accuracy": 0.2180153101682663,
      "num_tokens": 119654699.0,
      "step": 68975
    },
    {
      "entropy": 6.032042121887207,
      "epoch": 5.366854697529663,
      "grad_norm": 1.53125,
      "learning_rate": 0.00025165567337619583,
      "loss": 5.0174,
      "mean_token_accuracy": 0.20992347300052644,
      "num_tokens": 119663082.0,
      "step": 68980
    },
    {
      "entropy": 5.9762200832366945,
      "epoch": 5.367243726901381,
      "grad_norm": 1.484375,
      "learning_rate": 0.000251628107542888,
      "loss": 4.9704,
      "mean_token_accuracy": 0.21943359822034836,
      "num_tokens": 119671268.0,
      "step": 68985
    },
    {
      "entropy": 5.938281059265137,
      "epoch": 5.367632756273099,
      "grad_norm": 1.546875,
      "learning_rate": 0.00025160054206421143,
      "loss": 4.9808,
      "mean_token_accuracy": 0.21456556469202043,
      "num_tokens": 119680070.0,
      "step": 68990
    },
    {
      "entropy": 5.933156585693359,
      "epoch": 5.3680217856448165,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002515729769405845,
      "loss": 4.8534,
      "mean_token_accuracy": 0.22508912831544875,
      "num_tokens": 119689403.0,
      "step": 68995
    },
    {
      "entropy": 5.866895914077759,
      "epoch": 5.368410815016533,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002515454121724253,
      "loss": 4.7769,
      "mean_token_accuracy": 0.23196562379598618,
      "num_tokens": 119697844.0,
      "step": 69000
    },
    {
      "epoch": 5.368410815016533,
      "eval_entropy": 5.6759912096157565,
      "eval_loss": 5.055768966674805,
      "eval_mean_token_accuracy": 0.2220340483213297,
      "eval_num_tokens": 119697844.0,
      "eval_runtime": 21.809,
      "eval_samples_per_second": 1905.547,
      "eval_steps_per_second": 238.205,
      "step": 69000
    },
    {
      "entropy": 6.009350156784057,
      "epoch": 5.368799844388251,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025151784776015216,
      "loss": 5.0461,
      "mean_token_accuracy": 0.21573394536972046,
      "num_tokens": 119707171.0,
      "step": 69005
    },
    {
      "entropy": 5.9767499446868895,
      "epoch": 5.369188873759969,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002514902837041833,
      "loss": 5.0589,
      "mean_token_accuracy": 0.21176643669605255,
      "num_tokens": 119715897.0,
      "step": 69010
    },
    {
      "entropy": 5.961584901809692,
      "epoch": 5.369577903131686,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025146272000493695,
      "loss": 4.9272,
      "mean_token_accuracy": 0.2251289814710617,
      "num_tokens": 119724480.0,
      "step": 69015
    },
    {
      "entropy": 6.04992642402649,
      "epoch": 5.369966932503404,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002514351566628313,
      "loss": 5.0425,
      "mean_token_accuracy": 0.2147315561771393,
      "num_tokens": 119733005.0,
      "step": 69020
    },
    {
      "entropy": 6.094898653030396,
      "epoch": 5.370355961875122,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002514075936782847,
      "loss": 5.0591,
      "mean_token_accuracy": 0.21183206588029863,
      "num_tokens": 119741243.0,
      "step": 69025
    },
    {
      "entropy": 5.960351753234863,
      "epoch": 5.3707449912468395,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002513800310517153,
      "loss": 5.0027,
      "mean_token_accuracy": 0.2116766095161438,
      "num_tokens": 119749343.0,
      "step": 69030
    },
    {
      "entropy": 5.962866926193238,
      "epoch": 5.371134020618557,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002513524687835414,
      "loss": 4.9898,
      "mean_token_accuracy": 0.21040924340486528,
      "num_tokens": 119757708.0,
      "step": 69035
    },
    {
      "entropy": 5.946325826644897,
      "epoch": 5.371523049990274,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002513249068741811,
      "loss": 4.9451,
      "mean_token_accuracy": 0.214374540746212,
      "num_tokens": 119767146.0,
      "step": 69040
    },
    {
      "entropy": 5.975495672225952,
      "epoch": 5.371912079361992,
      "grad_norm": 1.5625,
      "learning_rate": 0.00025129734532405265,
      "loss": 4.9711,
      "mean_token_accuracy": 0.2156537339091301,
      "num_tokens": 119775046.0,
      "step": 69045
    },
    {
      "entropy": 6.006795835494995,
      "epoch": 5.372301108733709,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025126978413357424,
      "loss": 4.9572,
      "mean_token_accuracy": 0.216485196352005,
      "num_tokens": 119783336.0,
      "step": 69050
    },
    {
      "entropy": 5.910364151000977,
      "epoch": 5.372690138105427,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002512422233031641,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21487195044755936,
      "num_tokens": 119792659.0,
      "step": 69055
    },
    {
      "entropy": 5.893155813217163,
      "epoch": 5.373079167477145,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002512146628332404,
      "loss": 4.892,
      "mean_token_accuracy": 0.211874321103096,
      "num_tokens": 119800913.0,
      "step": 69060
    },
    {
      "entropy": 6.048960638046265,
      "epoch": 5.3734681968488625,
      "grad_norm": 1.6875,
      "learning_rate": 0.0002511871027242214,
      "loss": 5.0652,
      "mean_token_accuracy": 0.20502713620662688,
      "num_tokens": 119809011.0,
      "step": 69065
    },
    {
      "entropy": 5.94715142250061,
      "epoch": 5.373857226220579,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002511595429765251,
      "loss": 4.8936,
      "mean_token_accuracy": 0.2197125792503357,
      "num_tokens": 119817839.0,
      "step": 69070
    },
    {
      "entropy": 6.030389451980591,
      "epoch": 5.374246255592297,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00025113198359056986,
      "loss": 5.0478,
      "mean_token_accuracy": 0.21966323852539063,
      "num_tokens": 119826309.0,
      "step": 69075
    },
    {
      "entropy": 5.977067613601685,
      "epoch": 5.374635284964015,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002511044245667738,
      "loss": 4.9675,
      "mean_token_accuracy": 0.21827393919229507,
      "num_tokens": 119834630.0,
      "step": 69080
    },
    {
      "entropy": 5.986836051940918,
      "epoch": 5.375024314335732,
      "grad_norm": 1.59375,
      "learning_rate": 0.000251076865905555,
      "loss": 4.9787,
      "mean_token_accuracy": 0.21972154527902604,
      "num_tokens": 119843079.0,
      "step": 69085
    },
    {
      "entropy": 6.018704652786255,
      "epoch": 5.37541334370745,
      "grad_norm": 1.5,
      "learning_rate": 0.00025104930760733163,
      "loss": 5.0091,
      "mean_token_accuracy": 0.2117392063140869,
      "num_tokens": 119851594.0,
      "step": 69090
    },
    {
      "entropy": 6.061796617507935,
      "epoch": 5.375802373079168,
      "grad_norm": 1.53125,
      "learning_rate": 0.000251021749672522,
      "loss": 5.0208,
      "mean_token_accuracy": 0.2097836196422577,
      "num_tokens": 119859882.0,
      "step": 69095
    },
    {
      "entropy": 5.951039886474609,
      "epoch": 5.3761914024508854,
      "grad_norm": 1.578125,
      "learning_rate": 0.00025099419210154413,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21719225645065307,
      "num_tokens": 119868424.0,
      "step": 69100
    },
    {
      "entropy": 5.882094097137451,
      "epoch": 5.376580431822602,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002509666348948162,
      "loss": 4.8073,
      "mean_token_accuracy": 0.22546813637018204,
      "num_tokens": 119877691.0,
      "step": 69105
    },
    {
      "entropy": 6.0222697257995605,
      "epoch": 5.37696946119432,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025093907805275636,
      "loss": 5.0194,
      "mean_token_accuracy": 0.20675883889198304,
      "num_tokens": 119886146.0,
      "step": 69110
    },
    {
      "entropy": 5.9277040481567385,
      "epoch": 5.377358490566038,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025091152157578266,
      "loss": 4.829,
      "mean_token_accuracy": 0.22571858018636703,
      "num_tokens": 119894692.0,
      "step": 69115
    },
    {
      "entropy": 6.020661449432373,
      "epoch": 5.377747519937755,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002508839654643133,
      "loss": 5.0239,
      "mean_token_accuracy": 0.21016445010900497,
      "num_tokens": 119904777.0,
      "step": 69120
    },
    {
      "entropy": 5.988295364379883,
      "epoch": 5.378136549309473,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002508564097187664,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2235170915722847,
      "num_tokens": 119913136.0,
      "step": 69125
    },
    {
      "entropy": 5.9395452499389645,
      "epoch": 5.378525578681191,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00025082885433956007,
      "loss": 4.8699,
      "mean_token_accuracy": 0.21961246579885482,
      "num_tokens": 119921571.0,
      "step": 69130
    },
    {
      "entropy": 5.9511559009552,
      "epoch": 5.3789146080529076,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025080129932711234,
      "loss": 4.9268,
      "mean_token_accuracy": 0.22589248716831206,
      "num_tokens": 119929520.0,
      "step": 69135
    },
    {
      "entropy": 5.959663963317871,
      "epoch": 5.379303637424625,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002507737446818414,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2242049589753151,
      "num_tokens": 119937477.0,
      "step": 69140
    },
    {
      "entropy": 5.9346459865570065,
      "epoch": 5.379692666796343,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002507461904041653,
      "loss": 4.8859,
      "mean_token_accuracy": 0.2235795497894287,
      "num_tokens": 119945927.0,
      "step": 69145
    },
    {
      "entropy": 5.9584068775177,
      "epoch": 5.380081696168061,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025071863649450216,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21586691290140153,
      "num_tokens": 119954066.0,
      "step": 69150
    },
    {
      "entropy": 5.970383834838867,
      "epoch": 5.380470725539778,
      "grad_norm": 1.53125,
      "learning_rate": 0.00025069108295327006,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2197898104786873,
      "num_tokens": 119962925.0,
      "step": 69155
    },
    {
      "entropy": 5.985500478744507,
      "epoch": 5.380859754911496,
      "grad_norm": 1.59375,
      "learning_rate": 0.00025066352978088706,
      "loss": 4.9597,
      "mean_token_accuracy": 0.22405468970537185,
      "num_tokens": 119971451.0,
      "step": 69160
    },
    {
      "entropy": 5.944410133361816,
      "epoch": 5.381248784283214,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002506359769777713,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22176973074674605,
      "num_tokens": 119979851.0,
      "step": 69165
    },
    {
      "entropy": 5.943067502975464,
      "epoch": 5.3816378136549305,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002506084245443408,
      "loss": 4.9228,
      "mean_token_accuracy": 0.21691677570343018,
      "num_tokens": 119988877.0,
      "step": 69170
    },
    {
      "entropy": 5.969129180908203,
      "epoch": 5.382026843026648,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002505808724810137,
      "loss": 5.0873,
      "mean_token_accuracy": 0.21134822815656662,
      "num_tokens": 119997324.0,
      "step": 69175
    },
    {
      "entropy": 5.942006778717041,
      "epoch": 5.382415872398366,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025055332078820794,
      "loss": 5.0195,
      "mean_token_accuracy": 0.2085981547832489,
      "num_tokens": 120006171.0,
      "step": 69180
    },
    {
      "entropy": 6.032240676879883,
      "epoch": 5.382804901770084,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002505257694663416,
      "loss": 4.9768,
      "mean_token_accuracy": 0.21140389889478683,
      "num_tokens": 120014812.0,
      "step": 69185
    },
    {
      "entropy": 6.0560871124267575,
      "epoch": 5.383193931141801,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002504982185158328,
      "loss": 5.0074,
      "mean_token_accuracy": 0.21245020031929016,
      "num_tokens": 120023437.0,
      "step": 69190
    },
    {
      "entropy": 5.920071125030518,
      "epoch": 5.383582960513519,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002504706679370995,
      "loss": 4.8072,
      "mean_token_accuracy": 0.228926420211792,
      "num_tokens": 120031540.0,
      "step": 69195
    },
    {
      "entropy": 5.942682266235352,
      "epoch": 5.383971989885237,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002504431177305598,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2171502470970154,
      "num_tokens": 120040093.0,
      "step": 69200
    },
    {
      "entropy": 5.996459531784057,
      "epoch": 5.3843610192569535,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002504155678966317,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21904054284095764,
      "num_tokens": 120047822.0,
      "step": 69205
    },
    {
      "entropy": 5.991739082336426,
      "epoch": 5.384750048628671,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002503880184357333,
      "loss": 5.0335,
      "mean_token_accuracy": 0.20824511498212814,
      "num_tokens": 120056295.0,
      "step": 69210
    },
    {
      "entropy": 5.976065826416016,
      "epoch": 5.385139078000389,
      "grad_norm": 1.59375,
      "learning_rate": 0.00025036046934828244,
      "loss": 4.8966,
      "mean_token_accuracy": 0.2248357802629471,
      "num_tokens": 120065274.0,
      "step": 69215
    },
    {
      "entropy": 6.010432338714599,
      "epoch": 5.385528107372107,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025033292063469714,
      "loss": 4.9251,
      "mean_token_accuracy": 0.21944453865289687,
      "num_tokens": 120074162.0,
      "step": 69220
    },
    {
      "entropy": 5.90080156326294,
      "epoch": 5.385917136743824,
      "grad_norm": 1.546875,
      "learning_rate": 0.00025030537229539567,
      "loss": 4.8955,
      "mean_token_accuracy": 0.22178859412670135,
      "num_tokens": 120083208.0,
      "step": 69225
    },
    {
      "entropy": 5.910979413986206,
      "epoch": 5.386306166115542,
      "grad_norm": 1.5,
      "learning_rate": 0.0002502778243307958,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21390208452939988,
      "num_tokens": 120092264.0,
      "step": 69230
    },
    {
      "entropy": 5.933062887191772,
      "epoch": 5.38669519548726,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025025027674131567,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21369398832321168,
      "num_tokens": 120100654.0,
      "step": 69235
    },
    {
      "entropy": 5.9579109191894535,
      "epoch": 5.3870842248589765,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00025022272952737317,
      "loss": 4.8984,
      "mean_token_accuracy": 0.21949789375066758,
      "num_tokens": 120108974.0,
      "step": 69240
    },
    {
      "entropy": 6.00877013206482,
      "epoch": 5.387473254230694,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002501951826893863,
      "loss": 4.9163,
      "mean_token_accuracy": 0.216525998711586,
      "num_tokens": 120118110.0,
      "step": 69245
    },
    {
      "entropy": 5.987317895889282,
      "epoch": 5.387862283602412,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002501676362277731,
      "loss": 4.9412,
      "mean_token_accuracy": 0.22301646322011948,
      "num_tokens": 120126110.0,
      "step": 69250
    },
    {
      "entropy": 5.912262868881226,
      "epoch": 5.38825131297413,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025014009014295136,
      "loss": 4.9656,
      "mean_token_accuracy": 0.21361180394887924,
      "num_tokens": 120135122.0,
      "step": 69255
    },
    {
      "entropy": 5.950713443756103,
      "epoch": 5.388640342345847,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002501125444353393,
      "loss": 4.9308,
      "mean_token_accuracy": 0.22395459264516832,
      "num_tokens": 120143745.0,
      "step": 69260
    },
    {
      "entropy": 5.993258094787597,
      "epoch": 5.389029371717565,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002500849991053547,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20731472820043564,
      "num_tokens": 120152005.0,
      "step": 69265
    },
    {
      "entropy": 5.922664165496826,
      "epoch": 5.389418401089282,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025005745415341565,
      "loss": 4.905,
      "mean_token_accuracy": 0.22606851011514664,
      "num_tokens": 120161206.0,
      "step": 69270
    },
    {
      "entropy": 5.994931745529175,
      "epoch": 5.3898074304609995,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00025002990957994,
      "loss": 4.9464,
      "mean_token_accuracy": 0.2155529424548149,
      "num_tokens": 120169199.0,
      "step": 69275
    },
    {
      "entropy": 5.94662446975708,
      "epoch": 5.390196459832717,
      "grad_norm": 1.546875,
      "learning_rate": 0.00025000236538534577,
      "loss": 4.9554,
      "mean_token_accuracy": 0.2191656529903412,
      "num_tokens": 120177959.0,
      "step": 69280
    },
    {
      "entropy": 5.960764312744141,
      "epoch": 5.390585489204435,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024997482157005074,
      "loss": 4.8922,
      "mean_token_accuracy": 0.22175944298505784,
      "num_tokens": 120186679.0,
      "step": 69285
    },
    {
      "entropy": 5.921430540084839,
      "epoch": 5.390974518576153,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000249947278134473,
      "loss": 4.8694,
      "mean_token_accuracy": 0.22561158686876298,
      "num_tokens": 120195186.0,
      "step": 69290
    },
    {
      "entropy": 5.995001173019409,
      "epoch": 5.39136354794787,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002499197350790305,
      "loss": 5.0223,
      "mean_token_accuracy": 0.21821577250957488,
      "num_tokens": 120202821.0,
      "step": 69295
    },
    {
      "entropy": 5.92721118927002,
      "epoch": 5.391752577319588,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024989219240414105,
      "loss": 4.9797,
      "mean_token_accuracy": 0.21684201806783676,
      "num_tokens": 120211429.0,
      "step": 69300
    },
    {
      "entropy": 5.98551926612854,
      "epoch": 5.392141606691305,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024986465011022263,
      "loss": 4.9803,
      "mean_token_accuracy": 0.21427216082811357,
      "num_tokens": 120220096.0,
      "step": 69305
    },
    {
      "entropy": 5.9349600315094,
      "epoch": 5.3925306360630225,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024983710819769325,
      "loss": 4.85,
      "mean_token_accuracy": 0.22334273606538774,
      "num_tokens": 120228512.0,
      "step": 69310
    },
    {
      "entropy": 5.979427671432495,
      "epoch": 5.39291966543474,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002498095666669706,
      "loss": 4.9899,
      "mean_token_accuracy": 0.21298328191041946,
      "num_tokens": 120236897.0,
      "step": 69315
    },
    {
      "entropy": 6.004055786132812,
      "epoch": 5.393308694806458,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002497820255184727,
      "loss": 4.9936,
      "mean_token_accuracy": 0.2093295305967331,
      "num_tokens": 120245533.0,
      "step": 69320
    },
    {
      "entropy": 6.024666595458984,
      "epoch": 5.393697724178176,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024975448475261737,
      "loss": 5.0007,
      "mean_token_accuracy": 0.2150074690580368,
      "num_tokens": 120255332.0,
      "step": 69325
    },
    {
      "entropy": 6.042718553543091,
      "epoch": 5.394086753549893,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024972694436982255,
      "loss": 5.0096,
      "mean_token_accuracy": 0.21146293878555297,
      "num_tokens": 120264441.0,
      "step": 69330
    },
    {
      "entropy": 5.890518522262573,
      "epoch": 5.39447578292161,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002496994043705062,
      "loss": 4.804,
      "mean_token_accuracy": 0.22933408468961716,
      "num_tokens": 120272372.0,
      "step": 69335
    },
    {
      "entropy": 5.912908840179443,
      "epoch": 5.394864812293328,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002496718647550861,
      "loss": 4.9016,
      "mean_token_accuracy": 0.21331263482570648,
      "num_tokens": 120280372.0,
      "step": 69340
    },
    {
      "entropy": 5.944162178039551,
      "epoch": 5.3952538416650455,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002496443255239801,
      "loss": 4.9548,
      "mean_token_accuracy": 0.21944860368967056,
      "num_tokens": 120288786.0,
      "step": 69345
    },
    {
      "entropy": 5.920022296905517,
      "epoch": 5.395642871036763,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002496167866776061,
      "loss": 4.8029,
      "mean_token_accuracy": 0.2265725016593933,
      "num_tokens": 120297624.0,
      "step": 69350
    },
    {
      "entropy": 5.915057516098022,
      "epoch": 5.396031900408481,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024958924821638195,
      "loss": 4.8956,
      "mean_token_accuracy": 0.22712543457746506,
      "num_tokens": 120305764.0,
      "step": 69355
    },
    {
      "entropy": 5.99413480758667,
      "epoch": 5.396420929780199,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024956171014072555,
      "loss": 5.0702,
      "mean_token_accuracy": 0.2094024211168289,
      "num_tokens": 120314337.0,
      "step": 69360
    },
    {
      "entropy": 6.000122737884522,
      "epoch": 5.396809959151916,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024953417245105467,
      "loss": 4.9776,
      "mean_token_accuracy": 0.22048172801733018,
      "num_tokens": 120322751.0,
      "step": 69365
    },
    {
      "entropy": 5.957219648361206,
      "epoch": 5.397198988523634,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024950663514778726,
      "loss": 4.9266,
      "mean_token_accuracy": 0.2189728155732155,
      "num_tokens": 120331581.0,
      "step": 69370
    },
    {
      "entropy": 5.924093198776245,
      "epoch": 5.397588017895351,
      "grad_norm": 1.5625,
      "learning_rate": 0.00024947909823134107,
      "loss": 4.9409,
      "mean_token_accuracy": 0.2202657863497734,
      "num_tokens": 120339686.0,
      "step": 69375
    },
    {
      "entropy": 5.965054512023926,
      "epoch": 5.3979770472670685,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002494515617021339,
      "loss": 4.9728,
      "mean_token_accuracy": 0.21553620398044587,
      "num_tokens": 120348328.0,
      "step": 69380
    },
    {
      "entropy": 5.949032545089722,
      "epoch": 5.398366076638786,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024942402556058364,
      "loss": 4.9011,
      "mean_token_accuracy": 0.22199678868055345,
      "num_tokens": 120356025.0,
      "step": 69385
    },
    {
      "entropy": 6.00901517868042,
      "epoch": 5.398755106010504,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002493964898071081,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20385624468326569,
      "num_tokens": 120364036.0,
      "step": 69390
    },
    {
      "entropy": 5.958004331588745,
      "epoch": 5.399144135382222,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002493689544421251,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21645919531583785,
      "num_tokens": 120373015.0,
      "step": 69395
    },
    {
      "entropy": 6.008584594726562,
      "epoch": 5.399533164753939,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024934141946605237,
      "loss": 4.939,
      "mean_token_accuracy": 0.21706321090459824,
      "num_tokens": 120381342.0,
      "step": 69400
    },
    {
      "entropy": 5.9857641696929935,
      "epoch": 5.399922194125656,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024931388487930777,
      "loss": 4.9121,
      "mean_token_accuracy": 0.20908040404319764,
      "num_tokens": 120389689.0,
      "step": 69405
    },
    {
      "entropy": 5.987439966201782,
      "epoch": 5.400311223497374,
      "grad_norm": 1.40625,
      "learning_rate": 0.000249286350682309,
      "loss": 4.9487,
      "mean_token_accuracy": 0.21842693984508516,
      "num_tokens": 120397983.0,
      "step": 69410
    },
    {
      "entropy": 5.9292192459106445,
      "epoch": 5.4007002528690915,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002492588168754741,
      "loss": 4.8886,
      "mean_token_accuracy": 0.22557025998830796,
      "num_tokens": 120406315.0,
      "step": 69415
    },
    {
      "entropy": 5.899976634979248,
      "epoch": 5.401089282240809,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024923128345922055,
      "loss": 4.9379,
      "mean_token_accuracy": 0.22425064593553543,
      "num_tokens": 120414771.0,
      "step": 69420
    },
    {
      "entropy": 5.856578350067139,
      "epoch": 5.401478311612527,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002492037504339663,
      "loss": 4.8301,
      "mean_token_accuracy": 0.22903923392295839,
      "num_tokens": 120424159.0,
      "step": 69425
    },
    {
      "entropy": 5.888695955276489,
      "epoch": 5.401867340984245,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00024917621780012905,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22793111205101013,
      "num_tokens": 120432187.0,
      "step": 69430
    },
    {
      "entropy": 5.946759843826294,
      "epoch": 5.402256370355962,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024914868555812663,
      "loss": 4.9704,
      "mean_token_accuracy": 0.2170004203915596,
      "num_tokens": 120441265.0,
      "step": 69435
    },
    {
      "entropy": 5.89326229095459,
      "epoch": 5.402645399727679,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002491211537083768,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22099369764328003,
      "num_tokens": 120450183.0,
      "step": 69440
    },
    {
      "entropy": 6.00045747756958,
      "epoch": 5.403034429099397,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0002490936222512972,
      "loss": 5.0367,
      "mean_token_accuracy": 0.21045224815607072,
      "num_tokens": 120458103.0,
      "step": 69445
    },
    {
      "entropy": 5.925220537185669,
      "epoch": 5.4034234584711145,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002490660911873057,
      "loss": 4.9341,
      "mean_token_accuracy": 0.21810067743062972,
      "num_tokens": 120466181.0,
      "step": 69450
    },
    {
      "entropy": 5.95880913734436,
      "epoch": 5.403812487842832,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024903856051681995,
      "loss": 4.8988,
      "mean_token_accuracy": 0.20949715822935105,
      "num_tokens": 120474240.0,
      "step": 69455
    },
    {
      "entropy": 5.940529251098633,
      "epoch": 5.40420151721455,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002490110302402577,
      "loss": 4.9871,
      "mean_token_accuracy": 0.21525032073259354,
      "num_tokens": 120483897.0,
      "step": 69460
    },
    {
      "entropy": 5.966724061965943,
      "epoch": 5.404590546586268,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002489835003580367,
      "loss": 4.96,
      "mean_token_accuracy": 0.22134848684072495,
      "num_tokens": 120492654.0,
      "step": 69465
    },
    {
      "entropy": 5.964642429351807,
      "epoch": 5.404979575957984,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024895597087057483,
      "loss": 4.949,
      "mean_token_accuracy": 0.22301851361989974,
      "num_tokens": 120501056.0,
      "step": 69470
    },
    {
      "entropy": 5.965786504745483,
      "epoch": 5.405368605329702,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002489284417782895,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21809854507446289,
      "num_tokens": 120509767.0,
      "step": 69475
    },
    {
      "entropy": 5.978460693359375,
      "epoch": 5.40575763470142,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002489009130815986,
      "loss": 4.9502,
      "mean_token_accuracy": 0.2145475998520851,
      "num_tokens": 120518484.0,
      "step": 69480
    },
    {
      "entropy": 5.927890348434448,
      "epoch": 5.4061466640731375,
      "grad_norm": 1.375,
      "learning_rate": 0.0002488733847809197,
      "loss": 4.8867,
      "mean_token_accuracy": 0.22907895892858504,
      "num_tokens": 120527130.0,
      "step": 69485
    },
    {
      "entropy": 5.998595333099365,
      "epoch": 5.406535693444855,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002488458568766707,
      "loss": 4.9292,
      "mean_token_accuracy": 0.2131692111492157,
      "num_tokens": 120536460.0,
      "step": 69490
    },
    {
      "entropy": 5.9556139469146725,
      "epoch": 5.406924722816573,
      "grad_norm": 1.625,
      "learning_rate": 0.0002488183293692692,
      "loss": 5.0511,
      "mean_token_accuracy": 0.21563382148742677,
      "num_tokens": 120545396.0,
      "step": 69495
    },
    {
      "entropy": 5.937748908996582,
      "epoch": 5.407313752188291,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002487908022591329,
      "loss": 4.9563,
      "mean_token_accuracy": 0.21728212386369705,
      "num_tokens": 120553896.0,
      "step": 69500
    },
    {
      "entropy": 5.903484630584717,
      "epoch": 5.407702781560007,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024876327554667943,
      "loss": 4.9107,
      "mean_token_accuracy": 0.2139052554965019,
      "num_tokens": 120562209.0,
      "step": 69505
    },
    {
      "entropy": 5.991656160354614,
      "epoch": 5.408091810931725,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002487357492323265,
      "loss": 4.9193,
      "mean_token_accuracy": 0.21190870255231858,
      "num_tokens": 120572040.0,
      "step": 69510
    },
    {
      "entropy": 5.999327278137207,
      "epoch": 5.408480840303443,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002487082233164918,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20574814230203628,
      "num_tokens": 120580308.0,
      "step": 69515
    },
    {
      "entropy": 6.063082695007324,
      "epoch": 5.4088698696751605,
      "grad_norm": 1.5,
      "learning_rate": 0.0002486806977995929,
      "loss": 5.0084,
      "mean_token_accuracy": 0.21703935265541077,
      "num_tokens": 120590070.0,
      "step": 69520
    },
    {
      "entropy": 6.001849746704101,
      "epoch": 5.409258899046878,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002486531726820476,
      "loss": 4.9806,
      "mean_token_accuracy": 0.21526347249746322,
      "num_tokens": 120598922.0,
      "step": 69525
    },
    {
      "entropy": 5.914220142364502,
      "epoch": 5.409647928418596,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024862564796427337,
      "loss": 4.9054,
      "mean_token_accuracy": 0.22563911974430084,
      "num_tokens": 120607312.0,
      "step": 69530
    },
    {
      "entropy": 5.940074825286866,
      "epoch": 5.410036957790314,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000248598123646688,
      "loss": 4.9293,
      "mean_token_accuracy": 0.22088136225938798,
      "num_tokens": 120616214.0,
      "step": 69535
    },
    {
      "entropy": 5.9978776454925535,
      "epoch": 5.41042598716203,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002485705997297091,
      "loss": 5.026,
      "mean_token_accuracy": 0.21436664909124375,
      "num_tokens": 120624784.0,
      "step": 69540
    },
    {
      "entropy": 5.99267611503601,
      "epoch": 5.410815016533748,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024854307621375427,
      "loss": 4.9853,
      "mean_token_accuracy": 0.21453779935836792,
      "num_tokens": 120634052.0,
      "step": 69545
    },
    {
      "entropy": 5.988218355178833,
      "epoch": 5.411204045905466,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002485155530992411,
      "loss": 4.9352,
      "mean_token_accuracy": 0.21579424440860748,
      "num_tokens": 120642441.0,
      "step": 69550
    },
    {
      "entropy": 5.917684507369995,
      "epoch": 5.4115930752771835,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002484880303865873,
      "loss": 4.9083,
      "mean_token_accuracy": 0.22172731161117554,
      "num_tokens": 120651342.0,
      "step": 69555
    },
    {
      "entropy": 5.918315982818603,
      "epoch": 5.411982104648901,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002484605080762104,
      "loss": 4.9718,
      "mean_token_accuracy": 0.2228367179632187,
      "num_tokens": 120661044.0,
      "step": 69560
    },
    {
      "entropy": 5.953747081756592,
      "epoch": 5.412371134020619,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002484329861685281,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22112565189599992,
      "num_tokens": 120668743.0,
      "step": 69565
    },
    {
      "entropy": 6.006809997558594,
      "epoch": 5.412760163392337,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024840546466395796,
      "loss": 5.0119,
      "mean_token_accuracy": 0.21986104249954225,
      "num_tokens": 120677379.0,
      "step": 69570
    },
    {
      "entropy": 5.989054155349732,
      "epoch": 5.413149192764053,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024837794356291756,
      "loss": 4.9263,
      "mean_token_accuracy": 0.2183444619178772,
      "num_tokens": 120685941.0,
      "step": 69575
    },
    {
      "entropy": 6.001258516311646,
      "epoch": 5.413538222135771,
      "grad_norm": 1.5,
      "learning_rate": 0.0002483504228658245,
      "loss": 5.0097,
      "mean_token_accuracy": 0.21729765981435775,
      "num_tokens": 120695155.0,
      "step": 69580
    },
    {
      "entropy": 5.945470571517944,
      "epoch": 5.413927251507489,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002483229025730963,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22013757526874542,
      "num_tokens": 120703564.0,
      "step": 69585
    },
    {
      "entropy": 5.97585096359253,
      "epoch": 5.4143162808792065,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002482953826851506,
      "loss": 4.884,
      "mean_token_accuracy": 0.2252918228507042,
      "num_tokens": 120712435.0,
      "step": 69590
    },
    {
      "entropy": 5.908626747131348,
      "epoch": 5.414705310250924,
      "grad_norm": 1.484375,
      "learning_rate": 0.000248267863202405,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2259386360645294,
      "num_tokens": 120720801.0,
      "step": 69595
    },
    {
      "entropy": 5.954751825332641,
      "epoch": 5.415094339622642,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024824034412527704,
      "loss": 4.9593,
      "mean_token_accuracy": 0.2264085367321968,
      "num_tokens": 120729199.0,
      "step": 69600
    },
    {
      "entropy": 5.922840547561646,
      "epoch": 5.415483368994359,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002482128254541842,
      "loss": 4.8272,
      "mean_token_accuracy": 0.23023559153079987,
      "num_tokens": 120737241.0,
      "step": 69605
    },
    {
      "entropy": 5.9119775772094725,
      "epoch": 5.415872398366076,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002481853071895442,
      "loss": 4.9398,
      "mean_token_accuracy": 0.21839763224124908,
      "num_tokens": 120745704.0,
      "step": 69610
    },
    {
      "entropy": 5.927948713302612,
      "epoch": 5.416261427737794,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0002481577893317744,
      "loss": 4.8873,
      "mean_token_accuracy": 0.22726092338562012,
      "num_tokens": 120753904.0,
      "step": 69615
    },
    {
      "entropy": 5.94631175994873,
      "epoch": 5.416650457109512,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002481302718812924,
      "loss": 4.8734,
      "mean_token_accuracy": 0.21951136142015457,
      "num_tokens": 120761573.0,
      "step": 69620
    },
    {
      "entropy": 6.019516658782959,
      "epoch": 5.4170394864812295,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002481027548385158,
      "loss": 4.9926,
      "mean_token_accuracy": 0.21075040251016616,
      "num_tokens": 120770336.0,
      "step": 69625
    },
    {
      "entropy": 5.954289150238037,
      "epoch": 5.417428515852947,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024807523820386207,
      "loss": 5.0097,
      "mean_token_accuracy": 0.21802036464214325,
      "num_tokens": 120779007.0,
      "step": 69630
    },
    {
      "entropy": 5.896920680999756,
      "epoch": 5.417817545224665,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002480477219777488,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22808157801628112,
      "num_tokens": 120787109.0,
      "step": 69635
    },
    {
      "entropy": 5.847638940811157,
      "epoch": 5.418206574596382,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002480202061605933,
      "loss": 4.7877,
      "mean_token_accuracy": 0.23531697541475297,
      "num_tokens": 120795810.0,
      "step": 69640
    },
    {
      "entropy": 5.897540187835693,
      "epoch": 5.418595603968099,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002479926907528134,
      "loss": 4.8606,
      "mean_token_accuracy": 0.22465434223413466,
      "num_tokens": 120804972.0,
      "step": 69645
    },
    {
      "entropy": 5.947163057327271,
      "epoch": 5.418984633339817,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024796517575482633,
      "loss": 4.9606,
      "mean_token_accuracy": 0.21651918143033982,
      "num_tokens": 120813820.0,
      "step": 69650
    },
    {
      "entropy": 5.965563344955444,
      "epoch": 5.419373662711535,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002479376611670498,
      "loss": 4.9037,
      "mean_token_accuracy": 0.22504474669694902,
      "num_tokens": 120822298.0,
      "step": 69655
    },
    {
      "entropy": 5.956828832626343,
      "epoch": 5.4197626920832525,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024791014698990116,
      "loss": 4.8917,
      "mean_token_accuracy": 0.22093381434679032,
      "num_tokens": 120830656.0,
      "step": 69660
    },
    {
      "entropy": 5.986466932296753,
      "epoch": 5.42015172145497,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024788263322379783,
      "loss": 4.9501,
      "mean_token_accuracy": 0.2209106683731079,
      "num_tokens": 120838750.0,
      "step": 69665
    },
    {
      "entropy": 5.942884540557861,
      "epoch": 5.420540750826687,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00024785511986915746,
      "loss": 4.895,
      "mean_token_accuracy": 0.21785439550876617,
      "num_tokens": 120847695.0,
      "step": 69670
    },
    {
      "entropy": 6.045891618728637,
      "epoch": 5.420929780198405,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002478276069263974,
      "loss": 5.033,
      "mean_token_accuracy": 0.2093837782740593,
      "num_tokens": 120856288.0,
      "step": 69675
    },
    {
      "entropy": 6.046911001205444,
      "epoch": 5.421318809570122,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024780009439593514,
      "loss": 5.0761,
      "mean_token_accuracy": 0.21329471617937087,
      "num_tokens": 120864952.0,
      "step": 69680
    },
    {
      "entropy": 5.942602729797363,
      "epoch": 5.42170783894184,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002477725822781881,
      "loss": 4.954,
      "mean_token_accuracy": 0.21525086015462874,
      "num_tokens": 120873574.0,
      "step": 69685
    },
    {
      "entropy": 5.891190671920777,
      "epoch": 5.422096868313558,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002477450705735739,
      "loss": 4.8851,
      "mean_token_accuracy": 0.22408080846071243,
      "num_tokens": 120881710.0,
      "step": 69690
    },
    {
      "entropy": 5.90842981338501,
      "epoch": 5.4224858976852754,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024771755928250984,
      "loss": 4.8589,
      "mean_token_accuracy": 0.227132585644722,
      "num_tokens": 120889864.0,
      "step": 69695
    },
    {
      "entropy": 5.922195625305176,
      "epoch": 5.422874927056993,
      "grad_norm": 1.5,
      "learning_rate": 0.00024769004840541337,
      "loss": 4.9384,
      "mean_token_accuracy": 0.21617465764284133,
      "num_tokens": 120898997.0,
      "step": 69700
    },
    {
      "entropy": 5.928155040740966,
      "epoch": 5.42326395642871,
      "grad_norm": 1.5,
      "learning_rate": 0.000247662537942702,
      "loss": 4.9349,
      "mean_token_accuracy": 0.22389423698186875,
      "num_tokens": 120907897.0,
      "step": 69705
    },
    {
      "entropy": 5.91570520401001,
      "epoch": 5.423652985800428,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002476350278947931,
      "loss": 4.8688,
      "mean_token_accuracy": 0.22564842700958251,
      "num_tokens": 120916729.0,
      "step": 69710
    },
    {
      "entropy": 5.929300785064697,
      "epoch": 5.424042015172145,
      "grad_norm": 1.578125,
      "learning_rate": 0.000247607518262104,
      "loss": 4.9721,
      "mean_token_accuracy": 0.21279922127723694,
      "num_tokens": 120926012.0,
      "step": 69715
    },
    {
      "entropy": 5.92682204246521,
      "epoch": 5.424431044543863,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002475800090450523,
      "loss": 4.8934,
      "mean_token_accuracy": 0.22135115414857864,
      "num_tokens": 120934016.0,
      "step": 69720
    },
    {
      "entropy": 5.997908735275269,
      "epoch": 5.424820073915581,
      "grad_norm": 1.734375,
      "learning_rate": 0.0002475525002440553,
      "loss": 5.0349,
      "mean_token_accuracy": 0.21130410581827164,
      "num_tokens": 120942415.0,
      "step": 69725
    },
    {
      "entropy": 5.962086725234985,
      "epoch": 5.425209103287298,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024752499185953043,
      "loss": 4.962,
      "mean_token_accuracy": 0.22431014031171798,
      "num_tokens": 120950143.0,
      "step": 69730
    },
    {
      "entropy": 5.842546033859253,
      "epoch": 5.425598132659016,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002474974838918951,
      "loss": 4.9152,
      "mean_token_accuracy": 0.2200164258480072,
      "num_tokens": 120958457.0,
      "step": 69735
    },
    {
      "entropy": 5.9410973072052,
      "epoch": 5.425987162030733,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002474699763415667,
      "loss": 4.9045,
      "mean_token_accuracy": 0.2279996857047081,
      "num_tokens": 120966520.0,
      "step": 69740
    },
    {
      "entropy": 5.954097175598145,
      "epoch": 5.426376191402451,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00024744246920896243,
      "loss": 4.8792,
      "mean_token_accuracy": 0.22382989078760146,
      "num_tokens": 120974805.0,
      "step": 69745
    },
    {
      "entropy": 6.013351011276245,
      "epoch": 5.426765220774168,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002474149624944999,
      "loss": 4.9675,
      "mean_token_accuracy": 0.21421680450439454,
      "num_tokens": 120983455.0,
      "step": 69750
    },
    {
      "entropy": 5.959715700149536,
      "epoch": 5.427154250145886,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024738745619859643,
      "loss": 4.9702,
      "mean_token_accuracy": 0.2144887387752533,
      "num_tokens": 120993310.0,
      "step": 69755
    },
    {
      "entropy": 5.972245597839356,
      "epoch": 5.427543279517604,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00024735995032166934,
      "loss": 4.9704,
      "mean_token_accuracy": 0.21893640160560607,
      "num_tokens": 121002002.0,
      "step": 69760
    },
    {
      "entropy": 5.994754028320313,
      "epoch": 5.427932308889321,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002473324448641361,
      "loss": 4.9916,
      "mean_token_accuracy": 0.21555853486061097,
      "num_tokens": 121010440.0,
      "step": 69765
    },
    {
      "entropy": 5.980138444900513,
      "epoch": 5.428321338261039,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024730493982641396,
      "loss": 4.8239,
      "mean_token_accuracy": 0.22795003205537795,
      "num_tokens": 121018231.0,
      "step": 69770
    },
    {
      "entropy": 5.995293712615966,
      "epoch": 5.428710367632756,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002472774352089202,
      "loss": 4.9731,
      "mean_token_accuracy": 0.21923116445541382,
      "num_tokens": 121027193.0,
      "step": 69775
    },
    {
      "entropy": 5.988741397857666,
      "epoch": 5.429099397004474,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002472499310120723,
      "loss": 4.9603,
      "mean_token_accuracy": 0.2227725550532341,
      "num_tokens": 121035472.0,
      "step": 69780
    },
    {
      "entropy": 5.943289375305175,
      "epoch": 5.429488426376191,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002472224272362875,
      "loss": 4.9276,
      "mean_token_accuracy": 0.21586958020925523,
      "num_tokens": 121044167.0,
      "step": 69785
    },
    {
      "entropy": 5.93031530380249,
      "epoch": 5.429877455747909,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024719492388198313,
      "loss": 4.8993,
      "mean_token_accuracy": 0.2209222748875618,
      "num_tokens": 121053444.0,
      "step": 69790
    },
    {
      "entropy": 5.9733682632446286,
      "epoch": 5.430266485119627,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002471674209495765,
      "loss": 4.8984,
      "mean_token_accuracy": 0.22360751628875733,
      "num_tokens": 121062733.0,
      "step": 69795
    },
    {
      "entropy": 5.999290561676025,
      "epoch": 5.430655514491344,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024713991843948506,
      "loss": 4.9702,
      "mean_token_accuracy": 0.21681024730205536,
      "num_tokens": 121071187.0,
      "step": 69800
    },
    {
      "entropy": 5.961117219924927,
      "epoch": 5.431044543863061,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024711241635212597,
      "loss": 4.9697,
      "mean_token_accuracy": 0.2162310317158699,
      "num_tokens": 121079926.0,
      "step": 69805
    },
    {
      "entropy": 5.946539878845215,
      "epoch": 5.431433573234779,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024708491468791656,
      "loss": 4.9375,
      "mean_token_accuracy": 0.22285141348838805,
      "num_tokens": 121089275.0,
      "step": 69810
    },
    {
      "entropy": 5.9817249298095705,
      "epoch": 5.431822602606497,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002470574134472741,
      "loss": 4.9,
      "mean_token_accuracy": 0.22950120121240616,
      "num_tokens": 121098020.0,
      "step": 69815
    },
    {
      "entropy": 6.008489036560059,
      "epoch": 5.432211631978214,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024702991263061595,
      "loss": 4.9999,
      "mean_token_accuracy": 0.21528828293085098,
      "num_tokens": 121106124.0,
      "step": 69820
    },
    {
      "entropy": 5.9804222106933596,
      "epoch": 5.432600661349932,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024700241223835935,
      "loss": 5.008,
      "mean_token_accuracy": 0.21622665226459503,
      "num_tokens": 121114958.0,
      "step": 69825
    },
    {
      "entropy": 6.023210859298706,
      "epoch": 5.43298969072165,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024697491227092163,
      "loss": 4.963,
      "mean_token_accuracy": 0.22363444715738295,
      "num_tokens": 121123463.0,
      "step": 69830
    },
    {
      "entropy": 6.025192832946777,
      "epoch": 5.433378720093367,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024694741272872,
      "loss": 5.049,
      "mean_token_accuracy": 0.21160695254802703,
      "num_tokens": 121132260.0,
      "step": 69835
    },
    {
      "entropy": 6.015255975723266,
      "epoch": 5.433767749465084,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024691991361217166,
      "loss": 4.9184,
      "mean_token_accuracy": 0.22024457454681395,
      "num_tokens": 121140409.0,
      "step": 69840
    },
    {
      "entropy": 5.899049711227417,
      "epoch": 5.434156778836802,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024689241492169403,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21644993722438813,
      "num_tokens": 121148834.0,
      "step": 69845
    },
    {
      "entropy": 5.970679426193238,
      "epoch": 5.43454580820852,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024686491665770424,
      "loss": 5.029,
      "mean_token_accuracy": 0.21138272881507875,
      "num_tokens": 121157953.0,
      "step": 69850
    },
    {
      "entropy": 6.019716548919678,
      "epoch": 5.434934837580237,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00024683741882061954,
      "loss": 4.9076,
      "mean_token_accuracy": 0.22652398198843002,
      "num_tokens": 121165774.0,
      "step": 69855
    },
    {
      "entropy": 5.978470659255981,
      "epoch": 5.435323866951955,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002468099214108572,
      "loss": 4.9415,
      "mean_token_accuracy": 0.22101059108972548,
      "num_tokens": 121174549.0,
      "step": 69860
    },
    {
      "entropy": 6.026112461090088,
      "epoch": 5.435712896323673,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024678242442883444,
      "loss": 5.0216,
      "mean_token_accuracy": 0.22077436298131942,
      "num_tokens": 121183588.0,
      "step": 69865
    },
    {
      "entropy": 5.961284065246582,
      "epoch": 5.43610192569539,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024675492787496844,
      "loss": 4.9946,
      "mean_token_accuracy": 0.2117645487189293,
      "num_tokens": 121192349.0,
      "step": 69870
    },
    {
      "entropy": 6.029409646987915,
      "epoch": 5.436490955067107,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00024672743174967645,
      "loss": 4.9313,
      "mean_token_accuracy": 0.22920598983764648,
      "num_tokens": 121200667.0,
      "step": 69875
    },
    {
      "entropy": 5.872901821136475,
      "epoch": 5.436879984438825,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002466999360533757,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22792194038629532,
      "num_tokens": 121209836.0,
      "step": 69880
    },
    {
      "entropy": 5.978393840789795,
      "epoch": 5.437269013810543,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002466724407864833,
      "loss": 4.9696,
      "mean_token_accuracy": 0.21688223034143447,
      "num_tokens": 121217535.0,
      "step": 69885
    },
    {
      "entropy": 6.003305578231812,
      "epoch": 5.43765804318226,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024664494594941664,
      "loss": 4.9092,
      "mean_token_accuracy": 0.22898366004228593,
      "num_tokens": 121226159.0,
      "step": 69890
    },
    {
      "entropy": 6.033805322647095,
      "epoch": 5.438047072553978,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024661745154259273,
      "loss": 5.0645,
      "mean_token_accuracy": 0.2165144294500351,
      "num_tokens": 121234717.0,
      "step": 69895
    },
    {
      "entropy": 5.991330575942993,
      "epoch": 5.438436101925696,
      "grad_norm": 1.734375,
      "learning_rate": 0.0002465899575664289,
      "loss": 4.9073,
      "mean_token_accuracy": 0.21948652863502502,
      "num_tokens": 121243086.0,
      "step": 69900
    },
    {
      "entropy": 5.989740467071533,
      "epoch": 5.438825131297413,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024656246402134216,
      "loss": 5.0348,
      "mean_token_accuracy": 0.21474512964487075,
      "num_tokens": 121251350.0,
      "step": 69905
    },
    {
      "entropy": 5.969817686080932,
      "epoch": 5.43921416066913,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002465349709077498,
      "loss": 4.9629,
      "mean_token_accuracy": 0.21102358102798463,
      "num_tokens": 121260455.0,
      "step": 69910
    },
    {
      "entropy": 5.951726770401001,
      "epoch": 5.439603190040848,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002465074782260689,
      "loss": 4.8888,
      "mean_token_accuracy": 0.22276460528373718,
      "num_tokens": 121269166.0,
      "step": 69915
    },
    {
      "entropy": 5.921156406402588,
      "epoch": 5.439992219412566,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024647998597671674,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23536086082458496,
      "num_tokens": 121277415.0,
      "step": 69920
    },
    {
      "entropy": 5.969387531280518,
      "epoch": 5.440381248784283,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002464524941601103,
      "loss": 4.9355,
      "mean_token_accuracy": 0.2188545823097229,
      "num_tokens": 121285961.0,
      "step": 69925
    },
    {
      "entropy": 5.963651132583618,
      "epoch": 5.440770278156001,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024642500277666696,
      "loss": 4.9366,
      "mean_token_accuracy": 0.22320512533187867,
      "num_tokens": 121294178.0,
      "step": 69930
    },
    {
      "entropy": 5.929413843154907,
      "epoch": 5.441159307527719,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024639751182680363,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21915046721696854,
      "num_tokens": 121303146.0,
      "step": 69935
    },
    {
      "entropy": 5.952569246292114,
      "epoch": 5.4415483368994355,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002463700213109376,
      "loss": 4.9174,
      "mean_token_accuracy": 0.21696809381246568,
      "num_tokens": 121312485.0,
      "step": 69940
    },
    {
      "entropy": 5.937078619003296,
      "epoch": 5.441937366271153,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024634253122948584,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2242192506790161,
      "num_tokens": 121320882.0,
      "step": 69945
    },
    {
      "entropy": 5.900988626480102,
      "epoch": 5.442326395642871,
      "grad_norm": 1.5,
      "learning_rate": 0.00024631504158286556,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21521260142326354,
      "num_tokens": 121329289.0,
      "step": 69950
    },
    {
      "entropy": 5.905853605270385,
      "epoch": 5.442715425014589,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024628755237149385,
      "loss": 4.8591,
      "mean_token_accuracy": 0.21939316689968108,
      "num_tokens": 121337448.0,
      "step": 69955
    },
    {
      "entropy": 5.970423603057862,
      "epoch": 5.443104454386306,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00024626006359578797,
      "loss": 4.9904,
      "mean_token_accuracy": 0.21098355650901796,
      "num_tokens": 121346671.0,
      "step": 69960
    },
    {
      "entropy": 5.925609922409057,
      "epoch": 5.443493483758024,
      "grad_norm": 1.578125,
      "learning_rate": 0.00024623257525616484,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21551347374916077,
      "num_tokens": 121355396.0,
      "step": 69965
    },
    {
      "entropy": 6.0288787364959715,
      "epoch": 5.443882513129742,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024620508735304146,
      "loss": 5.0395,
      "mean_token_accuracy": 0.21164844185113907,
      "num_tokens": 121363574.0,
      "step": 69970
    },
    {
      "entropy": 5.947305059432983,
      "epoch": 5.4442715425014585,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002461775998868352,
      "loss": 4.9259,
      "mean_token_accuracy": 0.2192602962255478,
      "num_tokens": 121372139.0,
      "step": 69975
    },
    {
      "entropy": 5.978057336807251,
      "epoch": 5.444660571873176,
      "grad_norm": 1.609375,
      "learning_rate": 0.00024615011285796293,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21770855635404587,
      "num_tokens": 121380975.0,
      "step": 69980
    },
    {
      "entropy": 5.966154384613037,
      "epoch": 5.445049601244894,
      "grad_norm": 1.5,
      "learning_rate": 0.00024612262626684176,
      "loss": 4.981,
      "mean_token_accuracy": 0.21439764499664307,
      "num_tokens": 121390340.0,
      "step": 69985
    },
    {
      "entropy": 5.938549566268921,
      "epoch": 5.445438630616612,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002460951401138888,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22770666778087617,
      "num_tokens": 121398807.0,
      "step": 69990
    },
    {
      "entropy": 5.986173057556153,
      "epoch": 5.445827659988329,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00024606765439952105,
      "loss": 4.9944,
      "mean_token_accuracy": 0.21809961199760436,
      "num_tokens": 121407703.0,
      "step": 69995
    },
    {
      "entropy": 5.935518741607666,
      "epoch": 5.446216689360047,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002460401691241556,
      "loss": 4.9828,
      "mean_token_accuracy": 0.22147472351789474,
      "num_tokens": 121416385.0,
      "step": 70000
    },
    {
      "entropy": 5.888468217849732,
      "epoch": 5.446605718731764,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00024601268428820947,
      "loss": 4.895,
      "mean_token_accuracy": 0.21349750757217406,
      "num_tokens": 121424651.0,
      "step": 70005
    },
    {
      "entropy": 5.931980323791504,
      "epoch": 5.4469947481034815,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024598519989209983,
      "loss": 4.9841,
      "mean_token_accuracy": 0.21779898405075074,
      "num_tokens": 121434248.0,
      "step": 70010
    },
    {
      "entropy": 5.920939302444458,
      "epoch": 5.447383777475199,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002459577159362434,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22563773691654204,
      "num_tokens": 121442762.0,
      "step": 70015
    },
    {
      "entropy": 6.018200492858886,
      "epoch": 5.447772806846917,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002459302324210575,
      "loss": 5.0529,
      "mean_token_accuracy": 0.21488068550825118,
      "num_tokens": 121451622.0,
      "step": 70020
    },
    {
      "entropy": 6.014619255065918,
      "epoch": 5.448161836218635,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002459027493469591,
      "loss": 4.9653,
      "mean_token_accuracy": 0.22404997646808625,
      "num_tokens": 121460308.0,
      "step": 70025
    },
    {
      "entropy": 5.974753379821777,
      "epoch": 5.448550865590352,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00024587526671436517,
      "loss": 4.9877,
      "mean_token_accuracy": 0.22001627534627916,
      "num_tokens": 121469429.0,
      "step": 70030
    },
    {
      "entropy": 5.900829553604126,
      "epoch": 5.44893989496207,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024584778452369264,
      "loss": 4.8591,
      "mean_token_accuracy": 0.21657415181398393,
      "num_tokens": 121478240.0,
      "step": 70035
    },
    {
      "entropy": 5.958044528961182,
      "epoch": 5.449328924333787,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024582030277535866,
      "loss": 4.9576,
      "mean_token_accuracy": 0.22291708290576934,
      "num_tokens": 121487480.0,
      "step": 70040
    },
    {
      "entropy": 5.991892719268799,
      "epoch": 5.4497179537055045,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024579282146978016,
      "loss": 5.0014,
      "mean_token_accuracy": 0.21514321714639664,
      "num_tokens": 121495653.0,
      "step": 70045
    },
    {
      "entropy": 5.984756565093994,
      "epoch": 5.450106983077222,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000245765340607374,
      "loss": 4.9785,
      "mean_token_accuracy": 0.2211349055171013,
      "num_tokens": 121503924.0,
      "step": 70050
    },
    {
      "entropy": 5.992353343963623,
      "epoch": 5.45049601244894,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002457378601885573,
      "loss": 5.0136,
      "mean_token_accuracy": 0.21979162693023682,
      "num_tokens": 121513062.0,
      "step": 70055
    },
    {
      "entropy": 5.950098848342895,
      "epoch": 5.450885041820658,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024571038021374705,
      "loss": 4.9498,
      "mean_token_accuracy": 0.20833095014095307,
      "num_tokens": 121522543.0,
      "step": 70060
    },
    {
      "entropy": 6.040549898147583,
      "epoch": 5.451274071192375,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00024568290068336014,
      "loss": 5.0694,
      "mean_token_accuracy": 0.20912840068340302,
      "num_tokens": 121531081.0,
      "step": 70065
    },
    {
      "entropy": 5.975691270828247,
      "epoch": 5.451663100564093,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002456554215978135,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21624995321035384,
      "num_tokens": 121540034.0,
      "step": 70070
    },
    {
      "entropy": 5.940537929534912,
      "epoch": 5.45205212993581,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002456279429575242,
      "loss": 4.964,
      "mean_token_accuracy": 0.21798825562000274,
      "num_tokens": 121549006.0,
      "step": 70075
    },
    {
      "entropy": 5.919402647018432,
      "epoch": 5.4524411593075275,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024560046476290905,
      "loss": 4.9021,
      "mean_token_accuracy": 0.21927562057971955,
      "num_tokens": 121557178.0,
      "step": 70080
    },
    {
      "entropy": 5.960303401947021,
      "epoch": 5.452830188679245,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00024557298701438504,
      "loss": 4.9218,
      "mean_token_accuracy": 0.2219780296087265,
      "num_tokens": 121565594.0,
      "step": 70085
    },
    {
      "entropy": 5.963353157043457,
      "epoch": 5.453219218050963,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002455455097123691,
      "loss": 4.9463,
      "mean_token_accuracy": 0.2226736918091774,
      "num_tokens": 121575115.0,
      "step": 70090
    },
    {
      "entropy": 6.024876451492309,
      "epoch": 5.453608247422681,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002455180328572783,
      "loss": 5.0851,
      "mean_token_accuracy": 0.21108345240354537,
      "num_tokens": 121584674.0,
      "step": 70095
    },
    {
      "entropy": 6.037327623367309,
      "epoch": 5.453997276794398,
      "grad_norm": 1.5,
      "learning_rate": 0.00024549055644952926,
      "loss": 4.9982,
      "mean_token_accuracy": 0.2171592190861702,
      "num_tokens": 121593647.0,
      "step": 70100
    },
    {
      "entropy": 6.053216409683228,
      "epoch": 5.454386306166116,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002454630804895392,
      "loss": 4.9197,
      "mean_token_accuracy": 0.22766707539558412,
      "num_tokens": 121603187.0,
      "step": 70105
    },
    {
      "entropy": 6.012965440750122,
      "epoch": 5.454775335537833,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002454356049777248,
      "loss": 5.0403,
      "mean_token_accuracy": 0.2095397263765335,
      "num_tokens": 121612479.0,
      "step": 70110
    },
    {
      "entropy": 5.963444232940674,
      "epoch": 5.4551643649095505,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024540812991450306,
      "loss": 4.9744,
      "mean_token_accuracy": 0.21937051862478257,
      "num_tokens": 121620859.0,
      "step": 70115
    },
    {
      "entropy": 5.983778476715088,
      "epoch": 5.455553394281268,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024538065530029093,
      "loss": 4.9885,
      "mean_token_accuracy": 0.22022458761930466,
      "num_tokens": 121629265.0,
      "step": 70120
    },
    {
      "entropy": 5.944018745422364,
      "epoch": 5.455942423652986,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002453531811355051,
      "loss": 4.942,
      "mean_token_accuracy": 0.22083521336317063,
      "num_tokens": 121638246.0,
      "step": 70125
    },
    {
      "entropy": 5.9625472068786625,
      "epoch": 5.456331453024704,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024532570742056257,
      "loss": 4.9415,
      "mean_token_accuracy": 0.22107749283313752,
      "num_tokens": 121646596.0,
      "step": 70130
    },
    {
      "entropy": 5.961402082443238,
      "epoch": 5.456720482396421,
      "grad_norm": 1.625,
      "learning_rate": 0.00024529823415588025,
      "loss": 4.9056,
      "mean_token_accuracy": 0.2189960226416588,
      "num_tokens": 121654935.0,
      "step": 70135
    },
    {
      "entropy": 5.988764381408691,
      "epoch": 5.457109511768138,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002452707613418749,
      "loss": 5.0471,
      "mean_token_accuracy": 0.2054786279797554,
      "num_tokens": 121663294.0,
      "step": 70140
    },
    {
      "entropy": 5.901919364929199,
      "epoch": 5.457498541139856,
      "grad_norm": 1.6875,
      "learning_rate": 0.00024524328897896345,
      "loss": 4.9023,
      "mean_token_accuracy": 0.22331960946321489,
      "num_tokens": 121671646.0,
      "step": 70145
    },
    {
      "entropy": 5.943007135391236,
      "epoch": 5.4578875705115735,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024521581706756266,
      "loss": 4.9567,
      "mean_token_accuracy": 0.21306896954774857,
      "num_tokens": 121681237.0,
      "step": 70150
    },
    {
      "entropy": 6.006966686248779,
      "epoch": 5.458276599883291,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024518834560808954,
      "loss": 4.9327,
      "mean_token_accuracy": 0.22010865360498427,
      "num_tokens": 121690295.0,
      "step": 70155
    },
    {
      "entropy": 6.011630582809448,
      "epoch": 5.458665629255009,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024516087460096074,
      "loss": 4.9376,
      "mean_token_accuracy": 0.22252710461616515,
      "num_tokens": 121698880.0,
      "step": 70160
    },
    {
      "entropy": 5.994739770889282,
      "epoch": 5.459054658626727,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002451334040465933,
      "loss": 5.0118,
      "mean_token_accuracy": 0.21929710656404494,
      "num_tokens": 121707068.0,
      "step": 70165
    },
    {
      "entropy": 5.940282726287842,
      "epoch": 5.459443687998444,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024510593394540383,
      "loss": 4.9002,
      "mean_token_accuracy": 0.22216132581233977,
      "num_tokens": 121715536.0,
      "step": 70170
    },
    {
      "entropy": 5.97674036026001,
      "epoch": 5.459832717370161,
      "grad_norm": 1.765625,
      "learning_rate": 0.0002450784642978092,
      "loss": 4.9839,
      "mean_token_accuracy": 0.21331264674663544,
      "num_tokens": 121723904.0,
      "step": 70175
    },
    {
      "entropy": 5.92652530670166,
      "epoch": 5.460221746741879,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002450509951042263,
      "loss": 4.9454,
      "mean_token_accuracy": 0.2235196575522423,
      "num_tokens": 121732396.0,
      "step": 70180
    },
    {
      "entropy": 5.9980486869812015,
      "epoch": 5.4606107761135965,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024502352636507184,
      "loss": 5.0436,
      "mean_token_accuracy": 0.21017932742834092,
      "num_tokens": 121741913.0,
      "step": 70185
    },
    {
      "entropy": 5.959969902038575,
      "epoch": 5.460999805485314,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024499605808076273,
      "loss": 4.9384,
      "mean_token_accuracy": 0.2214701235294342,
      "num_tokens": 121750649.0,
      "step": 70190
    },
    {
      "entropy": 5.908649682998657,
      "epoch": 5.461388834857032,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024496859025171564,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21795813590288163,
      "num_tokens": 121759296.0,
      "step": 70195
    },
    {
      "entropy": 5.991874265670776,
      "epoch": 5.46177786422875,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002449411228783474,
      "loss": 4.8579,
      "mean_token_accuracy": 0.22521924078464509,
      "num_tokens": 121767593.0,
      "step": 70200
    },
    {
      "entropy": 5.950638628005981,
      "epoch": 5.462166893600466,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00024491365596107477,
      "loss": 4.8902,
      "mean_token_accuracy": 0.2189406245946884,
      "num_tokens": 121776473.0,
      "step": 70205
    },
    {
      "entropy": 5.947278928756714,
      "epoch": 5.462555922972184,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024488618950031457,
      "loss": 4.9638,
      "mean_token_accuracy": 0.21975332349538804,
      "num_tokens": 121784648.0,
      "step": 70210
    },
    {
      "entropy": 5.908992004394531,
      "epoch": 5.462944952343902,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00024485872349648335,
      "loss": 4.9032,
      "mean_token_accuracy": 0.22434411495923995,
      "num_tokens": 121793650.0,
      "step": 70215
    },
    {
      "entropy": 5.929314088821411,
      "epoch": 5.4633339817156195,
      "grad_norm": 1.5,
      "learning_rate": 0.00024483125794999815,
      "loss": 4.911,
      "mean_token_accuracy": 0.22596397548913955,
      "num_tokens": 121803072.0,
      "step": 70220
    },
    {
      "entropy": 5.982689476013183,
      "epoch": 5.463723011087337,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002448037928612756,
      "loss": 4.8846,
      "mean_token_accuracy": 0.22459494173526764,
      "num_tokens": 121811394.0,
      "step": 70225
    },
    {
      "entropy": 5.9788548946380615,
      "epoch": 5.464112040459055,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002447763282307324,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21643175482749938,
      "num_tokens": 121820504.0,
      "step": 70230
    },
    {
      "entropy": 6.007728958129883,
      "epoch": 5.4645010698307726,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002447488640587854,
      "loss": 4.9672,
      "mean_token_accuracy": 0.22175848931074144,
      "num_tokens": 121829199.0,
      "step": 70235
    },
    {
      "entropy": 5.934974575042725,
      "epoch": 5.46489009920249,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024472140034585113,
      "loss": 4.872,
      "mean_token_accuracy": 0.22756956666707992,
      "num_tokens": 121837454.0,
      "step": 70240
    },
    {
      "entropy": 5.915169620513916,
      "epoch": 5.465279128574207,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00024469393709234647,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22027171701192855,
      "num_tokens": 121845326.0,
      "step": 70245
    },
    {
      "entropy": 5.914072799682617,
      "epoch": 5.465668157945925,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024466647429868804,
      "loss": 4.9518,
      "mean_token_accuracy": 0.22001068890094758,
      "num_tokens": 121854649.0,
      "step": 70250
    },
    {
      "entropy": 5.981341981887818,
      "epoch": 5.4660571873176425,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002446390119652926,
      "loss": 5.0305,
      "mean_token_accuracy": 0.2188127100467682,
      "num_tokens": 121863535.0,
      "step": 70255
    },
    {
      "entropy": 6.028214836120606,
      "epoch": 5.46644621668936,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024461155009257684,
      "loss": 4.9386,
      "mean_token_accuracy": 0.22699596136808395,
      "num_tokens": 121872256.0,
      "step": 70260
    },
    {
      "entropy": 5.877805137634278,
      "epoch": 5.466835246061078,
      "grad_norm": 1.375,
      "learning_rate": 0.00024458408868095744,
      "loss": 4.848,
      "mean_token_accuracy": 0.2158252105116844,
      "num_tokens": 121881075.0,
      "step": 70265
    },
    {
      "entropy": 5.975247430801391,
      "epoch": 5.4672242754327955,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024455662773085107,
      "loss": 5.0396,
      "mean_token_accuracy": 0.21376084089279174,
      "num_tokens": 121890051.0,
      "step": 70270
    },
    {
      "entropy": 5.973037958145142,
      "epoch": 5.467613304804512,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002445291672426744,
      "loss": 4.9329,
      "mean_token_accuracy": 0.2202228859066963,
      "num_tokens": 121898638.0,
      "step": 70275
    },
    {
      "entropy": 5.914977979660034,
      "epoch": 5.46800233417623,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024450170721684416,
      "loss": 4.8458,
      "mean_token_accuracy": 0.22978811711072922,
      "num_tokens": 121907422.0,
      "step": 70280
    },
    {
      "entropy": 6.048287963867187,
      "epoch": 5.468391363547948,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002444742476537769,
      "loss": 5.0081,
      "mean_token_accuracy": 0.21591449826955794,
      "num_tokens": 121916623.0,
      "step": 70285
    },
    {
      "entropy": 5.932656908035279,
      "epoch": 5.4687803929196654,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024444678855388934,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21673074662685393,
      "num_tokens": 121924907.0,
      "step": 70290
    },
    {
      "entropy": 5.9612071990966795,
      "epoch": 5.469169422291383,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024441932991759824,
      "loss": 4.911,
      "mean_token_accuracy": 0.21317079067230224,
      "num_tokens": 121933224.0,
      "step": 70295
    },
    {
      "entropy": 6.016350603103637,
      "epoch": 5.469558451663101,
      "grad_norm": 1.640625,
      "learning_rate": 0.0002443918717453201,
      "loss": 5.0505,
      "mean_token_accuracy": 0.21678704172372817,
      "num_tokens": 121941249.0,
      "step": 70300
    },
    {
      "entropy": 6.012908935546875,
      "epoch": 5.4699474810348185,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002443644140374715,
      "loss": 5.0635,
      "mean_token_accuracy": 0.204950512945652,
      "num_tokens": 121950427.0,
      "step": 70305
    },
    {
      "entropy": 6.003415203094482,
      "epoch": 5.470336510406535,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00024433695679446934,
      "loss": 4.9789,
      "mean_token_accuracy": 0.2115243375301361,
      "num_tokens": 121959471.0,
      "step": 70310
    },
    {
      "entropy": 5.958139085769654,
      "epoch": 5.470725539778253,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002443095000167299,
      "loss": 4.9483,
      "mean_token_accuracy": 0.22107537537813188,
      "num_tokens": 121968910.0,
      "step": 70315
    },
    {
      "entropy": 6.004298448562622,
      "epoch": 5.471114569149971,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00024428204370467,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21145312786102294,
      "num_tokens": 121976904.0,
      "step": 70320
    },
    {
      "entropy": 5.933706665039063,
      "epoch": 5.471503598521688,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024425458785870617,
      "loss": 4.8372,
      "mean_token_accuracy": 0.2207672953605652,
      "num_tokens": 121985716.0,
      "step": 70325
    },
    {
      "entropy": 5.926695013046265,
      "epoch": 5.471892627893406,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024422713247925507,
      "loss": 4.946,
      "mean_token_accuracy": 0.21529849916696547,
      "num_tokens": 121994047.0,
      "step": 70330
    },
    {
      "entropy": 5.930426406860351,
      "epoch": 5.472281657265124,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002441996775667332,
      "loss": 4.9246,
      "mean_token_accuracy": 0.22008053511381148,
      "num_tokens": 122003249.0,
      "step": 70335
    },
    {
      "entropy": 5.983847618103027,
      "epoch": 5.472670686636841,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002441722231215572,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2199401617050171,
      "num_tokens": 122011144.0,
      "step": 70340
    },
    {
      "entropy": 6.031998205184936,
      "epoch": 5.473059716008558,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024414476914414373,
      "loss": 5.017,
      "mean_token_accuracy": 0.21338811963796617,
      "num_tokens": 122020389.0,
      "step": 70345
    },
    {
      "entropy": 5.963103151321411,
      "epoch": 5.473448745380276,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00024411731563490918,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21821746081113816,
      "num_tokens": 122028348.0,
      "step": 70350
    },
    {
      "entropy": 6.017904615402221,
      "epoch": 5.473837774751994,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00024408986259427025,
      "loss": 4.96,
      "mean_token_accuracy": 0.21725095063447952,
      "num_tokens": 122036638.0,
      "step": 70355
    },
    {
      "entropy": 6.047570848464966,
      "epoch": 5.474226804123711,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024406241002264346,
      "loss": 5.0338,
      "mean_token_accuracy": 0.21034547537565232,
      "num_tokens": 122045311.0,
      "step": 70360
    },
    {
      "entropy": 5.907257986068726,
      "epoch": 5.474615833495429,
      "grad_norm": 1.609375,
      "learning_rate": 0.00024403495792044532,
      "loss": 4.9158,
      "mean_token_accuracy": 0.22249471843242646,
      "num_tokens": 122054748.0,
      "step": 70365
    },
    {
      "entropy": 5.8872373580932615,
      "epoch": 5.475004862867147,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024400750628809247,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22901204377412795,
      "num_tokens": 122063548.0,
      "step": 70370
    },
    {
      "entropy": 5.932893466949463,
      "epoch": 5.475393892238864,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024398005512600134,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2194409966468811,
      "num_tokens": 122071907.0,
      "step": 70375
    },
    {
      "entropy": 5.89904088973999,
      "epoch": 5.475782921610581,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024395260443458856,
      "loss": 4.8418,
      "mean_token_accuracy": 0.23071974962949754,
      "num_tokens": 122079930.0,
      "step": 70380
    },
    {
      "entropy": 5.95824089050293,
      "epoch": 5.476171950982299,
      "grad_norm": 1.5,
      "learning_rate": 0.00024392515421427054,
      "loss": 4.932,
      "mean_token_accuracy": 0.21748286187648774,
      "num_tokens": 122088540.0,
      "step": 70385
    },
    {
      "entropy": 5.96919436454773,
      "epoch": 5.476560980354017,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024389770446546388,
      "loss": 4.9642,
      "mean_token_accuracy": 0.21824264526367188,
      "num_tokens": 122097799.0,
      "step": 70390
    },
    {
      "entropy": 5.999470520019531,
      "epoch": 5.476950009725734,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024387025518858503,
      "loss": 4.9607,
      "mean_token_accuracy": 0.21786169558763505,
      "num_tokens": 122106761.0,
      "step": 70395
    },
    {
      "entropy": 5.982647180557251,
      "epoch": 5.477339039097452,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002438428063840505,
      "loss": 5.0274,
      "mean_token_accuracy": 0.21347213685512542,
      "num_tokens": 122115585.0,
      "step": 70400
    },
    {
      "entropy": 5.890702772140503,
      "epoch": 5.47772806846917,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024381535805227677,
      "loss": 4.899,
      "mean_token_accuracy": 0.2185283601284027,
      "num_tokens": 122123701.0,
      "step": 70405
    },
    {
      "entropy": 5.995426845550537,
      "epoch": 5.478117097840887,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024378791019368036,
      "loss": 4.9952,
      "mean_token_accuracy": 0.2249425619840622,
      "num_tokens": 122132707.0,
      "step": 70410
    },
    {
      "entropy": 5.9993959903717045,
      "epoch": 5.478506127212604,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024376046280867765,
      "loss": 4.9941,
      "mean_token_accuracy": 0.21296221017837524,
      "num_tokens": 122140611.0,
      "step": 70415
    },
    {
      "entropy": 5.9595160484313965,
      "epoch": 5.478895156584322,
      "grad_norm": 1.578125,
      "learning_rate": 0.00024373301589768531,
      "loss": 4.907,
      "mean_token_accuracy": 0.22458894550800323,
      "num_tokens": 122149453.0,
      "step": 70420
    },
    {
      "entropy": 5.8908594131469725,
      "epoch": 5.47928418595604,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002437055694611197,
      "loss": 4.866,
      "mean_token_accuracy": 0.2256157413125038,
      "num_tokens": 122158177.0,
      "step": 70425
    },
    {
      "entropy": 5.964514875411988,
      "epoch": 5.479673215327757,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024367812349939723,
      "loss": 4.9467,
      "mean_token_accuracy": 0.21685306429862977,
      "num_tokens": 122166860.0,
      "step": 70430
    },
    {
      "entropy": 5.96753339767456,
      "epoch": 5.480062244699475,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00024365067801293438,
      "loss": 4.9799,
      "mean_token_accuracy": 0.21732509583234788,
      "num_tokens": 122175378.0,
      "step": 70435
    },
    {
      "entropy": 5.936414766311645,
      "epoch": 5.480451274071193,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00024362323300214755,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22556928396224976,
      "num_tokens": 122184348.0,
      "step": 70440
    },
    {
      "entropy": 6.016794490814209,
      "epoch": 5.48084030344291,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002435957884674533,
      "loss": 4.9997,
      "mean_token_accuracy": 0.21757253557443618,
      "num_tokens": 122192861.0,
      "step": 70445
    },
    {
      "entropy": 5.896386766433716,
      "epoch": 5.481229332814627,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002435683444092679,
      "loss": 4.8296,
      "mean_token_accuracy": 0.22956524640321732,
      "num_tokens": 122202076.0,
      "step": 70450
    },
    {
      "entropy": 5.9378471851348875,
      "epoch": 5.481618362186345,
      "grad_norm": 1.375,
      "learning_rate": 0.0002435409008280079,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21909688860177995,
      "num_tokens": 122211482.0,
      "step": 70455
    },
    {
      "entropy": 6.017807960510254,
      "epoch": 5.482007391558063,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002435134577240896,
      "loss": 4.96,
      "mean_token_accuracy": 0.21332394778728486,
      "num_tokens": 122219385.0,
      "step": 70460
    },
    {
      "entropy": 5.946179580688477,
      "epoch": 5.48239642092978,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024348601509792956,
      "loss": 4.9768,
      "mean_token_accuracy": 0.2127937227487564,
      "num_tokens": 122228544.0,
      "step": 70465
    },
    {
      "entropy": 5.911778688430786,
      "epoch": 5.482785450301498,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00024345857294994405,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22565930038690568,
      "num_tokens": 122237870.0,
      "step": 70470
    },
    {
      "entropy": 5.997793817520142,
      "epoch": 5.483174479673215,
      "grad_norm": 1.578125,
      "learning_rate": 0.00024343113128054945,
      "loss": 4.9711,
      "mean_token_accuracy": 0.22394575774669648,
      "num_tokens": 122245525.0,
      "step": 70475
    },
    {
      "entropy": 5.944190931320191,
      "epoch": 5.483563509044933,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002434036900901621,
      "loss": 4.8172,
      "mean_token_accuracy": 0.23269631415605546,
      "num_tokens": 122253689.0,
      "step": 70480
    },
    {
      "entropy": 5.936705780029297,
      "epoch": 5.48395253841665,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00024337624937919856,
      "loss": 4.9918,
      "mean_token_accuracy": 0.21378928124904634,
      "num_tokens": 122262179.0,
      "step": 70485
    },
    {
      "entropy": 5.905461835861206,
      "epoch": 5.484341567788368,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00024334880914807512,
      "loss": 4.9208,
      "mean_token_accuracy": 0.22155172675848006,
      "num_tokens": 122270304.0,
      "step": 70490
    },
    {
      "entropy": 5.910798120498657,
      "epoch": 5.484730597160086,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024332136939720806,
      "loss": 4.8617,
      "mean_token_accuracy": 0.2284513831138611,
      "num_tokens": 122278480.0,
      "step": 70495
    },
    {
      "entropy": 5.919939947128296,
      "epoch": 5.485119626531803,
      "grad_norm": 1.5625,
      "learning_rate": 0.00024329393012701396,
      "loss": 4.9232,
      "mean_token_accuracy": 0.21460044980049134,
      "num_tokens": 122286425.0,
      "step": 70500
    },
    {
      "entropy": 5.948812103271484,
      "epoch": 5.485508655903521,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024326649133790885,
      "loss": 4.9301,
      "mean_token_accuracy": 0.21724251359701158,
      "num_tokens": 122295053.0,
      "step": 70505
    },
    {
      "entropy": 5.945113039016723,
      "epoch": 5.485897685275238,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024323905303030925,
      "loss": 4.859,
      "mean_token_accuracy": 0.23535158783197402,
      "num_tokens": 122304227.0,
      "step": 70510
    },
    {
      "entropy": 5.919388103485107,
      "epoch": 5.486286714646956,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024321161520463147,
      "loss": 4.9161,
      "mean_token_accuracy": 0.22310811877250672,
      "num_tokens": 122312930.0,
      "step": 70515
    },
    {
      "entropy": 5.915830373764038,
      "epoch": 5.486675744018673,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024318417786129188,
      "loss": 4.8991,
      "mean_token_accuracy": 0.2226574882864952,
      "num_tokens": 122321340.0,
      "step": 70520
    },
    {
      "entropy": 5.961959409713745,
      "epoch": 5.487064773390391,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024315674100070672,
      "loss": 4.9588,
      "mean_token_accuracy": 0.21752821803092956,
      "num_tokens": 122329697.0,
      "step": 70525
    },
    {
      "entropy": 5.994410610198974,
      "epoch": 5.487453802762109,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002431293046232923,
      "loss": 4.9802,
      "mean_token_accuracy": 0.21960243135690688,
      "num_tokens": 122337713.0,
      "step": 70530
    },
    {
      "entropy": 5.9285054206848145,
      "epoch": 5.487842832133826,
      "grad_norm": 1.5,
      "learning_rate": 0.00024310186872946498,
      "loss": 4.8602,
      "mean_token_accuracy": 0.22690443247556685,
      "num_tokens": 122346992.0,
      "step": 70535
    },
    {
      "entropy": 5.905802488327026,
      "epoch": 5.488231861505543,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024307443331964107,
      "loss": 4.8706,
      "mean_token_accuracy": 0.21943615525960922,
      "num_tokens": 122355002.0,
      "step": 70540
    },
    {
      "entropy": 5.9703226566314695,
      "epoch": 5.488620890877261,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00024304699839423677,
      "loss": 4.9387,
      "mean_token_accuracy": 0.2181223973631859,
      "num_tokens": 122363300.0,
      "step": 70545
    },
    {
      "entropy": 5.981680774688721,
      "epoch": 5.489009920248979,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024301956395366838,
      "loss": 4.9603,
      "mean_token_accuracy": 0.2173103481531143,
      "num_tokens": 122372650.0,
      "step": 70550
    },
    {
      "entropy": 5.958162260055542,
      "epoch": 5.489398949620696,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024299212999835223,
      "loss": 4.8677,
      "mean_token_accuracy": 0.22220796644687651,
      "num_tokens": 122381754.0,
      "step": 70555
    },
    {
      "entropy": 5.957894897460937,
      "epoch": 5.489787978992414,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024296469652870456,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21959279030561446,
      "num_tokens": 122390468.0,
      "step": 70560
    },
    {
      "entropy": 5.974099493026733,
      "epoch": 5.490177008364132,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024293726354514166,
      "loss": 4.9713,
      "mean_token_accuracy": 0.22042764127254486,
      "num_tokens": 122398934.0,
      "step": 70565
    },
    {
      "entropy": 5.941821718215943,
      "epoch": 5.490566037735849,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002429098310480798,
      "loss": 4.9452,
      "mean_token_accuracy": 0.220921990275383,
      "num_tokens": 122407226.0,
      "step": 70570
    },
    {
      "entropy": 5.979077386856079,
      "epoch": 5.490955067107567,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00024288239903793508,
      "loss": 4.987,
      "mean_token_accuracy": 0.21571199893951415,
      "num_tokens": 122415565.0,
      "step": 70575
    },
    {
      "entropy": 5.966078567504883,
      "epoch": 5.491344096479284,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024285496751512393,
      "loss": 4.9687,
      "mean_token_accuracy": 0.2156738445162773,
      "num_tokens": 122425506.0,
      "step": 70580
    },
    {
      "entropy": 6.020027589797974,
      "epoch": 5.491733125851002,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002428275364800624,
      "loss": 5.039,
      "mean_token_accuracy": 0.2182349979877472,
      "num_tokens": 122434074.0,
      "step": 70585
    },
    {
      "entropy": 5.928857469558716,
      "epoch": 5.492122155222719,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024280010593316683,
      "loss": 4.9511,
      "mean_token_accuracy": 0.21847590059041977,
      "num_tokens": 122443601.0,
      "step": 70590
    },
    {
      "entropy": 6.032143926620483,
      "epoch": 5.492511184594437,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002427726758748534,
      "loss": 5.0084,
      "mean_token_accuracy": 0.20755247175693511,
      "num_tokens": 122453282.0,
      "step": 70595
    },
    {
      "entropy": 5.992221355438232,
      "epoch": 5.492900213966155,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002427452463055383,
      "loss": 4.9152,
      "mean_token_accuracy": 0.22200459837913514,
      "num_tokens": 122461307.0,
      "step": 70600
    },
    {
      "entropy": 5.894187641143799,
      "epoch": 5.493289243337872,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024271781722563773,
      "loss": 4.7974,
      "mean_token_accuracy": 0.23077647387981415,
      "num_tokens": 122470258.0,
      "step": 70605
    },
    {
      "entropy": 5.943425416946411,
      "epoch": 5.493678272709589,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00024269038863556798,
      "loss": 4.9642,
      "mean_token_accuracy": 0.21794824451208114,
      "num_tokens": 122479019.0,
      "step": 70610
    },
    {
      "entropy": 5.924500322341919,
      "epoch": 5.494067302081307,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00024266296053574506,
      "loss": 4.9006,
      "mean_token_accuracy": 0.22078728675842285,
      "num_tokens": 122487592.0,
      "step": 70615
    },
    {
      "entropy": 5.951456928253174,
      "epoch": 5.494456331453025,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002426355329265853,
      "loss": 4.93,
      "mean_token_accuracy": 0.21787372827529908,
      "num_tokens": 122495648.0,
      "step": 70620
    },
    {
      "entropy": 5.995060825347901,
      "epoch": 5.494845360824742,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024260810580850479,
      "loss": 4.9654,
      "mean_token_accuracy": 0.2104364737868309,
      "num_tokens": 122504244.0,
      "step": 70625
    },
    {
      "entropy": 5.8834315776824955,
      "epoch": 5.49523439019646,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024258067918191971,
      "loss": 4.8144,
      "mean_token_accuracy": 0.23290845900774002,
      "num_tokens": 122512940.0,
      "step": 70630
    },
    {
      "entropy": 5.971502447128296,
      "epoch": 5.495623419568178,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002425532530472463,
      "loss": 4.992,
      "mean_token_accuracy": 0.2123175159096718,
      "num_tokens": 122522658.0,
      "step": 70635
    },
    {
      "entropy": 5.972108983993531,
      "epoch": 5.496012448939895,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00024252582740490053,
      "loss": 4.9415,
      "mean_token_accuracy": 0.22343245297670364,
      "num_tokens": 122531349.0,
      "step": 70640
    },
    {
      "entropy": 6.027799987792969,
      "epoch": 5.496401478311612,
      "grad_norm": 1.625,
      "learning_rate": 0.00024249840225529874,
      "loss": 4.9581,
      "mean_token_accuracy": 0.2242451012134552,
      "num_tokens": 122539891.0,
      "step": 70645
    },
    {
      "entropy": 5.999165821075439,
      "epoch": 5.49679050768333,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002424709775988569,
      "loss": 5.0093,
      "mean_token_accuracy": 0.20905782729387284,
      "num_tokens": 122548003.0,
      "step": 70650
    },
    {
      "entropy": 5.9686705589294435,
      "epoch": 5.497179537055048,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024244355343599116,
      "loss": 4.926,
      "mean_token_accuracy": 0.2171271786093712,
      "num_tokens": 122556778.0,
      "step": 70655
    },
    {
      "entropy": 5.98237624168396,
      "epoch": 5.497568566426765,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024241612976711786,
      "loss": 4.9161,
      "mean_token_accuracy": 0.22354793697595596,
      "num_tokens": 122565470.0,
      "step": 70660
    },
    {
      "entropy": 5.933294296264648,
      "epoch": 5.497957595798483,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00024238870659265278,
      "loss": 4.9522,
      "mean_token_accuracy": 0.2193521499633789,
      "num_tokens": 122573843.0,
      "step": 70665
    },
    {
      "entropy": 5.918291997909546,
      "epoch": 5.498346625170201,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024236128391301215,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22633672654628753,
      "num_tokens": 122582680.0,
      "step": 70670
    },
    {
      "entropy": 5.989422082901001,
      "epoch": 5.4987356545419175,
      "grad_norm": 1.609375,
      "learning_rate": 0.00024233386172861217,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21458499878644943,
      "num_tokens": 122591364.0,
      "step": 70675
    },
    {
      "entropy": 6.03534426689148,
      "epoch": 5.499124683913635,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024230644003986869,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20704880952835084,
      "num_tokens": 122599885.0,
      "step": 70680
    },
    {
      "entropy": 6.012503433227539,
      "epoch": 5.499513713285353,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024227901884719806,
      "loss": 4.9205,
      "mean_token_accuracy": 0.22470949739217758,
      "num_tokens": 122609087.0,
      "step": 70685
    },
    {
      "entropy": 6.026303434371949,
      "epoch": 5.499902742657071,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024225159815101626,
      "loss": 5.0268,
      "mean_token_accuracy": 0.2101210370659828,
      "num_tokens": 122617997.0,
      "step": 70690
    },
    {
      "entropy": 5.966021776199341,
      "epoch": 5.500291772028788,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024222417795173929,
      "loss": 4.949,
      "mean_token_accuracy": 0.21887064278125762,
      "num_tokens": 122626488.0,
      "step": 70695
    },
    {
      "entropy": 5.971127319335937,
      "epoch": 5.500680801400506,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024219675824978332,
      "loss": 4.9706,
      "mean_token_accuracy": 0.21735353767871857,
      "num_tokens": 122635279.0,
      "step": 70700
    },
    {
      "entropy": 5.941038656234741,
      "epoch": 5.501069830772224,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024216933904556425,
      "loss": 4.8744,
      "mean_token_accuracy": 0.23166452944278718,
      "num_tokens": 122643129.0,
      "step": 70705
    },
    {
      "entropy": 5.957791042327881,
      "epoch": 5.501458860143941,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024214192033949827,
      "loss": 4.9801,
      "mean_token_accuracy": 0.21714633256196975,
      "num_tokens": 122651191.0,
      "step": 70710
    },
    {
      "entropy": 6.003350877761841,
      "epoch": 5.501847889515658,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024211450213200132,
      "loss": 5.01,
      "mean_token_accuracy": 0.2171599641442299,
      "num_tokens": 122660422.0,
      "step": 70715
    },
    {
      "entropy": 6.02469048500061,
      "epoch": 5.502236918887376,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024208708442348943,
      "loss": 5.0324,
      "mean_token_accuracy": 0.20958724915981292,
      "num_tokens": 122668714.0,
      "step": 70720
    },
    {
      "entropy": 5.898644304275512,
      "epoch": 5.502625948259094,
      "grad_norm": 1.5625,
      "learning_rate": 0.00024205966721437867,
      "loss": 4.9012,
      "mean_token_accuracy": 0.22218665331602097,
      "num_tokens": 122677814.0,
      "step": 70725
    },
    {
      "entropy": 5.977047157287598,
      "epoch": 5.503014977630811,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024203225050508503,
      "loss": 4.9393,
      "mean_token_accuracy": 0.2231607675552368,
      "num_tokens": 122686213.0,
      "step": 70730
    },
    {
      "entropy": 6.026789426803589,
      "epoch": 5.503404007002529,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002420048342960245,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21268612891435623,
      "num_tokens": 122694938.0,
      "step": 70735
    },
    {
      "entropy": 5.976043939590454,
      "epoch": 5.503793036374246,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024197741858761318,
      "loss": 5.0041,
      "mean_token_accuracy": 0.21414270848035813,
      "num_tokens": 122703121.0,
      "step": 70740
    },
    {
      "entropy": 5.974841165542602,
      "epoch": 5.5041820657459635,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002419500033802669,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21258212327957154,
      "num_tokens": 122711741.0,
      "step": 70745
    },
    {
      "entropy": 5.99524245262146,
      "epoch": 5.504571095117681,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002419225886744016,
      "loss": 4.8789,
      "mean_token_accuracy": 0.21784351766109467,
      "num_tokens": 122720730.0,
      "step": 70750
    },
    {
      "entropy": 5.948440408706665,
      "epoch": 5.504960124489399,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024189517447043352,
      "loss": 4.9176,
      "mean_token_accuracy": 0.21818767338991166,
      "num_tokens": 122729483.0,
      "step": 70755
    },
    {
      "entropy": 5.881301927566528,
      "epoch": 5.505349153861117,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024186776076877843,
      "loss": 4.844,
      "mean_token_accuracy": 0.22190718203783036,
      "num_tokens": 122738479.0,
      "step": 70760
    },
    {
      "entropy": 5.969672393798828,
      "epoch": 5.505738183232834,
      "grad_norm": 1.5625,
      "learning_rate": 0.00024184034756985235,
      "loss": 4.9749,
      "mean_token_accuracy": 0.21714142858982086,
      "num_tokens": 122747524.0,
      "step": 70765
    },
    {
      "entropy": 5.947131204605102,
      "epoch": 5.506127212604552,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024181293487407125,
      "loss": 4.9363,
      "mean_token_accuracy": 0.22481646686792373,
      "num_tokens": 122756565.0,
      "step": 70770
    },
    {
      "entropy": 5.896308135986328,
      "epoch": 5.50651624197627,
      "grad_norm": 1.46875,
      "learning_rate": 0.000241785522681851,
      "loss": 4.9046,
      "mean_token_accuracy": 0.2226010650396347,
      "num_tokens": 122764915.0,
      "step": 70775
    },
    {
      "entropy": 5.9126574993133545,
      "epoch": 5.5069052713479865,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024175811099360757,
      "loss": 4.8833,
      "mean_token_accuracy": 0.21962203234434127,
      "num_tokens": 122773933.0,
      "step": 70780
    },
    {
      "entropy": 5.958370018005371,
      "epoch": 5.507294300719704,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024173069980975691,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2189391851425171,
      "num_tokens": 122782440.0,
      "step": 70785
    },
    {
      "entropy": 6.003009939193726,
      "epoch": 5.507683330091422,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00024170328913071493,
      "loss": 4.9145,
      "mean_token_accuracy": 0.2255675032734871,
      "num_tokens": 122790916.0,
      "step": 70790
    },
    {
      "entropy": 6.013731575012207,
      "epoch": 5.5080723594631396,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00024167587895689753,
      "loss": 4.9635,
      "mean_token_accuracy": 0.22456276118755342,
      "num_tokens": 122799660.0,
      "step": 70795
    },
    {
      "entropy": 5.941084384918213,
      "epoch": 5.508461388834857,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024164846928872064,
      "loss": 4.9129,
      "mean_token_accuracy": 0.21477831155061722,
      "num_tokens": 122808191.0,
      "step": 70800
    },
    {
      "entropy": 5.932675457000732,
      "epoch": 5.508850418206575,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024162106012660012,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2162783920764923,
      "num_tokens": 122816604.0,
      "step": 70805
    },
    {
      "entropy": 6.02034912109375,
      "epoch": 5.509239447578292,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002415936514709519,
      "loss": 5.0228,
      "mean_token_accuracy": 0.21489442735910416,
      "num_tokens": 122824539.0,
      "step": 70810
    },
    {
      "entropy": 5.966786098480225,
      "epoch": 5.5096284769500095,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002415662433221918,
      "loss": 4.9584,
      "mean_token_accuracy": 0.2179805800318718,
      "num_tokens": 122832438.0,
      "step": 70815
    },
    {
      "entropy": 6.106759357452392,
      "epoch": 5.510017506321727,
      "grad_norm": 1.703125,
      "learning_rate": 0.00024153883568073576,
      "loss": 5.1205,
      "mean_token_accuracy": 0.20764787644147872,
      "num_tokens": 122840850.0,
      "step": 70820
    },
    {
      "entropy": 6.029980564117432,
      "epoch": 5.510406535693445,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002415114285469997,
      "loss": 5.0045,
      "mean_token_accuracy": 0.2158176839351654,
      "num_tokens": 122849790.0,
      "step": 70825
    },
    {
      "entropy": 5.910354471206665,
      "epoch": 5.5107955650651625,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024148402192139935,
      "loss": 4.8439,
      "mean_token_accuracy": 0.2267097771167755,
      "num_tokens": 122857059.0,
      "step": 70830
    },
    {
      "entropy": 5.986735773086548,
      "epoch": 5.51118459443688,
      "grad_norm": 1.5,
      "learning_rate": 0.00024145661580435063,
      "loss": 4.9685,
      "mean_token_accuracy": 0.2182190537452698,
      "num_tokens": 122865897.0,
      "step": 70835
    },
    {
      "entropy": 6.002157354354859,
      "epoch": 5.511573623808598,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024142921019626933,
      "loss": 4.967,
      "mean_token_accuracy": 0.21898044496774674,
      "num_tokens": 122875371.0,
      "step": 70840
    },
    {
      "entropy": 5.927171421051026,
      "epoch": 5.511962653180315,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024140180509757143,
      "loss": 4.9082,
      "mean_token_accuracy": 0.22391510903835296,
      "num_tokens": 122883369.0,
      "step": 70845
    },
    {
      "entropy": 5.9694311141967775,
      "epoch": 5.5123516825520325,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024137440050867265,
      "loss": 4.9922,
      "mean_token_accuracy": 0.2128205418586731,
      "num_tokens": 122892518.0,
      "step": 70850
    },
    {
      "entropy": 5.959693098068238,
      "epoch": 5.51274071192375,
      "grad_norm": 1.5625,
      "learning_rate": 0.00024134699642998882,
      "loss": 4.9322,
      "mean_token_accuracy": 0.21817529201507568,
      "num_tokens": 122901257.0,
      "step": 70855
    },
    {
      "entropy": 5.9683057308197025,
      "epoch": 5.513129741295468,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024131959286193572,
      "loss": 4.9674,
      "mean_token_accuracy": 0.22104318737983703,
      "num_tokens": 122910294.0,
      "step": 70860
    },
    {
      "entropy": 5.960390281677246,
      "epoch": 5.5135187706671855,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024129218980492917,
      "loss": 4.9926,
      "mean_token_accuracy": 0.2167594015598297,
      "num_tokens": 122919122.0,
      "step": 70865
    },
    {
      "entropy": 5.967777824401855,
      "epoch": 5.513907800038903,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000241264787259385,
      "loss": 4.9277,
      "mean_token_accuracy": 0.21707086712121965,
      "num_tokens": 122927230.0,
      "step": 70870
    },
    {
      "entropy": 5.9385534763336185,
      "epoch": 5.51429682941062,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024123738522571902,
      "loss": 4.902,
      "mean_token_accuracy": 0.22274251729249955,
      "num_tokens": 122936028.0,
      "step": 70875
    },
    {
      "entropy": 5.963991308212281,
      "epoch": 5.514685858782338,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002412099837043469,
      "loss": 4.9035,
      "mean_token_accuracy": 0.22129716277122496,
      "num_tokens": 122943713.0,
      "step": 70880
    },
    {
      "entropy": 6.009473276138306,
      "epoch": 5.515074888154055,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024118258269568456,
      "loss": 4.9558,
      "mean_token_accuracy": 0.22478728443384172,
      "num_tokens": 122952525.0,
      "step": 70885
    },
    {
      "entropy": 5.9801506996154785,
      "epoch": 5.515463917525773,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024115518220014771,
      "loss": 5.0321,
      "mean_token_accuracy": 0.21471601575613022,
      "num_tokens": 122961735.0,
      "step": 70890
    },
    {
      "entropy": 5.937101030349732,
      "epoch": 5.515852946897491,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00024112778221815206,
      "loss": 4.8996,
      "mean_token_accuracy": 0.21993622183799744,
      "num_tokens": 122969956.0,
      "step": 70895
    },
    {
      "entropy": 5.951824951171875,
      "epoch": 5.5162419762692085,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00024110038275011343,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21865475922822952,
      "num_tokens": 122978724.0,
      "step": 70900
    },
    {
      "entropy": 5.956900405883789,
      "epoch": 5.516631005640926,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00024107298379644754,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21959673911333083,
      "num_tokens": 122986904.0,
      "step": 70905
    },
    {
      "entropy": 5.860125350952148,
      "epoch": 5.517020035012644,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002410455853575701,
      "loss": 4.8156,
      "mean_token_accuracy": 0.23102680295705796,
      "num_tokens": 122995676.0,
      "step": 70910
    },
    {
      "entropy": 5.867335510253906,
      "epoch": 5.517409064384361,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00024101818743389687,
      "loss": 4.8778,
      "mean_token_accuracy": 0.2254628911614418,
      "num_tokens": 123004044.0,
      "step": 70915
    },
    {
      "entropy": 5.908370065689087,
      "epoch": 5.517798093756078,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00024099079002584355,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21371461749076842,
      "num_tokens": 123014416.0,
      "step": 70920
    },
    {
      "entropy": 5.933271026611328,
      "epoch": 5.518187123127796,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00024096339313382586,
      "loss": 4.9146,
      "mean_token_accuracy": 0.22008589953184127,
      "num_tokens": 123023049.0,
      "step": 70925
    },
    {
      "entropy": 5.931514549255371,
      "epoch": 5.518576152499514,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024093599675825954,
      "loss": 4.8129,
      "mean_token_accuracy": 0.22620275616645813,
      "num_tokens": 123031420.0,
      "step": 70930
    },
    {
      "entropy": 6.034665489196778,
      "epoch": 5.5189651818712315,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00024090860089956024,
      "loss": 4.9872,
      "mean_token_accuracy": 0.2136535793542862,
      "num_tokens": 123039515.0,
      "step": 70935
    },
    {
      "entropy": 5.934506416320801,
      "epoch": 5.519354211242949,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002408812055581437,
      "loss": 4.9822,
      "mean_token_accuracy": 0.22170058339834214,
      "num_tokens": 123048131.0,
      "step": 70940
    },
    {
      "entropy": 5.979631233215332,
      "epoch": 5.519743240614666,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024085381073442536,
      "loss": 4.9534,
      "mean_token_accuracy": 0.21521830707788467,
      "num_tokens": 123056665.0,
      "step": 70945
    },
    {
      "entropy": 6.0260542869567875,
      "epoch": 5.520132269986384,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024082641642882125,
      "loss": 5.0597,
      "mean_token_accuracy": 0.20866467505693437,
      "num_tokens": 123065855.0,
      "step": 70950
    },
    {
      "entropy": 5.940960836410523,
      "epoch": 5.520521299358101,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00024079902264174686,
      "loss": 4.8715,
      "mean_token_accuracy": 0.22485389858484267,
      "num_tokens": 123073325.0,
      "step": 70955
    },
    {
      "entropy": 5.916835689544678,
      "epoch": 5.520910328729819,
      "grad_norm": 1.5,
      "learning_rate": 0.0002407716293736179,
      "loss": 4.9367,
      "mean_token_accuracy": 0.22424513846635818,
      "num_tokens": 123081932.0,
      "step": 70960
    },
    {
      "entropy": 5.892676973342896,
      "epoch": 5.521299358101537,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024074423662484996,
      "loss": 4.8671,
      "mean_token_accuracy": 0.22854380756616594,
      "num_tokens": 123090801.0,
      "step": 70965
    },
    {
      "entropy": 5.94747405052185,
      "epoch": 5.5216883874732545,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002407168443958587,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2157708302140236,
      "num_tokens": 123098901.0,
      "step": 70970
    },
    {
      "entropy": 5.992716646194458,
      "epoch": 5.522077416844972,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024068945268705978,
      "loss": 4.9861,
      "mean_token_accuracy": 0.2144833117723465,
      "num_tokens": 123106902.0,
      "step": 70975
    },
    {
      "entropy": 5.972788381576538,
      "epoch": 5.522466446216689,
      "grad_norm": 1.578125,
      "learning_rate": 0.00024066206149886876,
      "loss": 4.9827,
      "mean_token_accuracy": 0.21229513138532638,
      "num_tokens": 123115548.0,
      "step": 70980
    },
    {
      "entropy": 5.978305053710938,
      "epoch": 5.522855475588407,
      "grad_norm": 1.578125,
      "learning_rate": 0.00024063467083170132,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21809609234333038,
      "num_tokens": 123123749.0,
      "step": 70985
    },
    {
      "entropy": 5.932640790939331,
      "epoch": 5.523244504960124,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002406072806859731,
      "loss": 4.9673,
      "mean_token_accuracy": 0.2177960082888603,
      "num_tokens": 123131963.0,
      "step": 70990
    },
    {
      "entropy": 5.8571524143219,
      "epoch": 5.523633534331842,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002405798910620996,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22494855672121047,
      "num_tokens": 123140011.0,
      "step": 70995
    },
    {
      "entropy": 6.021400213241577,
      "epoch": 5.52402256370356,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002405525019604965,
      "loss": 5.0384,
      "mean_token_accuracy": 0.21281415820121766,
      "num_tokens": 123148470.0,
      "step": 71000
    },
    {
      "entropy": 5.991381025314331,
      "epoch": 5.5244115930752775,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024052511338157928,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21436567604541779,
      "num_tokens": 123157354.0,
      "step": 71005
    },
    {
      "entropy": 6.002187061309814,
      "epoch": 5.524800622446994,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00024049772532576364,
      "loss": 5.0168,
      "mean_token_accuracy": 0.21916416734457017,
      "num_tokens": 123166088.0,
      "step": 71010
    },
    {
      "entropy": 5.880302381515503,
      "epoch": 5.525189651818712,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024047033779346505,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21861686557531357,
      "num_tokens": 123174675.0,
      "step": 71015
    },
    {
      "entropy": 5.975411510467529,
      "epoch": 5.52557868119043,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002404429507850992,
      "loss": 4.9869,
      "mean_token_accuracy": 0.22323040813207626,
      "num_tokens": 123184362.0,
      "step": 71020
    },
    {
      "entropy": 5.929631996154785,
      "epoch": 5.525967710562147,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002404155643010815,
      "loss": 4.9412,
      "mean_token_accuracy": 0.21899281293153763,
      "num_tokens": 123193393.0,
      "step": 71025
    },
    {
      "entropy": 5.995051383972168,
      "epoch": 5.526356739933865,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024038817834182762,
      "loss": 4.9717,
      "mean_token_accuracy": 0.21933116912841796,
      "num_tokens": 123202125.0,
      "step": 71030
    },
    {
      "entropy": 5.898466920852661,
      "epoch": 5.526745769305583,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024036079290775308,
      "loss": 4.9067,
      "mean_token_accuracy": 0.22351279407739638,
      "num_tokens": 123210399.0,
      "step": 71035
    },
    {
      "entropy": 6.002439641952515,
      "epoch": 5.5271347986773005,
      "grad_norm": 1.703125,
      "learning_rate": 0.00024033340799927338,
      "loss": 4.9907,
      "mean_token_accuracy": 0.2154357135295868,
      "num_tokens": 123219247.0,
      "step": 71040
    },
    {
      "entropy": 5.9523979187011715,
      "epoch": 5.527523828049017,
      "grad_norm": 1.5625,
      "learning_rate": 0.000240306023616804,
      "loss": 4.9539,
      "mean_token_accuracy": 0.2259290933609009,
      "num_tokens": 123227614.0,
      "step": 71045
    },
    {
      "entropy": 5.948361206054687,
      "epoch": 5.527912857420735,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024027863976076047,
      "loss": 4.9603,
      "mean_token_accuracy": 0.2179491013288498,
      "num_tokens": 123235456.0,
      "step": 71050
    },
    {
      "entropy": 5.915952491760254,
      "epoch": 5.528301886792453,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002402512564315583,
      "loss": 4.854,
      "mean_token_accuracy": 0.2244955286383629,
      "num_tokens": 123244411.0,
      "step": 71055
    },
    {
      "entropy": 5.96636734008789,
      "epoch": 5.52869091616417,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024022387362961306,
      "loss": 4.9092,
      "mean_token_accuracy": 0.2226063758134842,
      "num_tokens": 123253343.0,
      "step": 71060
    },
    {
      "entropy": 6.028327131271363,
      "epoch": 5.529079945535888,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024019649135534015,
      "loss": 5.0369,
      "mean_token_accuracy": 0.21102530062198638,
      "num_tokens": 123262528.0,
      "step": 71065
    },
    {
      "entropy": 5.91817717552185,
      "epoch": 5.529468974907606,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024016910960915512,
      "loss": 4.969,
      "mean_token_accuracy": 0.2161789357662201,
      "num_tokens": 123272081.0,
      "step": 71070
    },
    {
      "entropy": 6.017677307128906,
      "epoch": 5.529858004279323,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024014172839147336,
      "loss": 4.9891,
      "mean_token_accuracy": 0.21666686832904816,
      "num_tokens": 123280650.0,
      "step": 71075
    },
    {
      "entropy": 6.002981567382813,
      "epoch": 5.53024703365104,
      "grad_norm": 1.546875,
      "learning_rate": 0.00024011434770271033,
      "loss": 4.9473,
      "mean_token_accuracy": 0.2170201227068901,
      "num_tokens": 123288987.0,
      "step": 71080
    },
    {
      "entropy": 6.000453281402588,
      "epoch": 5.530636063022758,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002400869675432817,
      "loss": 4.9936,
      "mean_token_accuracy": 0.22091503143310548,
      "num_tokens": 123296898.0,
      "step": 71085
    },
    {
      "entropy": 5.917473697662354,
      "epoch": 5.531025092394476,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024005958791360266,
      "loss": 4.9257,
      "mean_token_accuracy": 0.21992261558771134,
      "num_tokens": 123305611.0,
      "step": 71090
    },
    {
      "entropy": 5.960154962539673,
      "epoch": 5.531414121766193,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024003220881408882,
      "loss": 4.9272,
      "mean_token_accuracy": 0.22469554990530013,
      "num_tokens": 123313794.0,
      "step": 71095
    },
    {
      "entropy": 5.978445100784302,
      "epoch": 5.531803151137911,
      "grad_norm": 1.5625,
      "learning_rate": 0.00024000483024515552,
      "loss": 4.9403,
      "mean_token_accuracy": 0.2225944072008133,
      "num_tokens": 123322088.0,
      "step": 71100
    },
    {
      "entropy": 5.9850749492645265,
      "epoch": 5.532192180509629,
      "grad_norm": 1.5,
      "learning_rate": 0.00023997745220721823,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21468393057584761,
      "num_tokens": 123330111.0,
      "step": 71105
    },
    {
      "entropy": 5.949058103561401,
      "epoch": 5.5325812098813465,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023995007470069236,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22267056852579117,
      "num_tokens": 123338589.0,
      "step": 71110
    },
    {
      "entropy": 5.995131540298462,
      "epoch": 5.532970239253063,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023992269772599335,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20801491141319275,
      "num_tokens": 123347941.0,
      "step": 71115
    },
    {
      "entropy": 6.005447101593018,
      "epoch": 5.533359268624781,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023989532128353654,
      "loss": 5.099,
      "mean_token_accuracy": 0.20693634897470475,
      "num_tokens": 123358431.0,
      "step": 71120
    },
    {
      "entropy": 5.914597082138061,
      "epoch": 5.533748297996499,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002398679453737374,
      "loss": 4.8777,
      "mean_token_accuracy": 0.2233620584011078,
      "num_tokens": 123366548.0,
      "step": 71125
    },
    {
      "entropy": 5.97241473197937,
      "epoch": 5.534137327368216,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023984056999701121,
      "loss": 4.9184,
      "mean_token_accuracy": 0.22294820994138717,
      "num_tokens": 123375438.0,
      "step": 71130
    },
    {
      "entropy": 5.941115474700927,
      "epoch": 5.534526356739934,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023981319515377347,
      "loss": 4.948,
      "mean_token_accuracy": 0.21383794844150544,
      "num_tokens": 123384340.0,
      "step": 71135
    },
    {
      "entropy": 5.94424524307251,
      "epoch": 5.534915386111652,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023978582084443939,
      "loss": 4.9306,
      "mean_token_accuracy": 0.22128111720085145,
      "num_tokens": 123393088.0,
      "step": 71140
    },
    {
      "entropy": 5.840986919403076,
      "epoch": 5.535304415483369,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002397584470694244,
      "loss": 4.6682,
      "mean_token_accuracy": 0.24049134701490402,
      "num_tokens": 123402198.0,
      "step": 71145
    },
    {
      "entropy": 5.985688161849976,
      "epoch": 5.535693444855086,
      "grad_norm": 1.484375,
      "learning_rate": 0.000239731073829144,
      "loss": 4.98,
      "mean_token_accuracy": 0.21640587896108626,
      "num_tokens": 123411143.0,
      "step": 71150
    },
    {
      "entropy": 6.006634521484375,
      "epoch": 5.536082474226804,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0002397037011240134,
      "loss": 5.0491,
      "mean_token_accuracy": 0.21639318466186525,
      "num_tokens": 123419779.0,
      "step": 71155
    },
    {
      "entropy": 5.944285726547241,
      "epoch": 5.536471503598522,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023967632895444796,
      "loss": 4.946,
      "mean_token_accuracy": 0.2263677030801773,
      "num_tokens": 123429002.0,
      "step": 71160
    },
    {
      "entropy": 5.875865697860718,
      "epoch": 5.536860532970239,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000239648957320863,
      "loss": 4.804,
      "mean_token_accuracy": 0.2277510493993759,
      "num_tokens": 123437178.0,
      "step": 71165
    },
    {
      "entropy": 5.963204383850098,
      "epoch": 5.537249562341957,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0002396215862236738,
      "loss": 4.8765,
      "mean_token_accuracy": 0.2362345963716507,
      "num_tokens": 123446927.0,
      "step": 71170
    },
    {
      "entropy": 5.982075977325439,
      "epoch": 5.537638591713675,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023959421566329574,
      "loss": 4.9526,
      "mean_token_accuracy": 0.21260907948017121,
      "num_tokens": 123456567.0,
      "step": 71175
    },
    {
      "entropy": 5.941429853439331,
      "epoch": 5.538027621085392,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023956684564014415,
      "loss": 4.8368,
      "mean_token_accuracy": 0.22583421915769578,
      "num_tokens": 123465142.0,
      "step": 71180
    },
    {
      "entropy": 5.939698982238769,
      "epoch": 5.538416650457109,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023953947615463423,
      "loss": 4.9472,
      "mean_token_accuracy": 0.21497834026813506,
      "num_tokens": 123474434.0,
      "step": 71185
    },
    {
      "entropy": 5.902551889419556,
      "epoch": 5.538805679828827,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002395121072071813,
      "loss": 4.9327,
      "mean_token_accuracy": 0.22042478173971175,
      "num_tokens": 123483381.0,
      "step": 71190
    },
    {
      "entropy": 5.986184978485108,
      "epoch": 5.539194709200545,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023948473879820072,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21151377707719804,
      "num_tokens": 123491607.0,
      "step": 71195
    },
    {
      "entropy": 5.994960784912109,
      "epoch": 5.539583738572262,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023945737092810764,
      "loss": 5.0565,
      "mean_token_accuracy": 0.21053973585367203,
      "num_tokens": 123500938.0,
      "step": 71200
    },
    {
      "entropy": 5.964706563949585,
      "epoch": 5.53997276794398,
      "grad_norm": 1.5,
      "learning_rate": 0.0002394300035973175,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21413474380970002,
      "num_tokens": 123509397.0,
      "step": 71205
    },
    {
      "entropy": 5.9748788356781,
      "epoch": 5.540361797315697,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023940263680624535,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22080014348030091,
      "num_tokens": 123518366.0,
      "step": 71210
    },
    {
      "entropy": 5.987403297424317,
      "epoch": 5.540750826687415,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023937527055530643,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21517172157764436,
      "num_tokens": 123527159.0,
      "step": 71215
    },
    {
      "entropy": 5.8696386337280275,
      "epoch": 5.541139856059132,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002393479048449162,
      "loss": 4.8529,
      "mean_token_accuracy": 0.22951883524656297,
      "num_tokens": 123536266.0,
      "step": 71220
    },
    {
      "entropy": 5.9507331371307375,
      "epoch": 5.54152888543085,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023932053967548971,
      "loss": 5.0116,
      "mean_token_accuracy": 0.21587826907634736,
      "num_tokens": 123545219.0,
      "step": 71225
    },
    {
      "entropy": 5.936486482620239,
      "epoch": 5.541917914802568,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023929317504744235,
      "loss": 4.8562,
      "mean_token_accuracy": 0.22185115963220597,
      "num_tokens": 123553920.0,
      "step": 71230
    },
    {
      "entropy": 6.017495346069336,
      "epoch": 5.542306944174285,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023926581096118916,
      "loss": 4.964,
      "mean_token_accuracy": 0.2178903803229332,
      "num_tokens": 123562255.0,
      "step": 71235
    },
    {
      "entropy": 5.936067295074463,
      "epoch": 5.542695973546003,
      "grad_norm": 1.5,
      "learning_rate": 0.0002392384474171454,
      "loss": 4.9048,
      "mean_token_accuracy": 0.220511095225811,
      "num_tokens": 123570931.0,
      "step": 71240
    },
    {
      "entropy": 5.972685432434082,
      "epoch": 5.543085002917721,
      "grad_norm": 1.609375,
      "learning_rate": 0.00023921108441572625,
      "loss": 5.0264,
      "mean_token_accuracy": 0.2090453565120697,
      "num_tokens": 123578737.0,
      "step": 71245
    },
    {
      "entropy": 5.930301475524902,
      "epoch": 5.543474032289438,
      "grad_norm": 1.640625,
      "learning_rate": 0.000239183721957347,
      "loss": 4.8875,
      "mean_token_accuracy": 0.2216387450695038,
      "num_tokens": 123586514.0,
      "step": 71250
    },
    {
      "entropy": 5.968838739395141,
      "epoch": 5.543863061661155,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023915636004242274,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22548809498548508,
      "num_tokens": 123595516.0,
      "step": 71255
    },
    {
      "entropy": 5.963003540039063,
      "epoch": 5.544252091032873,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002391289986713686,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21644422858953477,
      "num_tokens": 123604128.0,
      "step": 71260
    },
    {
      "entropy": 5.912517023086548,
      "epoch": 5.544641120404591,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023910163784459986,
      "loss": 4.9121,
      "mean_token_accuracy": 0.2228151962161064,
      "num_tokens": 123612891.0,
      "step": 71265
    },
    {
      "entropy": 5.914330625534058,
      "epoch": 5.545030149776308,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002390742775625317,
      "loss": 4.9417,
      "mean_token_accuracy": 0.2147146314382553,
      "num_tokens": 123621443.0,
      "step": 71270
    },
    {
      "entropy": 6.066943788528443,
      "epoch": 5.545419179148025,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023904691782557908,
      "loss": 5.0733,
      "mean_token_accuracy": 0.20723872631788254,
      "num_tokens": 123630581.0,
      "step": 71275
    },
    {
      "entropy": 6.018024349212647,
      "epoch": 5.545808208519743,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023901955863415725,
      "loss": 4.9512,
      "mean_token_accuracy": 0.21435263901948928,
      "num_tokens": 123639580.0,
      "step": 71280
    },
    {
      "entropy": 5.968200588226319,
      "epoch": 5.546197237891461,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00023899219998868142,
      "loss": 4.9217,
      "mean_token_accuracy": 0.22358225882053376,
      "num_tokens": 123648024.0,
      "step": 71285
    },
    {
      "entropy": 5.930078983306885,
      "epoch": 5.546586267263178,
      "grad_norm": 1.5625,
      "learning_rate": 0.00023896484188956664,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2146504834294319,
      "num_tokens": 123656274.0,
      "step": 71290
    },
    {
      "entropy": 5.968483972549438,
      "epoch": 5.546975296634896,
      "grad_norm": 1.4375,
      "learning_rate": 0.000238937484337228,
      "loss": 5.0237,
      "mean_token_accuracy": 0.21366752982139586,
      "num_tokens": 123665176.0,
      "step": 71295
    },
    {
      "entropy": 6.0183202743530275,
      "epoch": 5.547364326006614,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023891012733208062,
      "loss": 5.0718,
      "mean_token_accuracy": 0.21795423179864884,
      "num_tokens": 123674460.0,
      "step": 71300
    },
    {
      "entropy": 6.0261688232421875,
      "epoch": 5.547753355378331,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00023888277087453962,
      "loss": 4.9266,
      "mean_token_accuracy": 0.2176585778594017,
      "num_tokens": 123683930.0,
      "step": 71305
    },
    {
      "entropy": 5.941478109359741,
      "epoch": 5.548142384750049,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00023885541496502017,
      "loss": 4.8627,
      "mean_token_accuracy": 0.22013467103242873,
      "num_tokens": 123692718.0,
      "step": 71310
    },
    {
      "entropy": 5.867500877380371,
      "epoch": 5.548531414121766,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023882805960393722,
      "loss": 4.8681,
      "mean_token_accuracy": 0.22218011021614076,
      "num_tokens": 123701526.0,
      "step": 71315
    },
    {
      "entropy": 5.950256299972534,
      "epoch": 5.548920443493484,
      "grad_norm": 1.5,
      "learning_rate": 0.0002388007047917059,
      "loss": 4.9841,
      "mean_token_accuracy": 0.21839507967233657,
      "num_tokens": 123710466.0,
      "step": 71320
    },
    {
      "entropy": 5.858599138259888,
      "epoch": 5.549309472865201,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023877335052874122,
      "loss": 4.8702,
      "mean_token_accuracy": 0.22504011839628218,
      "num_tokens": 123718902.0,
      "step": 71325
    },
    {
      "entropy": 5.913662052154541,
      "epoch": 5.549698502236919,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023874599681545833,
      "loss": 4.9548,
      "mean_token_accuracy": 0.22411651760339737,
      "num_tokens": 123727818.0,
      "step": 71330
    },
    {
      "entropy": 5.949743032455444,
      "epoch": 5.550087531608637,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002387186436522722,
      "loss": 4.9493,
      "mean_token_accuracy": 0.2268570989370346,
      "num_tokens": 123735773.0,
      "step": 71335
    },
    {
      "entropy": 6.037681341171265,
      "epoch": 5.550476560980354,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002386912910395979,
      "loss": 5.0616,
      "mean_token_accuracy": 0.21308959573507308,
      "num_tokens": 123744708.0,
      "step": 71340
    },
    {
      "entropy": 5.946996831893921,
      "epoch": 5.550865590352071,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023866393897785049,
      "loss": 4.9163,
      "mean_token_accuracy": 0.21804601699113846,
      "num_tokens": 123753081.0,
      "step": 71345
    },
    {
      "entropy": 5.9165685176849365,
      "epoch": 5.551254619723789,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023863658746744494,
      "loss": 4.9385,
      "mean_token_accuracy": 0.22170434892177582,
      "num_tokens": 123761581.0,
      "step": 71350
    },
    {
      "entropy": 5.951578092575073,
      "epoch": 5.551643649095507,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023860923650879636,
      "loss": 4.8939,
      "mean_token_accuracy": 0.22442959249019623,
      "num_tokens": 123770000.0,
      "step": 71355
    },
    {
      "entropy": 5.996806478500366,
      "epoch": 5.552032678467224,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002385818861023197,
      "loss": 5.0129,
      "mean_token_accuracy": 0.21791805028915406,
      "num_tokens": 123778540.0,
      "step": 71360
    },
    {
      "entropy": 5.9745598316192625,
      "epoch": 5.552421707838942,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023855453624842986,
      "loss": 4.8575,
      "mean_token_accuracy": 0.2265525281429291,
      "num_tokens": 123787086.0,
      "step": 71365
    },
    {
      "entropy": 5.9080057621002195,
      "epoch": 5.55281073721066,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000238527186947542,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21715307086706162,
      "num_tokens": 123795040.0,
      "step": 71370
    },
    {
      "entropy": 6.031479787826538,
      "epoch": 5.553199766582377,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000238499838200071,
      "loss": 5.1178,
      "mean_token_accuracy": 0.20820316970348357,
      "num_tokens": 123803610.0,
      "step": 71375
    },
    {
      "entropy": 5.93375940322876,
      "epoch": 5.553588795954094,
      "grad_norm": 1.640625,
      "learning_rate": 0.00023847249000643185,
      "loss": 4.8991,
      "mean_token_accuracy": 0.21728943437337875,
      "num_tokens": 123812044.0,
      "step": 71380
    },
    {
      "entropy": 5.982574558258056,
      "epoch": 5.553977825325812,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023844514236703952,
      "loss": 4.9728,
      "mean_token_accuracy": 0.219694884121418,
      "num_tokens": 123820665.0,
      "step": 71385
    },
    {
      "entropy": 5.951496124267578,
      "epoch": 5.5543668546975296,
      "grad_norm": 1.46875,
      "learning_rate": 0.000238417795282309,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22410383373498916,
      "num_tokens": 123828896.0,
      "step": 71390
    },
    {
      "entropy": 5.983322477340698,
      "epoch": 5.554755884069247,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023839044875265526,
      "loss": 4.9956,
      "mean_token_accuracy": 0.2158604696393013,
      "num_tokens": 123837955.0,
      "step": 71395
    },
    {
      "entropy": 6.008596706390381,
      "epoch": 5.555144913440965,
      "grad_norm": 1.625,
      "learning_rate": 0.0002383631027784931,
      "loss": 5.0014,
      "mean_token_accuracy": 0.2183651804924011,
      "num_tokens": 123846421.0,
      "step": 71400
    },
    {
      "entropy": 6.046213006973266,
      "epoch": 5.555533942812683,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023833575736023754,
      "loss": 5.0653,
      "mean_token_accuracy": 0.20254047363996505,
      "num_tokens": 123855208.0,
      "step": 71405
    },
    {
      "entropy": 6.034641599655151,
      "epoch": 5.5559229721843995,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023830841249830348,
      "loss": 5.1275,
      "mean_token_accuracy": 0.20732738673686982,
      "num_tokens": 123863702.0,
      "step": 71410
    },
    {
      "entropy": 5.982849168777466,
      "epoch": 5.556312001556117,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002382810681931059,
      "loss": 4.9369,
      "mean_token_accuracy": 0.21981185376644136,
      "num_tokens": 123872602.0,
      "step": 71415
    },
    {
      "entropy": 6.028792524337769,
      "epoch": 5.556701030927835,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00023825372444505967,
      "loss": 4.9992,
      "mean_token_accuracy": 0.21216824799776077,
      "num_tokens": 123881614.0,
      "step": 71420
    },
    {
      "entropy": 5.92593297958374,
      "epoch": 5.5570900602995525,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023822638125457974,
      "loss": 4.8565,
      "mean_token_accuracy": 0.21412606835365294,
      "num_tokens": 123889630.0,
      "step": 71425
    },
    {
      "entropy": 5.975437974929809,
      "epoch": 5.55747908967127,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023819903862208093,
      "loss": 4.9517,
      "mean_token_accuracy": 0.22103043794631957,
      "num_tokens": 123898288.0,
      "step": 71430
    },
    {
      "entropy": 5.9636406898498535,
      "epoch": 5.557868119042988,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002381716965479781,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2189297005534172,
      "num_tokens": 123906540.0,
      "step": 71435
    },
    {
      "entropy": 5.870230627059937,
      "epoch": 5.558257148414706,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00023814435503268617,
      "loss": 4.8061,
      "mean_token_accuracy": 0.22460798174142838,
      "num_tokens": 123914355.0,
      "step": 71440
    },
    {
      "entropy": 5.932128238677978,
      "epoch": 5.558646177786423,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023811701407661995,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21508176922798156,
      "num_tokens": 123923313.0,
      "step": 71445
    },
    {
      "entropy": 5.91254677772522,
      "epoch": 5.55903520715814,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00023808967368019434,
      "loss": 4.97,
      "mean_token_accuracy": 0.2198936551809311,
      "num_tokens": 123931862.0,
      "step": 71450
    },
    {
      "entropy": 5.989532423019409,
      "epoch": 5.559424236529858,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002380623338438242,
      "loss": 4.9754,
      "mean_token_accuracy": 0.219772270321846,
      "num_tokens": 123940813.0,
      "step": 71455
    },
    {
      "entropy": 5.994790506362915,
      "epoch": 5.5598132659015755,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002380349945679244,
      "loss": 4.9915,
      "mean_token_accuracy": 0.2157420828938484,
      "num_tokens": 123950037.0,
      "step": 71460
    },
    {
      "entropy": 5.958417701721191,
      "epoch": 5.560202295273293,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023800765585290967,
      "loss": 4.9772,
      "mean_token_accuracy": 0.21437259912490844,
      "num_tokens": 123958624.0,
      "step": 71465
    },
    {
      "entropy": 5.916739082336425,
      "epoch": 5.560591324645011,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023798031769919488,
      "loss": 4.8062,
      "mean_token_accuracy": 0.22598819434642792,
      "num_tokens": 123967537.0,
      "step": 71470
    },
    {
      "entropy": 5.994444513320923,
      "epoch": 5.560980354016729,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002379529801071949,
      "loss": 4.9401,
      "mean_token_accuracy": 0.22464111000299453,
      "num_tokens": 123976330.0,
      "step": 71475
    },
    {
      "entropy": 5.989505100250244,
      "epoch": 5.561369383388445,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002379256430773244,
      "loss": 4.941,
      "mean_token_accuracy": 0.22138318568468093,
      "num_tokens": 123984549.0,
      "step": 71480
    },
    {
      "entropy": 5.965972900390625,
      "epoch": 5.561758412760163,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023789830660999822,
      "loss": 5.0534,
      "mean_token_accuracy": 0.20846115797758102,
      "num_tokens": 123993441.0,
      "step": 71485
    },
    {
      "entropy": 5.954560232162476,
      "epoch": 5.562147442131881,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002378709707056313,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22652447372674941,
      "num_tokens": 124001631.0,
      "step": 71490
    },
    {
      "entropy": 5.958728170394897,
      "epoch": 5.5625364715035985,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023784363536463827,
      "loss": 4.9532,
      "mean_token_accuracy": 0.2124000981450081,
      "num_tokens": 124010159.0,
      "step": 71495
    },
    {
      "entropy": 6.00135908126831,
      "epoch": 5.562925500875316,
      "grad_norm": 1.5,
      "learning_rate": 0.00023781630058743392,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2214615151286125,
      "num_tokens": 124018881.0,
      "step": 71500
    },
    {
      "entropy": 5.911443376541138,
      "epoch": 5.563314530247034,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002377889663744331,
      "loss": 4.9463,
      "mean_token_accuracy": 0.22110024690628052,
      "num_tokens": 124027528.0,
      "step": 71505
    },
    {
      "entropy": 5.922202014923096,
      "epoch": 5.563703559618752,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002377616327260504,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22080135047435762,
      "num_tokens": 124035733.0,
      "step": 71510
    },
    {
      "entropy": 6.026410102844238,
      "epoch": 5.564092588990468,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023773429964270065,
      "loss": 5.0004,
      "mean_token_accuracy": 0.21989874839782714,
      "num_tokens": 124044875.0,
      "step": 71515
    },
    {
      "entropy": 5.967165946960449,
      "epoch": 5.564481618362186,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002377069671247986,
      "loss": 4.8847,
      "mean_token_accuracy": 0.2225658968091011,
      "num_tokens": 124053223.0,
      "step": 71520
    },
    {
      "entropy": 5.950085210800171,
      "epoch": 5.564870647733904,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023767963517275903,
      "loss": 4.9221,
      "mean_token_accuracy": 0.21639414578676225,
      "num_tokens": 124061739.0,
      "step": 71525
    },
    {
      "entropy": 5.860366916656494,
      "epoch": 5.5652596771056215,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002376523037869965,
      "loss": 4.8219,
      "mean_token_accuracy": 0.23611561357975006,
      "num_tokens": 124070264.0,
      "step": 71530
    },
    {
      "entropy": 5.9864359378814695,
      "epoch": 5.565648706477339,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023762497296792585,
      "loss": 4.9964,
      "mean_token_accuracy": 0.2178785890340805,
      "num_tokens": 124078999.0,
      "step": 71535
    },
    {
      "entropy": 5.989340543746948,
      "epoch": 5.566037735849057,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023759764271596173,
      "loss": 5.033,
      "mean_token_accuracy": 0.2139681950211525,
      "num_tokens": 124088060.0,
      "step": 71540
    },
    {
      "entropy": 5.877062273025513,
      "epoch": 5.566426765220774,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023757031303151888,
      "loss": 4.8453,
      "mean_token_accuracy": 0.2268519416451454,
      "num_tokens": 124096683.0,
      "step": 71545
    },
    {
      "entropy": 5.935090160369873,
      "epoch": 5.566815794592491,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023754298391501195,
      "loss": 4.9294,
      "mean_token_accuracy": 0.22316647320985794,
      "num_tokens": 124104639.0,
      "step": 71550
    },
    {
      "entropy": 5.921485424041748,
      "epoch": 5.567204823964209,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002375156553668556,
      "loss": 4.8652,
      "mean_token_accuracy": 0.21907831579446793,
      "num_tokens": 124112830.0,
      "step": 71555
    },
    {
      "entropy": 5.980562925338745,
      "epoch": 5.567593853335927,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002374883273874645,
      "loss": 5.024,
      "mean_token_accuracy": 0.21733318865299225,
      "num_tokens": 124121109.0,
      "step": 71560
    },
    {
      "entropy": 5.993254041671753,
      "epoch": 5.5679828827076445,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002374609999772534,
      "loss": 4.9299,
      "mean_token_accuracy": 0.2244493305683136,
      "num_tokens": 124130297.0,
      "step": 71565
    },
    {
      "entropy": 6.029901647567749,
      "epoch": 5.568371912079362,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023743367313663682,
      "loss": 4.9665,
      "mean_token_accuracy": 0.21727085411548613,
      "num_tokens": 124138489.0,
      "step": 71570
    },
    {
      "entropy": 5.978700780868531,
      "epoch": 5.56876094145108,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023740634686602945,
      "loss": 4.9527,
      "mean_token_accuracy": 0.2138536974787712,
      "num_tokens": 124146850.0,
      "step": 71575
    },
    {
      "entropy": 5.911853504180908,
      "epoch": 5.569149970822798,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023737902116584598,
      "loss": 4.8997,
      "mean_token_accuracy": 0.2219911575317383,
      "num_tokens": 124155673.0,
      "step": 71580
    },
    {
      "entropy": 5.938500213623047,
      "epoch": 5.569539000194514,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023735169603650104,
      "loss": 4.8231,
      "mean_token_accuracy": 0.23067893832921982,
      "num_tokens": 124164531.0,
      "step": 71585
    },
    {
      "entropy": 5.970232200622559,
      "epoch": 5.569928029566232,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023732437147840906,
      "loss": 4.9169,
      "mean_token_accuracy": 0.21559476107358932,
      "num_tokens": 124173147.0,
      "step": 71590
    },
    {
      "entropy": 5.911458730697632,
      "epoch": 5.57031705893795,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023729704749198482,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2239275321364403,
      "num_tokens": 124182403.0,
      "step": 71595
    },
    {
      "entropy": 5.990103435516358,
      "epoch": 5.5707060883096675,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023726972407764286,
      "loss": 5.0058,
      "mean_token_accuracy": 0.216741643846035,
      "num_tokens": 124191352.0,
      "step": 71600
    },
    {
      "entropy": 5.961564302444458,
      "epoch": 5.571095117681385,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023724240123579782,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22624872475862504,
      "num_tokens": 124199948.0,
      "step": 71605
    },
    {
      "entropy": 5.948271703720093,
      "epoch": 5.571484147053102,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00023721507896686412,
      "loss": 4.8261,
      "mean_token_accuracy": 0.2308749496936798,
      "num_tokens": 124207935.0,
      "step": 71610
    },
    {
      "entropy": 5.907149410247802,
      "epoch": 5.57187317642482,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00023718775727125658,
      "loss": 4.8696,
      "mean_token_accuracy": 0.224971042573452,
      "num_tokens": 124216811.0,
      "step": 71615
    },
    {
      "entropy": 5.919652843475342,
      "epoch": 5.572262205796537,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023716043614938965,
      "loss": 4.9405,
      "mean_token_accuracy": 0.21835584938526154,
      "num_tokens": 124225667.0,
      "step": 71620
    },
    {
      "entropy": 5.94203143119812,
      "epoch": 5.572651235168255,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023713311560167783,
      "loss": 4.9393,
      "mean_token_accuracy": 0.21003020703792571,
      "num_tokens": 124234094.0,
      "step": 71625
    },
    {
      "entropy": 6.03167724609375,
      "epoch": 5.573040264539973,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023710579562853573,
      "loss": 4.9745,
      "mean_token_accuracy": 0.2214861512184143,
      "num_tokens": 124242178.0,
      "step": 71630
    },
    {
      "entropy": 5.961128902435303,
      "epoch": 5.5734292939116905,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00023707847623037783,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2162827581167221,
      "num_tokens": 124250708.0,
      "step": 71635
    },
    {
      "entropy": 5.905124378204346,
      "epoch": 5.573818323283408,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023705115740761874,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22338349521160125,
      "num_tokens": 124259160.0,
      "step": 71640
    },
    {
      "entropy": 5.987390613555908,
      "epoch": 5.574207352655126,
      "grad_norm": 1.375,
      "learning_rate": 0.00023702383916067294,
      "loss": 5.0122,
      "mean_token_accuracy": 0.2139311969280243,
      "num_tokens": 124268211.0,
      "step": 71645
    },
    {
      "entropy": 6.071311378479004,
      "epoch": 5.574596382026843,
      "grad_norm": 1.5,
      "learning_rate": 0.00023699652148995488,
      "loss": 5.0456,
      "mean_token_accuracy": 0.20477796196937562,
      "num_tokens": 124277530.0,
      "step": 71650
    },
    {
      "entropy": 5.903726816177368,
      "epoch": 5.57498541139856,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023696920439587917,
      "loss": 4.8136,
      "mean_token_accuracy": 0.2322118654847145,
      "num_tokens": 124286020.0,
      "step": 71655
    },
    {
      "entropy": 5.995105409622193,
      "epoch": 5.575374440770278,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00023694188787886023,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21706831455230713,
      "num_tokens": 124294498.0,
      "step": 71660
    },
    {
      "entropy": 5.959561824798584,
      "epoch": 5.575763470141996,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002369145719393126,
      "loss": 5.0157,
      "mean_token_accuracy": 0.2133713573217392,
      "num_tokens": 124303160.0,
      "step": 71665
    },
    {
      "entropy": 5.945098829269409,
      "epoch": 5.5761524995137135,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002368872565776507,
      "loss": 4.9916,
      "mean_token_accuracy": 0.21537958979606628,
      "num_tokens": 124312684.0,
      "step": 71670
    },
    {
      "entropy": 5.959701251983643,
      "epoch": 5.576541528885431,
      "grad_norm": 1.609375,
      "learning_rate": 0.00023685994179428887,
      "loss": 4.8778,
      "mean_token_accuracy": 0.23108814507722855,
      "num_tokens": 124321426.0,
      "step": 71675
    },
    {
      "entropy": 5.9943990230560305,
      "epoch": 5.576930558257148,
      "grad_norm": 1.625,
      "learning_rate": 0.00023683262758964185,
      "loss": 4.997,
      "mean_token_accuracy": 0.2153536483645439,
      "num_tokens": 124329842.0,
      "step": 71680
    },
    {
      "entropy": 5.920341157913208,
      "epoch": 5.577319587628866,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023680531396412397,
      "loss": 4.9043,
      "mean_token_accuracy": 0.22194890379905702,
      "num_tokens": 124338906.0,
      "step": 71685
    },
    {
      "entropy": 5.923767137527466,
      "epoch": 5.577708617000583,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00023677800091814966,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21430815011262894,
      "num_tokens": 124346861.0,
      "step": 71690
    },
    {
      "entropy": 5.896508693695068,
      "epoch": 5.578097646372301,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002367506884521334,
      "loss": 4.9511,
      "mean_token_accuracy": 0.21623628437519074,
      "num_tokens": 124356138.0,
      "step": 71695
    },
    {
      "entropy": 5.967844247817993,
      "epoch": 5.578486675744019,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023672337656648951,
      "loss": 4.9159,
      "mean_token_accuracy": 0.22357018738985063,
      "num_tokens": 124365459.0,
      "step": 71700
    },
    {
      "entropy": 5.962654638290405,
      "epoch": 5.5788757051157365,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002366960652616324,
      "loss": 4.9666,
      "mean_token_accuracy": 0.21764759421348573,
      "num_tokens": 124374222.0,
      "step": 71705
    },
    {
      "entropy": 6.005130004882813,
      "epoch": 5.579264734487454,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002366687545379766,
      "loss": 5.043,
      "mean_token_accuracy": 0.2138952061533928,
      "num_tokens": 124383242.0,
      "step": 71710
    },
    {
      "entropy": 5.957384300231934,
      "epoch": 5.579653763859171,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023664144439593644,
      "loss": 4.9966,
      "mean_token_accuracy": 0.2260822609066963,
      "num_tokens": 124392080.0,
      "step": 71715
    },
    {
      "entropy": 5.988266944885254,
      "epoch": 5.580042793230889,
      "grad_norm": 1.5,
      "learning_rate": 0.00023661413483592626,
      "loss": 4.948,
      "mean_token_accuracy": 0.21648921817541122,
      "num_tokens": 124400376.0,
      "step": 71720
    },
    {
      "entropy": 5.966048002243042,
      "epoch": 5.580431822602606,
      "grad_norm": 1.375,
      "learning_rate": 0.00023658682585836049,
      "loss": 5.0265,
      "mean_token_accuracy": 0.21196640878915787,
      "num_tokens": 124409743.0,
      "step": 71725
    },
    {
      "entropy": 5.947758483886719,
      "epoch": 5.580820851974324,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023655951746365356,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21657426059246063,
      "num_tokens": 124418740.0,
      "step": 71730
    },
    {
      "entropy": 6.006952810287475,
      "epoch": 5.581209881346042,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023653220965221973,
      "loss": 4.9914,
      "mean_token_accuracy": 0.21133291572332383,
      "num_tokens": 124426857.0,
      "step": 71735
    },
    {
      "entropy": 5.985870313644409,
      "epoch": 5.5815989107177595,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023650490242447338,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2240399584174156,
      "num_tokens": 124435749.0,
      "step": 71740
    },
    {
      "entropy": 5.965119504928589,
      "epoch": 5.581987940089476,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00023647759578082883,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21598538309335708,
      "num_tokens": 124444697.0,
      "step": 71745
    },
    {
      "entropy": 5.916910839080811,
      "epoch": 5.582376969461194,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023645028972170046,
      "loss": 4.8603,
      "mean_token_accuracy": 0.22530157268047332,
      "num_tokens": 124452889.0,
      "step": 71750
    },
    {
      "entropy": 5.985986518859863,
      "epoch": 5.582765998832912,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023642298424750264,
      "loss": 5.0247,
      "mean_token_accuracy": 0.21501931995153428,
      "num_tokens": 124462321.0,
      "step": 71755
    },
    {
      "entropy": 6.000534105300903,
      "epoch": 5.583155028204629,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00023639567935864958,
      "loss": 4.9849,
      "mean_token_accuracy": 0.21923601180315017,
      "num_tokens": 124470903.0,
      "step": 71760
    },
    {
      "entropy": 5.943886756896973,
      "epoch": 5.583544057576347,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002363683750555557,
      "loss": 5.0358,
      "mean_token_accuracy": 0.212981915473938,
      "num_tokens": 124479903.0,
      "step": 71765
    },
    {
      "entropy": 5.955171680450439,
      "epoch": 5.583933086948065,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002363410713386352,
      "loss": 4.9222,
      "mean_token_accuracy": 0.22077054381370545,
      "num_tokens": 124488114.0,
      "step": 71770
    },
    {
      "entropy": 6.001496124267578,
      "epoch": 5.5843221163197825,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023631376820830248,
      "loss": 4.95,
      "mean_token_accuracy": 0.21600372195243836,
      "num_tokens": 124496994.0,
      "step": 71775
    },
    {
      "entropy": 5.9759202003479,
      "epoch": 5.5847111456915,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002362864656649717,
      "loss": 4.9412,
      "mean_token_accuracy": 0.22384363859891893,
      "num_tokens": 124505678.0,
      "step": 71780
    },
    {
      "entropy": 5.993854093551636,
      "epoch": 5.585100175063217,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023625916370905714,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21485771387815475,
      "num_tokens": 124514675.0,
      "step": 71785
    },
    {
      "entropy": 5.967865657806397,
      "epoch": 5.585489204434935,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023623186234097315,
      "loss": 4.9971,
      "mean_token_accuracy": 0.21939122974872588,
      "num_tokens": 124523405.0,
      "step": 71790
    },
    {
      "entropy": 5.983066129684448,
      "epoch": 5.585878233806652,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023620456156113397,
      "loss": 5.0388,
      "mean_token_accuracy": 0.21103414744138718,
      "num_tokens": 124532100.0,
      "step": 71795
    },
    {
      "entropy": 6.036317539215088,
      "epoch": 5.58626726317837,
      "grad_norm": 1.625,
      "learning_rate": 0.00023617726136995377,
      "loss": 5.0315,
      "mean_token_accuracy": 0.21090627908706666,
      "num_tokens": 124540821.0,
      "step": 71800
    },
    {
      "entropy": 6.011054992675781,
      "epoch": 5.586656292550088,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023614996176784687,
      "loss": 4.9054,
      "mean_token_accuracy": 0.21940771937370301,
      "num_tokens": 124549359.0,
      "step": 71805
    },
    {
      "entropy": 5.964091491699219,
      "epoch": 5.5870453219218055,
      "grad_norm": 1.5625,
      "learning_rate": 0.00023612266275522737,
      "loss": 5.0015,
      "mean_token_accuracy": 0.21439261734485626,
      "num_tokens": 124558888.0,
      "step": 71810
    },
    {
      "entropy": 5.975882863998413,
      "epoch": 5.587434351293522,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00023609536433250966,
      "loss": 4.9324,
      "mean_token_accuracy": 0.2247634083032608,
      "num_tokens": 124567458.0,
      "step": 71815
    },
    {
      "entropy": 5.978480195999145,
      "epoch": 5.58782338066524,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00023606806650010786,
      "loss": 4.9635,
      "mean_token_accuracy": 0.21545154601335526,
      "num_tokens": 124576329.0,
      "step": 71820
    },
    {
      "entropy": 5.946493244171142,
      "epoch": 5.588212410036958,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023604076925843622,
      "loss": 4.9317,
      "mean_token_accuracy": 0.22226136028766633,
      "num_tokens": 124585615.0,
      "step": 71825
    },
    {
      "entropy": 5.915390253067017,
      "epoch": 5.588601439408675,
      "grad_norm": 1.5625,
      "learning_rate": 0.00023601347260790884,
      "loss": 4.8817,
      "mean_token_accuracy": 0.22215299308300018,
      "num_tokens": 124594170.0,
      "step": 71830
    },
    {
      "entropy": 5.989613008499146,
      "epoch": 5.588990468780393,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023598617654894006,
      "loss": 5.0036,
      "mean_token_accuracy": 0.21270175129175187,
      "num_tokens": 124602671.0,
      "step": 71835
    },
    {
      "entropy": 6.010264253616333,
      "epoch": 5.589379498152111,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023595888108194385,
      "loss": 4.9661,
      "mean_token_accuracy": 0.22624698877334595,
      "num_tokens": 124611207.0,
      "step": 71840
    },
    {
      "entropy": 5.964342403411865,
      "epoch": 5.5897685275238285,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002359315862073345,
      "loss": 4.8904,
      "mean_token_accuracy": 0.21458305716514586,
      "num_tokens": 124619490.0,
      "step": 71845
    },
    {
      "entropy": 5.886595773696899,
      "epoch": 5.590157556895545,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002359042919255262,
      "loss": 4.8749,
      "mean_token_accuracy": 0.21880633682012557,
      "num_tokens": 124628330.0,
      "step": 71850
    },
    {
      "entropy": 5.91149525642395,
      "epoch": 5.590546586267263,
      "grad_norm": 1.375,
      "learning_rate": 0.00023587699823693304,
      "loss": 4.9149,
      "mean_token_accuracy": 0.2214462161064148,
      "num_tokens": 124637064.0,
      "step": 71855
    },
    {
      "entropy": 5.947497415542602,
      "epoch": 5.590935615638981,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023584970514196912,
      "loss": 4.9763,
      "mean_token_accuracy": 0.22130292803049087,
      "num_tokens": 124645121.0,
      "step": 71860
    },
    {
      "entropy": 5.952523422241211,
      "epoch": 5.591324645010698,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00023582241264104864,
      "loss": 4.9039,
      "mean_token_accuracy": 0.21909817755222322,
      "num_tokens": 124653541.0,
      "step": 71865
    },
    {
      "entropy": 5.929447364807129,
      "epoch": 5.591713674382416,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023579512073458565,
      "loss": 4.9516,
      "mean_token_accuracy": 0.2154666930437088,
      "num_tokens": 124662010.0,
      "step": 71870
    },
    {
      "entropy": 5.936817026138305,
      "epoch": 5.592102703754134,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023576782942299424,
      "loss": 4.8992,
      "mean_token_accuracy": 0.219418565928936,
      "num_tokens": 124669875.0,
      "step": 71875
    },
    {
      "entropy": 5.969080829620362,
      "epoch": 5.592491733125851,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002357405387066886,
      "loss": 4.9803,
      "mean_token_accuracy": 0.21397658586502075,
      "num_tokens": 124678676.0,
      "step": 71880
    },
    {
      "entropy": 6.0098330020904545,
      "epoch": 5.592880762497568,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002357132485860829,
      "loss": 5.0184,
      "mean_token_accuracy": 0.21298391968011857,
      "num_tokens": 124687508.0,
      "step": 71885
    },
    {
      "entropy": 5.931096696853638,
      "epoch": 5.593269791869286,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023568595906159107,
      "loss": 4.8973,
      "mean_token_accuracy": 0.22850116789340974,
      "num_tokens": 124696080.0,
      "step": 71890
    },
    {
      "entropy": 5.962214469909668,
      "epoch": 5.593658821241004,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023565867013362725,
      "loss": 5.0052,
      "mean_token_accuracy": 0.21833460181951522,
      "num_tokens": 124704294.0,
      "step": 71895
    },
    {
      "entropy": 6.009474229812622,
      "epoch": 5.594047850612721,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023563138180260552,
      "loss": 4.9997,
      "mean_token_accuracy": 0.2181362807750702,
      "num_tokens": 124712893.0,
      "step": 71900
    },
    {
      "entropy": 5.946059608459473,
      "epoch": 5.594436879984439,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023560409406893986,
      "loss": 4.8804,
      "mean_token_accuracy": 0.2253958761692047,
      "num_tokens": 124720697.0,
      "step": 71905
    },
    {
      "entropy": 5.936862373352051,
      "epoch": 5.594825909356157,
      "grad_norm": 1.609375,
      "learning_rate": 0.00023557680693304435,
      "loss": 4.933,
      "mean_token_accuracy": 0.22109376937150954,
      "num_tokens": 124729183.0,
      "step": 71910
    },
    {
      "entropy": 5.989626693725586,
      "epoch": 5.595214938727874,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00023554952039533307,
      "loss": 4.9741,
      "mean_token_accuracy": 0.22106578201055527,
      "num_tokens": 124737995.0,
      "step": 71915
    },
    {
      "entropy": 5.905414772033692,
      "epoch": 5.595603968099591,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023552223445622006,
      "loss": 4.8322,
      "mean_token_accuracy": 0.23181974291801452,
      "num_tokens": 124747090.0,
      "step": 71920
    },
    {
      "entropy": 5.972628164291382,
      "epoch": 5.595992997471309,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023549494911611924,
      "loss": 4.9543,
      "mean_token_accuracy": 0.212968909740448,
      "num_tokens": 124755330.0,
      "step": 71925
    },
    {
      "entropy": 5.927337980270385,
      "epoch": 5.596382026843027,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023546766437544477,
      "loss": 4.8983,
      "mean_token_accuracy": 0.21769261807203294,
      "num_tokens": 124764480.0,
      "step": 71930
    },
    {
      "entropy": 6.003054809570313,
      "epoch": 5.596771056214744,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023544038023461057,
      "loss": 4.9413,
      "mean_token_accuracy": 0.21740732043981553,
      "num_tokens": 124772843.0,
      "step": 71935
    },
    {
      "entropy": 5.928217649459839,
      "epoch": 5.597160085586462,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002354130966940306,
      "loss": 4.8486,
      "mean_token_accuracy": 0.22657188922166824,
      "num_tokens": 124781379.0,
      "step": 71940
    },
    {
      "entropy": 6.013958263397217,
      "epoch": 5.597549114958179,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023538581375411876,
      "loss": 5.0525,
      "mean_token_accuracy": 0.2110696777701378,
      "num_tokens": 124790688.0,
      "step": 71945
    },
    {
      "entropy": 6.013502168655395,
      "epoch": 5.597938144329897,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023535853141528923,
      "loss": 4.9704,
      "mean_token_accuracy": 0.22131878584623338,
      "num_tokens": 124799349.0,
      "step": 71950
    },
    {
      "entropy": 5.947861576080323,
      "epoch": 5.598327173701614,
      "grad_norm": 1.71875,
      "learning_rate": 0.00023533124967795594,
      "loss": 4.8718,
      "mean_token_accuracy": 0.2199745923280716,
      "num_tokens": 124807585.0,
      "step": 71955
    },
    {
      "entropy": 5.990926361083984,
      "epoch": 5.598716203073332,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023530396854253276,
      "loss": 5.0191,
      "mean_token_accuracy": 0.22077550739049911,
      "num_tokens": 124815849.0,
      "step": 71960
    },
    {
      "entropy": 6.03556809425354,
      "epoch": 5.59910523244505,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023527668800943374,
      "loss": 5.0182,
      "mean_token_accuracy": 0.21271554082632066,
      "num_tokens": 124824312.0,
      "step": 71965
    },
    {
      "entropy": 6.039052820205688,
      "epoch": 5.599494261816767,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002352494080790727,
      "loss": 5.0218,
      "mean_token_accuracy": 0.21654243022203445,
      "num_tokens": 124832821.0,
      "step": 71970
    },
    {
      "entropy": 5.958756971359253,
      "epoch": 5.599883291188485,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002352221287518636,
      "loss": 4.9511,
      "mean_token_accuracy": 0.21951511949300767,
      "num_tokens": 124841726.0,
      "step": 71975
    },
    {
      "entropy": 5.94097409248352,
      "epoch": 5.600272320560203,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023519485002822038,
      "loss": 4.8688,
      "mean_token_accuracy": 0.22075520306825638,
      "num_tokens": 124851053.0,
      "step": 71980
    },
    {
      "entropy": 5.993810939788818,
      "epoch": 5.6006613499319196,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023516757190855705,
      "loss": 4.9346,
      "mean_token_accuracy": 0.21474522054195405,
      "num_tokens": 124859657.0,
      "step": 71985
    },
    {
      "entropy": 5.99341311454773,
      "epoch": 5.601050379303637,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023514029439328734,
      "loss": 4.9951,
      "mean_token_accuracy": 0.21699948906898497,
      "num_tokens": 124868938.0,
      "step": 71990
    },
    {
      "entropy": 6.011152696609497,
      "epoch": 5.601439408675355,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002351130174828252,
      "loss": 4.9851,
      "mean_token_accuracy": 0.2157683938741684,
      "num_tokens": 124877104.0,
      "step": 71995
    },
    {
      "entropy": 6.0121537208557125,
      "epoch": 5.601828438047073,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023508574117758453,
      "loss": 4.9603,
      "mean_token_accuracy": 0.2224699079990387,
      "num_tokens": 124885693.0,
      "step": 72000
    },
    {
      "epoch": 5.601828438047073,
      "eval_entropy": 5.676884120651113,
      "eval_loss": 5.04976749420166,
      "eval_mean_token_accuracy": 0.22283494280609978,
      "eval_num_tokens": 124885693.0,
      "eval_runtime": 21.8114,
      "eval_samples_per_second": 1905.335,
      "eval_steps_per_second": 238.178,
      "step": 72000
    },
    {
      "entropy": 5.932283544540406,
      "epoch": 5.60221746741879,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023505846547797928,
      "loss": 5.0017,
      "mean_token_accuracy": 0.2147320106625557,
      "num_tokens": 124894162.0,
      "step": 72005
    },
    {
      "entropy": 5.9542478084564205,
      "epoch": 5.602606496790508,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023503119038442312,
      "loss": 4.8871,
      "mean_token_accuracy": 0.22071647942066192,
      "num_tokens": 124903603.0,
      "step": 72010
    },
    {
      "entropy": 5.936869144439697,
      "epoch": 5.602995526162225,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002350039158973301,
      "loss": 4.8465,
      "mean_token_accuracy": 0.23021106272935868,
      "num_tokens": 124912418.0,
      "step": 72015
    },
    {
      "entropy": 6.0223236083984375,
      "epoch": 5.6033845555339425,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000234976642017114,
      "loss": 5.0263,
      "mean_token_accuracy": 0.21576027423143387,
      "num_tokens": 124920539.0,
      "step": 72020
    },
    {
      "entropy": 5.9755617618560795,
      "epoch": 5.60377358490566,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023494936874418864,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2067431852221489,
      "num_tokens": 124928580.0,
      "step": 72025
    },
    {
      "entropy": 5.933668565750122,
      "epoch": 5.604162614277378,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023492209607896793,
      "loss": 4.855,
      "mean_token_accuracy": 0.22139349728822708,
      "num_tokens": 124937453.0,
      "step": 72030
    },
    {
      "entropy": 5.980785703659057,
      "epoch": 5.604551643649096,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00023489482402186558,
      "loss": 4.9383,
      "mean_token_accuracy": 0.22610079646110534,
      "num_tokens": 124945543.0,
      "step": 72035
    },
    {
      "entropy": 5.99549708366394,
      "epoch": 5.604940673020813,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002348675525732954,
      "loss": 4.9984,
      "mean_token_accuracy": 0.21345447599887848,
      "num_tokens": 124954710.0,
      "step": 72040
    },
    {
      "entropy": 6.005642890930176,
      "epoch": 5.605329702392531,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023484028173367127,
      "loss": 4.9299,
      "mean_token_accuracy": 0.22087893038988113,
      "num_tokens": 124963295.0,
      "step": 72045
    },
    {
      "entropy": 6.059423494338989,
      "epoch": 5.605718731764248,
      "grad_norm": 1.5,
      "learning_rate": 0.000234813011503407,
      "loss": 5.1092,
      "mean_token_accuracy": 0.20318211019039153,
      "num_tokens": 124972117.0,
      "step": 72050
    },
    {
      "entropy": 6.029190158843994,
      "epoch": 5.6061077611359655,
      "grad_norm": 1.609375,
      "learning_rate": 0.00023478574188291623,
      "loss": 5.0191,
      "mean_token_accuracy": 0.21411577612161636,
      "num_tokens": 124980802.0,
      "step": 72055
    },
    {
      "entropy": 6.020864057540893,
      "epoch": 5.606496790507683,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023475847287261282,
      "loss": 4.9441,
      "mean_token_accuracy": 0.22262782752513885,
      "num_tokens": 124989997.0,
      "step": 72060
    },
    {
      "entropy": 6.043002986907959,
      "epoch": 5.606885819879401,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023473120447291057,
      "loss": 4.9472,
      "mean_token_accuracy": 0.21670212596654892,
      "num_tokens": 124998661.0,
      "step": 72065
    },
    {
      "entropy": 5.984840059280396,
      "epoch": 5.607274849251119,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002347039366842232,
      "loss": 4.9379,
      "mean_token_accuracy": 0.2203179657459259,
      "num_tokens": 125007265.0,
      "step": 72070
    },
    {
      "entropy": 5.982944297790527,
      "epoch": 5.607663878622836,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023467666950696442,
      "loss": 4.9688,
      "mean_token_accuracy": 0.21401740312576295,
      "num_tokens": 125017061.0,
      "step": 72075
    },
    {
      "entropy": 5.994307327270508,
      "epoch": 5.608052907994553,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00023464940294154796,
      "loss": 4.9964,
      "mean_token_accuracy": 0.21360096484422683,
      "num_tokens": 125025773.0,
      "step": 72080
    },
    {
      "entropy": 5.941303491592407,
      "epoch": 5.608441937366271,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023462213698838762,
      "loss": 4.8405,
      "mean_token_accuracy": 0.22683088779449462,
      "num_tokens": 125035399.0,
      "step": 72085
    },
    {
      "entropy": 6.026165580749511,
      "epoch": 5.6088309667379885,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023459487164789717,
      "loss": 4.9237,
      "mean_token_accuracy": 0.22086972147226333,
      "num_tokens": 125043924.0,
      "step": 72090
    },
    {
      "entropy": 6.050693798065185,
      "epoch": 5.609219996109706,
      "grad_norm": 1.5,
      "learning_rate": 0.0002345676069204901,
      "loss": 5.0198,
      "mean_token_accuracy": 0.21340699642896652,
      "num_tokens": 125052277.0,
      "step": 72095
    },
    {
      "entropy": 6.020176649093628,
      "epoch": 5.609609025481424,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023454034280658036,
      "loss": 5.0035,
      "mean_token_accuracy": 0.218339903652668,
      "num_tokens": 125061395.0,
      "step": 72100
    },
    {
      "entropy": 5.959775829315186,
      "epoch": 5.609998054853142,
      "grad_norm": 1.609375,
      "learning_rate": 0.00023451307930658145,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22010305523872375,
      "num_tokens": 125070063.0,
      "step": 72105
    },
    {
      "entropy": 5.924965524673462,
      "epoch": 5.610387084224859,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023448581642090716,
      "loss": 4.8369,
      "mean_token_accuracy": 0.2277142435312271,
      "num_tokens": 125078340.0,
      "step": 72110
    },
    {
      "entropy": 5.953242015838623,
      "epoch": 5.610776113596577,
      "grad_norm": 1.5625,
      "learning_rate": 0.00023445855414997108,
      "loss": 4.9541,
      "mean_token_accuracy": 0.22071052938699723,
      "num_tokens": 125087253.0,
      "step": 72115
    },
    {
      "entropy": 5.965456199645996,
      "epoch": 5.611165142968294,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00023443129249418694,
      "loss": 4.9371,
      "mean_token_accuracy": 0.21766656041145324,
      "num_tokens": 125096351.0,
      "step": 72120
    },
    {
      "entropy": 6.015374088287354,
      "epoch": 5.6115541723400115,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023440403145396838,
      "loss": 5.0012,
      "mean_token_accuracy": 0.21136986762285231,
      "num_tokens": 125104573.0,
      "step": 72125
    },
    {
      "entropy": 5.940572023391724,
      "epoch": 5.611943201711729,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023437677102972903,
      "loss": 4.914,
      "mean_token_accuracy": 0.22316266149282454,
      "num_tokens": 125112609.0,
      "step": 72130
    },
    {
      "entropy": 5.943961429595947,
      "epoch": 5.612332231083447,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023434951122188247,
      "loss": 4.9341,
      "mean_token_accuracy": 0.22291814535856247,
      "num_tokens": 125121452.0,
      "step": 72135
    },
    {
      "entropy": 5.958939123153686,
      "epoch": 5.612721260455165,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002343222520308423,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21974676102399826,
      "num_tokens": 125129541.0,
      "step": 72140
    },
    {
      "entropy": 5.9088176727294925,
      "epoch": 5.613110289826882,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002342949934570223,
      "loss": 4.9001,
      "mean_token_accuracy": 0.21167769581079482,
      "num_tokens": 125137604.0,
      "step": 72145
    },
    {
      "entropy": 5.9892223358154295,
      "epoch": 5.613499319198599,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023426773550083602,
      "loss": 4.969,
      "mean_token_accuracy": 0.21296900361776352,
      "num_tokens": 125146863.0,
      "step": 72150
    },
    {
      "entropy": 6.025477409362793,
      "epoch": 5.613888348570317,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023424047816269705,
      "loss": 4.9954,
      "mean_token_accuracy": 0.21871867626905442,
      "num_tokens": 125156126.0,
      "step": 72155
    },
    {
      "entropy": 6.0128502368927,
      "epoch": 5.6142773779420345,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023421322144301889,
      "loss": 5.0132,
      "mean_token_accuracy": 0.21121589243412017,
      "num_tokens": 125165583.0,
      "step": 72160
    },
    {
      "entropy": 5.939020347595215,
      "epoch": 5.614666407313752,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002341859653422152,
      "loss": 4.8423,
      "mean_token_accuracy": 0.22626532912254332,
      "num_tokens": 125174346.0,
      "step": 72165
    },
    {
      "entropy": 5.9511617660522464,
      "epoch": 5.61505543668547,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023415870986069948,
      "loss": 4.8835,
      "mean_token_accuracy": 0.21967267990112305,
      "num_tokens": 125182978.0,
      "step": 72170
    },
    {
      "entropy": 5.986528968811035,
      "epoch": 5.615444466057188,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023413145499888533,
      "loss": 4.9858,
      "mean_token_accuracy": 0.21494191735982895,
      "num_tokens": 125191318.0,
      "step": 72175
    },
    {
      "entropy": 6.000048875808716,
      "epoch": 5.615833495428905,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023410420075718635,
      "loss": 4.9383,
      "mean_token_accuracy": 0.21492552608251572,
      "num_tokens": 125199863.0,
      "step": 72180
    },
    {
      "entropy": 5.932898998260498,
      "epoch": 5.616222524800622,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023407694713601602,
      "loss": 4.9544,
      "mean_token_accuracy": 0.21613789200782776,
      "num_tokens": 125209035.0,
      "step": 72185
    },
    {
      "entropy": 5.957965230941772,
      "epoch": 5.61661155417234,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023404969413578787,
      "loss": 4.9726,
      "mean_token_accuracy": 0.22133685648441315,
      "num_tokens": 125217772.0,
      "step": 72190
    },
    {
      "entropy": 6.0245826721191404,
      "epoch": 5.6170005835440575,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002340224417569154,
      "loss": 4.897,
      "mean_token_accuracy": 0.21534340232610702,
      "num_tokens": 125226420.0,
      "step": 72195
    },
    {
      "entropy": 5.944337272644043,
      "epoch": 5.617389612915775,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002339951899998122,
      "loss": 4.9375,
      "mean_token_accuracy": 0.21636035293340683,
      "num_tokens": 125234437.0,
      "step": 72200
    },
    {
      "entropy": 5.950929164886475,
      "epoch": 5.617778642287493,
      "grad_norm": 1.59375,
      "learning_rate": 0.00023396793886489177,
      "loss": 4.9645,
      "mean_token_accuracy": 0.22458522766828537,
      "num_tokens": 125242672.0,
      "step": 72205
    },
    {
      "entropy": 5.915388059616089,
      "epoch": 5.618167671659211,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023394068835256738,
      "loss": 4.9252,
      "mean_token_accuracy": 0.22042891681194304,
      "num_tokens": 125251407.0,
      "step": 72210
    },
    {
      "entropy": 5.949826669692993,
      "epoch": 5.618556701030927,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023391343846325286,
      "loss": 5.0074,
      "mean_token_accuracy": 0.2167529821395874,
      "num_tokens": 125259329.0,
      "step": 72215
    },
    {
      "entropy": 5.9700507640838625,
      "epoch": 5.618945730402645,
      "grad_norm": 1.609375,
      "learning_rate": 0.00023388618919736143,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21182293444871902,
      "num_tokens": 125268499.0,
      "step": 72220
    },
    {
      "entropy": 6.024780416488648,
      "epoch": 5.619334759774363,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00023385894055530666,
      "loss": 4.9634,
      "mean_token_accuracy": 0.2239742621779442,
      "num_tokens": 125276574.0,
      "step": 72225
    },
    {
      "entropy": 6.007212972640991,
      "epoch": 5.6197237891460805,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023383169253750203,
      "loss": 4.9275,
      "mean_token_accuracy": 0.224239319562912,
      "num_tokens": 125286000.0,
      "step": 72230
    },
    {
      "entropy": 5.883499193191528,
      "epoch": 5.620112818517798,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023380444514436093,
      "loss": 4.8676,
      "mean_token_accuracy": 0.22815196365118026,
      "num_tokens": 125294839.0,
      "step": 72235
    },
    {
      "entropy": 5.887261772155762,
      "epoch": 5.620501847889516,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002337771983762968,
      "loss": 4.8033,
      "mean_token_accuracy": 0.22126456201076508,
      "num_tokens": 125303657.0,
      "step": 72240
    },
    {
      "entropy": 5.93742823600769,
      "epoch": 5.620890877261234,
      "grad_norm": 1.625,
      "learning_rate": 0.00023374995223372308,
      "loss": 4.9091,
      "mean_token_accuracy": 0.22160035520792007,
      "num_tokens": 125312596.0,
      "step": 72245
    },
    {
      "entropy": 5.94787564277649,
      "epoch": 5.62127990663295,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023372270671705313,
      "loss": 4.8995,
      "mean_token_accuracy": 0.22147901356220245,
      "num_tokens": 125321044.0,
      "step": 72250
    },
    {
      "entropy": 5.976075124740601,
      "epoch": 5.621668936004668,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023369546182670044,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21912810504436492,
      "num_tokens": 125329313.0,
      "step": 72255
    },
    {
      "entropy": 5.983640861511231,
      "epoch": 5.622057965376386,
      "grad_norm": 1.609375,
      "learning_rate": 0.00023366821756307837,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21212354749441148,
      "num_tokens": 125337768.0,
      "step": 72260
    },
    {
      "entropy": 6.002986764907837,
      "epoch": 5.6224469947481035,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023364097392660033,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21368985027074813,
      "num_tokens": 125346007.0,
      "step": 72265
    },
    {
      "entropy": 5.964420938491822,
      "epoch": 5.622836024119821,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023361373091767963,
      "loss": 4.9276,
      "mean_token_accuracy": 0.21605204790830612,
      "num_tokens": 125353737.0,
      "step": 72270
    },
    {
      "entropy": 6.041881513595581,
      "epoch": 5.623225053491539,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002335864885367297,
      "loss": 4.9757,
      "mean_token_accuracy": 0.2135939195752144,
      "num_tokens": 125361963.0,
      "step": 72275
    },
    {
      "entropy": 5.969851303100586,
      "epoch": 5.623614082863256,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002335592467841639,
      "loss": 4.9642,
      "mean_token_accuracy": 0.22105509638786316,
      "num_tokens": 125371572.0,
      "step": 72280
    },
    {
      "entropy": 5.909169054031372,
      "epoch": 5.624003112234973,
      "grad_norm": 1.625,
      "learning_rate": 0.00023353200566039557,
      "loss": 4.8335,
      "mean_token_accuracy": 0.237398761510849,
      "num_tokens": 125380350.0,
      "step": 72285
    },
    {
      "entropy": 5.98205451965332,
      "epoch": 5.624392141606691,
      "grad_norm": 1.5,
      "learning_rate": 0.00023350476516583807,
      "loss": 4.963,
      "mean_token_accuracy": 0.21581938862800598,
      "num_tokens": 125389107.0,
      "step": 72290
    },
    {
      "entropy": 5.9235419750213625,
      "epoch": 5.624781170978409,
      "grad_norm": 1.5625,
      "learning_rate": 0.00023347752530090472,
      "loss": 4.811,
      "mean_token_accuracy": 0.2233251675963402,
      "num_tokens": 125397723.0,
      "step": 72295
    },
    {
      "entropy": 5.907048559188842,
      "epoch": 5.6251702003501265,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002334502860660088,
      "loss": 4.8641,
      "mean_token_accuracy": 0.2215641349554062,
      "num_tokens": 125406450.0,
      "step": 72300
    },
    {
      "entropy": 5.897864484786988,
      "epoch": 5.625559229721844,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023342304746156367,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22244266867637635,
      "num_tokens": 125415592.0,
      "step": 72305
    },
    {
      "entropy": 6.000604724884033,
      "epoch": 5.625948259093562,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023339580948798262,
      "loss": 4.9124,
      "mean_token_accuracy": 0.2217046409845352,
      "num_tokens": 125424048.0,
      "step": 72310
    },
    {
      "entropy": 5.99835696220398,
      "epoch": 5.62633728846528,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023336857214567897,
      "loss": 4.9449,
      "mean_token_accuracy": 0.22609375715255736,
      "num_tokens": 125432402.0,
      "step": 72315
    },
    {
      "entropy": 6.005631923675537,
      "epoch": 5.626726317836996,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023334133543506598,
      "loss": 5.0075,
      "mean_token_accuracy": 0.21291139125823974,
      "num_tokens": 125440744.0,
      "step": 72320
    },
    {
      "entropy": 5.997346878051758,
      "epoch": 5.627115347208714,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002333140993565569,
      "loss": 5.0208,
      "mean_token_accuracy": 0.2097142145037651,
      "num_tokens": 125449073.0,
      "step": 72325
    },
    {
      "entropy": 5.927989721298218,
      "epoch": 5.627504376580432,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023328686391056497,
      "loss": 4.8193,
      "mean_token_accuracy": 0.2400944396853447,
      "num_tokens": 125457363.0,
      "step": 72330
    },
    {
      "entropy": 5.960924243927002,
      "epoch": 5.6278934059521495,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023325962909750349,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23475531190633775,
      "num_tokens": 125465823.0,
      "step": 72335
    },
    {
      "entropy": 5.9974676132202145,
      "epoch": 5.628282435323867,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023323239491778564,
      "loss": 4.9344,
      "mean_token_accuracy": 0.2222289815545082,
      "num_tokens": 125474567.0,
      "step": 72340
    },
    {
      "entropy": 5.929247760772705,
      "epoch": 5.628671464695585,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023320516137182484,
      "loss": 5.029,
      "mean_token_accuracy": 0.21580374538898467,
      "num_tokens": 125482761.0,
      "step": 72345
    },
    {
      "entropy": 5.951905584335327,
      "epoch": 5.629060494067302,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023317792846003416,
      "loss": 4.9166,
      "mean_token_accuracy": 0.2155584678053856,
      "num_tokens": 125491443.0,
      "step": 72350
    },
    {
      "entropy": 5.956709146499634,
      "epoch": 5.629449523439019,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023315069618282686,
      "loss": 4.8988,
      "mean_token_accuracy": 0.22171830534934997,
      "num_tokens": 125500038.0,
      "step": 72355
    },
    {
      "entropy": 5.993507146835327,
      "epoch": 5.629838552810737,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002331234645406161,
      "loss": 4.9582,
      "mean_token_accuracy": 0.22184862196445465,
      "num_tokens": 125508310.0,
      "step": 72360
    },
    {
      "entropy": 5.937461948394775,
      "epoch": 5.630227582182455,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023309623353381516,
      "loss": 4.8975,
      "mean_token_accuracy": 0.2235136076807976,
      "num_tokens": 125516717.0,
      "step": 72365
    },
    {
      "entropy": 5.938158464431763,
      "epoch": 5.6306166115541725,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023306900316283713,
      "loss": 4.9048,
      "mean_token_accuracy": 0.22072543501853942,
      "num_tokens": 125525441.0,
      "step": 72370
    },
    {
      "entropy": 5.875360059738159,
      "epoch": 5.63100564092589,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002330417734280953,
      "loss": 4.8202,
      "mean_token_accuracy": 0.22180796563625335,
      "num_tokens": 125533436.0,
      "step": 72375
    },
    {
      "entropy": 6.038035583496094,
      "epoch": 5.631394670297608,
      "grad_norm": 1.5,
      "learning_rate": 0.00023301454433000268,
      "loss": 4.9833,
      "mean_token_accuracy": 0.20745279639959335,
      "num_tokens": 125541876.0,
      "step": 72380
    },
    {
      "entropy": 5.940767049789429,
      "epoch": 5.631783699669325,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002329873158689726,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2292938143014908,
      "num_tokens": 125551027.0,
      "step": 72385
    },
    {
      "entropy": 5.982883405685425,
      "epoch": 5.632172729041042,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023296008804541812,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21341394931077956,
      "num_tokens": 125560282.0,
      "step": 72390
    },
    {
      "entropy": 6.025834751129151,
      "epoch": 5.63256175841276,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002329328608597524,
      "loss": 4.9363,
      "mean_token_accuracy": 0.2222495898604393,
      "num_tokens": 125569999.0,
      "step": 72395
    },
    {
      "entropy": 5.921707725524902,
      "epoch": 5.632950787784478,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023290563431238852,
      "loss": 4.8257,
      "mean_token_accuracy": 0.22903242707252502,
      "num_tokens": 125578183.0,
      "step": 72400
    },
    {
      "entropy": 5.9493790626525875,
      "epoch": 5.6333398171561955,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0002328784084037397,
      "loss": 4.9649,
      "mean_token_accuracy": 0.21857262700796126,
      "num_tokens": 125586303.0,
      "step": 72405
    },
    {
      "entropy": 5.926051616668701,
      "epoch": 5.633728846527913,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00023285118313421887,
      "loss": 4.8858,
      "mean_token_accuracy": 0.22560517638921737,
      "num_tokens": 125594724.0,
      "step": 72410
    },
    {
      "entropy": 5.960558319091797,
      "epoch": 5.63411787589963,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002328239585042393,
      "loss": 4.9557,
      "mean_token_accuracy": 0.218736432492733,
      "num_tokens": 125602552.0,
      "step": 72415
    },
    {
      "entropy": 6.067883014678955,
      "epoch": 5.634506905271348,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023279673451421406,
      "loss": 5.0724,
      "mean_token_accuracy": 0.21245322972536088,
      "num_tokens": 125611775.0,
      "step": 72420
    },
    {
      "entropy": 5.96911358833313,
      "epoch": 5.634895934643065,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023276951116455626,
      "loss": 5.0316,
      "mean_token_accuracy": 0.21345360577106476,
      "num_tokens": 125620629.0,
      "step": 72425
    },
    {
      "entropy": 5.927129125595092,
      "epoch": 5.635284964014783,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023274228845567886,
      "loss": 4.8474,
      "mean_token_accuracy": 0.23167373538017272,
      "num_tokens": 125629749.0,
      "step": 72430
    },
    {
      "entropy": 5.925306797027588,
      "epoch": 5.635673993386501,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023271506638799494,
      "loss": 4.9404,
      "mean_token_accuracy": 0.22155393064022064,
      "num_tokens": 125638752.0,
      "step": 72435
    },
    {
      "entropy": 5.947801303863526,
      "epoch": 5.6360630227582185,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023268784496191758,
      "loss": 4.9126,
      "mean_token_accuracy": 0.21734260618686677,
      "num_tokens": 125647128.0,
      "step": 72440
    },
    {
      "entropy": 6.037830591201782,
      "epoch": 5.636452052129936,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023266062417785988,
      "loss": 4.9747,
      "mean_token_accuracy": 0.22290472388267518,
      "num_tokens": 125655448.0,
      "step": 72445
    },
    {
      "entropy": 5.978285503387451,
      "epoch": 5.636841081501654,
      "grad_norm": 1.5,
      "learning_rate": 0.00023263340403623477,
      "loss": 4.9955,
      "mean_token_accuracy": 0.2092864453792572,
      "num_tokens": 125664460.0,
      "step": 72450
    },
    {
      "entropy": 6.049075412750244,
      "epoch": 5.637230110873371,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002326061845374553,
      "loss": 5.0416,
      "mean_token_accuracy": 0.2111099526286125,
      "num_tokens": 125674710.0,
      "step": 72455
    },
    {
      "entropy": 6.004948091506958,
      "epoch": 5.637619140245088,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023257896568193459,
      "loss": 4.8775,
      "mean_token_accuracy": 0.22306284308433533,
      "num_tokens": 125683097.0,
      "step": 72460
    },
    {
      "entropy": 5.8943016052246096,
      "epoch": 5.638008169616806,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002325517474700855,
      "loss": 4.8825,
      "mean_token_accuracy": 0.2329620748758316,
      "num_tokens": 125692026.0,
      "step": 72465
    },
    {
      "entropy": 5.972401428222656,
      "epoch": 5.638397198988524,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002325245299023211,
      "loss": 4.9676,
      "mean_token_accuracy": 0.22386868596076964,
      "num_tokens": 125701338.0,
      "step": 72470
    },
    {
      "entropy": 5.93399567604065,
      "epoch": 5.6387862283602415,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023249731297905423,
      "loss": 4.9474,
      "mean_token_accuracy": 0.22112037986516953,
      "num_tokens": 125710008.0,
      "step": 72475
    },
    {
      "entropy": 6.0598766803741455,
      "epoch": 5.639175257731958,
      "grad_norm": 1.375,
      "learning_rate": 0.00023247009670069812,
      "loss": 5.01,
      "mean_token_accuracy": 0.21524305790662765,
      "num_tokens": 125719349.0,
      "step": 72480
    },
    {
      "entropy": 5.902631235122681,
      "epoch": 5.639564287103676,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002324428810676656,
      "loss": 4.7537,
      "mean_token_accuracy": 0.2408243179321289,
      "num_tokens": 125728266.0,
      "step": 72485
    },
    {
      "entropy": 5.9141027450561525,
      "epoch": 5.639953316475394,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023241566608036957,
      "loss": 5.0054,
      "mean_token_accuracy": 0.2137009918689728,
      "num_tokens": 125737676.0,
      "step": 72490
    },
    {
      "entropy": 5.900844573974609,
      "epoch": 5.640342345847111,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023238845173922308,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22120697796344757,
      "num_tokens": 125745830.0,
      "step": 72495
    },
    {
      "entropy": 5.9556175708770756,
      "epoch": 5.640731375218829,
      "grad_norm": 1.5625,
      "learning_rate": 0.00023236123804463898,
      "loss": 4.8854,
      "mean_token_accuracy": 0.22525913119316102,
      "num_tokens": 125755006.0,
      "step": 72500
    },
    {
      "entropy": 5.88781418800354,
      "epoch": 5.641120404590547,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023233402499703027,
      "loss": 4.8525,
      "mean_token_accuracy": 0.2285912185907364,
      "num_tokens": 125763887.0,
      "step": 72505
    },
    {
      "entropy": 5.96304178237915,
      "epoch": 5.6415094339622645,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002323068125968098,
      "loss": 4.9318,
      "mean_token_accuracy": 0.22406024038791655,
      "num_tokens": 125772869.0,
      "step": 72510
    },
    {
      "entropy": 5.971529769897461,
      "epoch": 5.641898463333982,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002322796008443905,
      "loss": 4.9351,
      "mean_token_accuracy": 0.22701532542705535,
      "num_tokens": 125781326.0,
      "step": 72515
    },
    {
      "entropy": 6.044512987136841,
      "epoch": 5.642287492705699,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002322523897401852,
      "loss": 4.9878,
      "mean_token_accuracy": 0.21242875307798387,
      "num_tokens": 125790791.0,
      "step": 72520
    },
    {
      "entropy": 6.0123247623443605,
      "epoch": 5.642676522077417,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00023222517928460685,
      "loss": 5.0455,
      "mean_token_accuracy": 0.21348005086183547,
      "num_tokens": 125799549.0,
      "step": 72525
    },
    {
      "entropy": 5.960880851745605,
      "epoch": 5.643065551449134,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023219796947806842,
      "loss": 4.9486,
      "mean_token_accuracy": 0.22254555076360702,
      "num_tokens": 125808779.0,
      "step": 72530
    },
    {
      "entropy": 5.952186155319214,
      "epoch": 5.643454580820852,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00023217076032098255,
      "loss": 4.8903,
      "mean_token_accuracy": 0.2272851973772049,
      "num_tokens": 125817738.0,
      "step": 72535
    },
    {
      "entropy": 5.950487899780273,
      "epoch": 5.64384361019257,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023214355181376222,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21804638653993608,
      "num_tokens": 125826300.0,
      "step": 72540
    },
    {
      "entropy": 5.899617385864258,
      "epoch": 5.6442326395642874,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023211634395682042,
      "loss": 4.8724,
      "mean_token_accuracy": 0.22489833384752272,
      "num_tokens": 125835499.0,
      "step": 72545
    },
    {
      "entropy": 5.960859775543213,
      "epoch": 5.644621668936004,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00023208913675056976,
      "loss": 4.8934,
      "mean_token_accuracy": 0.2238628402352333,
      "num_tokens": 125844313.0,
      "step": 72550
    },
    {
      "entropy": 5.87387981414795,
      "epoch": 5.645010698307722,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002320619301954231,
      "loss": 4.8366,
      "mean_token_accuracy": 0.22894085049629212,
      "num_tokens": 125852741.0,
      "step": 72555
    },
    {
      "entropy": 5.941502475738526,
      "epoch": 5.64539972767944,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002320347242917934,
      "loss": 4.942,
      "mean_token_accuracy": 0.22158868312835694,
      "num_tokens": 125860567.0,
      "step": 72560
    },
    {
      "entropy": 5.908563184738159,
      "epoch": 5.645788757051157,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023200751904009332,
      "loss": 4.9319,
      "mean_token_accuracy": 0.22615647315979004,
      "num_tokens": 125869149.0,
      "step": 72565
    },
    {
      "entropy": 5.923479175567627,
      "epoch": 5.646177786422875,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023198031444073575,
      "loss": 4.8987,
      "mean_token_accuracy": 0.22287586629390715,
      "num_tokens": 125877795.0,
      "step": 72570
    },
    {
      "entropy": 5.9272908687591555,
      "epoch": 5.646566815794593,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023195311049413337,
      "loss": 4.8757,
      "mean_token_accuracy": 0.22326212972402573,
      "num_tokens": 125886399.0,
      "step": 72575
    },
    {
      "entropy": 5.916382455825806,
      "epoch": 5.64695584516631,
      "grad_norm": 1.5,
      "learning_rate": 0.00023192590720069907,
      "loss": 4.8778,
      "mean_token_accuracy": 0.22502481192350388,
      "num_tokens": 125894797.0,
      "step": 72580
    },
    {
      "entropy": 5.935520982742309,
      "epoch": 5.647344874538027,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023189870456084561,
      "loss": 4.9265,
      "mean_token_accuracy": 0.22271972596645356,
      "num_tokens": 125903132.0,
      "step": 72585
    },
    {
      "entropy": 5.922742176055908,
      "epoch": 5.647733903909745,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023187150257498567,
      "loss": 4.985,
      "mean_token_accuracy": 0.21649032086133957,
      "num_tokens": 125911987.0,
      "step": 72590
    },
    {
      "entropy": 5.965186357498169,
      "epoch": 5.648122933281463,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000231844301243532,
      "loss": 4.887,
      "mean_token_accuracy": 0.2257716402411461,
      "num_tokens": 125920166.0,
      "step": 72595
    },
    {
      "entropy": 5.9584949016571045,
      "epoch": 5.64851196265318,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002318171005668974,
      "loss": 4.9463,
      "mean_token_accuracy": 0.2202530711889267,
      "num_tokens": 125928563.0,
      "step": 72600
    },
    {
      "entropy": 5.969449615478515,
      "epoch": 5.648900992024898,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023178990054549447,
      "loss": 4.9451,
      "mean_token_accuracy": 0.2166808307170868,
      "num_tokens": 125937294.0,
      "step": 72605
    },
    {
      "entropy": 5.935456132888794,
      "epoch": 5.649290021396616,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023176270117973607,
      "loss": 4.93,
      "mean_token_accuracy": 0.22122231423854827,
      "num_tokens": 125946278.0,
      "step": 72610
    },
    {
      "entropy": 5.888427734375,
      "epoch": 5.6496790507683325,
      "grad_norm": 1.40625,
      "learning_rate": 0.000231735502470035,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23065802603960037,
      "num_tokens": 125954717.0,
      "step": 72615
    },
    {
      "entropy": 5.935413074493408,
      "epoch": 5.65006808014005,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023170830441680368,
      "loss": 4.954,
      "mean_token_accuracy": 0.21861388981342317,
      "num_tokens": 125962738.0,
      "step": 72620
    },
    {
      "entropy": 5.919857692718506,
      "epoch": 5.650457109511768,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023168110702045496,
      "loss": 4.9319,
      "mean_token_accuracy": 0.22469876259565352,
      "num_tokens": 125972048.0,
      "step": 72625
    },
    {
      "entropy": 6.009675979614258,
      "epoch": 5.650846138883486,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00023165391028140154,
      "loss": 4.9535,
      "mean_token_accuracy": 0.22055282741785048,
      "num_tokens": 125980703.0,
      "step": 72630
    },
    {
      "entropy": 6.042292213439941,
      "epoch": 5.651235168255203,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000231626714200056,
      "loss": 4.9546,
      "mean_token_accuracy": 0.21929629892110825,
      "num_tokens": 125989905.0,
      "step": 72635
    },
    {
      "entropy": 5.944596147537231,
      "epoch": 5.651624197626921,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023159951877683099,
      "loss": 4.9037,
      "mean_token_accuracy": 0.22625365257263183,
      "num_tokens": 125997855.0,
      "step": 72640
    },
    {
      "entropy": 5.939704036712646,
      "epoch": 5.652013226998639,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002315723240121393,
      "loss": 4.9185,
      "mean_token_accuracy": 0.21861237585544585,
      "num_tokens": 126007016.0,
      "step": 72645
    },
    {
      "entropy": 5.938398551940918,
      "epoch": 5.652402256370356,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023154512990639336,
      "loss": 4.9041,
      "mean_token_accuracy": 0.22398823648691177,
      "num_tokens": 126016885.0,
      "step": 72650
    },
    {
      "entropy": 5.961942434310913,
      "epoch": 5.652791285742073,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023151793646000602,
      "loss": 4.8896,
      "mean_token_accuracy": 0.21971345841884612,
      "num_tokens": 126024863.0,
      "step": 72655
    },
    {
      "entropy": 5.967657279968262,
      "epoch": 5.653180315113791,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023149074367338974,
      "loss": 4.9571,
      "mean_token_accuracy": 0.2179378792643547,
      "num_tokens": 126032983.0,
      "step": 72660
    },
    {
      "entropy": 5.981151866912842,
      "epoch": 5.653569344485509,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002314635515469572,
      "loss": 5.0034,
      "mean_token_accuracy": 0.21470242291688918,
      "num_tokens": 126041744.0,
      "step": 72665
    },
    {
      "entropy": 5.948379707336426,
      "epoch": 5.653958373857226,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023143636008112096,
      "loss": 4.8981,
      "mean_token_accuracy": 0.22322926819324493,
      "num_tokens": 126050150.0,
      "step": 72670
    },
    {
      "entropy": 5.986972379684448,
      "epoch": 5.654347403228944,
      "grad_norm": 1.671875,
      "learning_rate": 0.00023140916927629352,
      "loss": 4.9456,
      "mean_token_accuracy": 0.22065083235502242,
      "num_tokens": 126058194.0,
      "step": 72675
    },
    {
      "entropy": 5.9589134693145756,
      "epoch": 5.654736432600662,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002313819791328876,
      "loss": 4.9368,
      "mean_token_accuracy": 0.2183288276195526,
      "num_tokens": 126066882.0,
      "step": 72680
    },
    {
      "entropy": 5.9557366371154785,
      "epoch": 5.6551254619723785,
      "grad_norm": 1.75,
      "learning_rate": 0.00023135478965131574,
      "loss": 4.9787,
      "mean_token_accuracy": 0.21369950473308563,
      "num_tokens": 126075511.0,
      "step": 72685
    },
    {
      "entropy": 5.930550146102905,
      "epoch": 5.655514491344096,
      "grad_norm": 1.5,
      "learning_rate": 0.0002313276008319905,
      "loss": 4.9099,
      "mean_token_accuracy": 0.219523061811924,
      "num_tokens": 126084111.0,
      "step": 72690
    },
    {
      "entropy": 5.97290940284729,
      "epoch": 5.655903520715814,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023130041267532443,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21629708409309387,
      "num_tokens": 126093346.0,
      "step": 72695
    },
    {
      "entropy": 5.972196435928344,
      "epoch": 5.656292550087532,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023127322518173,
      "loss": 4.9362,
      "mean_token_accuracy": 0.2242979884147644,
      "num_tokens": 126102644.0,
      "step": 72700
    },
    {
      "entropy": 5.9347411632537845,
      "epoch": 5.656681579459249,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002312460383516198,
      "loss": 4.9117,
      "mean_token_accuracy": 0.22051851451396942,
      "num_tokens": 126111054.0,
      "step": 72705
    },
    {
      "entropy": 5.99338002204895,
      "epoch": 5.657070608830967,
      "grad_norm": 1.5,
      "learning_rate": 0.00023121885218540627,
      "loss": 4.9385,
      "mean_token_accuracy": 0.2115892291069031,
      "num_tokens": 126120108.0,
      "step": 72710
    },
    {
      "entropy": 5.907380771636963,
      "epoch": 5.657459638202685,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023119166668350205,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22198890894651413,
      "num_tokens": 126128731.0,
      "step": 72715
    },
    {
      "entropy": 5.960259294509887,
      "epoch": 5.6578486675744015,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002311644818463195,
      "loss": 4.9555,
      "mean_token_accuracy": 0.21847422569990158,
      "num_tokens": 126137299.0,
      "step": 72720
    },
    {
      "entropy": 5.917295837402344,
      "epoch": 5.658237696946119,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002311372976742712,
      "loss": 4.954,
      "mean_token_accuracy": 0.2151455119252205,
      "num_tokens": 126145858.0,
      "step": 72725
    },
    {
      "entropy": 5.859869432449341,
      "epoch": 5.658626726317837,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002311101141677696,
      "loss": 4.8715,
      "mean_token_accuracy": 0.2240202933549881,
      "num_tokens": 126154397.0,
      "step": 72730
    },
    {
      "entropy": 6.0130939960479735,
      "epoch": 5.659015755689555,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00023108293132722714,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21839783042669297,
      "num_tokens": 126163518.0,
      "step": 72735
    },
    {
      "entropy": 6.024412345886231,
      "epoch": 5.659404785061272,
      "grad_norm": 1.59375,
      "learning_rate": 0.00023105574915305627,
      "loss": 5.0041,
      "mean_token_accuracy": 0.21046695113182068,
      "num_tokens": 126171982.0,
      "step": 72740
    },
    {
      "entropy": 5.9236854076385494,
      "epoch": 5.65979381443299,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0002310285676456695,
      "loss": 4.9198,
      "mean_token_accuracy": 0.2125382497906685,
      "num_tokens": 126181320.0,
      "step": 72745
    },
    {
      "entropy": 5.949506378173828,
      "epoch": 5.660182843804707,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023100138680547921,
      "loss": 4.9429,
      "mean_token_accuracy": 0.22496840059757234,
      "num_tokens": 126189949.0,
      "step": 72750
    },
    {
      "entropy": 5.9301587581634525,
      "epoch": 5.6605718731764245,
      "grad_norm": 1.640625,
      "learning_rate": 0.00023097420663289786,
      "loss": 4.8921,
      "mean_token_accuracy": 0.22438059747219086,
      "num_tokens": 126197555.0,
      "step": 72755
    },
    {
      "entropy": 5.95957350730896,
      "epoch": 5.660960902548142,
      "grad_norm": 1.640625,
      "learning_rate": 0.00023094702712833786,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21903351694345474,
      "num_tokens": 126205853.0,
      "step": 72760
    },
    {
      "entropy": 5.935816240310669,
      "epoch": 5.66134993191986,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023091984829221157,
      "loss": 4.9377,
      "mean_token_accuracy": 0.2224680468440056,
      "num_tokens": 126214622.0,
      "step": 72765
    },
    {
      "entropy": 5.962105083465576,
      "epoch": 5.661738961291578,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023089267012493141,
      "loss": 4.9467,
      "mean_token_accuracy": 0.21980153024196625,
      "num_tokens": 126223014.0,
      "step": 72770
    },
    {
      "entropy": 5.9509642124176025,
      "epoch": 5.662127990663295,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023086549262690993,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21574712097644805,
      "num_tokens": 126231591.0,
      "step": 72775
    },
    {
      "entropy": 5.924649715423584,
      "epoch": 5.662517020035013,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023083831579855918,
      "loss": 4.8804,
      "mean_token_accuracy": 0.2249688372015953,
      "num_tokens": 126239982.0,
      "step": 72780
    },
    {
      "entropy": 6.028954982757568,
      "epoch": 5.662906049406731,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023081113964029177,
      "loss": 5.0196,
      "mean_token_accuracy": 0.2145646870136261,
      "num_tokens": 126248451.0,
      "step": 72785
    },
    {
      "entropy": 5.912058925628662,
      "epoch": 5.6632950787784475,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023078396415251995,
      "loss": 4.8909,
      "mean_token_accuracy": 0.2184933677315712,
      "num_tokens": 126257094.0,
      "step": 72790
    },
    {
      "entropy": 5.978049373626709,
      "epoch": 5.663684108150165,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00023075678933565615,
      "loss": 4.9998,
      "mean_token_accuracy": 0.2084408223628998,
      "num_tokens": 126266143.0,
      "step": 72795
    },
    {
      "entropy": 5.955077314376831,
      "epoch": 5.664073137521883,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002307296151901126,
      "loss": 4.9556,
      "mean_token_accuracy": 0.22117069363594055,
      "num_tokens": 126275142.0,
      "step": 72800
    },
    {
      "entropy": 6.03149642944336,
      "epoch": 5.664462166893601,
      "grad_norm": 1.625,
      "learning_rate": 0.0002307024417163016,
      "loss": 5.1248,
      "mean_token_accuracy": 0.21094515323638915,
      "num_tokens": 126284414.0,
      "step": 72805
    },
    {
      "entropy": 6.1144472599029545,
      "epoch": 5.664851196265318,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023067526891463563,
      "loss": 5.0786,
      "mean_token_accuracy": 0.21219465583562852,
      "num_tokens": 126293230.0,
      "step": 72810
    },
    {
      "entropy": 6.0772865295410154,
      "epoch": 5.665240225637035,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002306480967855269,
      "loss": 5.0817,
      "mean_token_accuracy": 0.20350730419158936,
      "num_tokens": 126302514.0,
      "step": 72815
    },
    {
      "entropy": 5.9674536228179935,
      "epoch": 5.665629255008753,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023062092532938772,
      "loss": 4.8386,
      "mean_token_accuracy": 0.23173818737268448,
      "num_tokens": 126310980.0,
      "step": 72820
    },
    {
      "entropy": 5.914691591262818,
      "epoch": 5.6660182843804705,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023059375454663033,
      "loss": 4.8306,
      "mean_token_accuracy": 0.230175244808197,
      "num_tokens": 126320581.0,
      "step": 72825
    },
    {
      "entropy": 5.931934785842896,
      "epoch": 5.666407313752188,
      "grad_norm": 1.625,
      "learning_rate": 0.0002305665844376671,
      "loss": 4.85,
      "mean_token_accuracy": 0.2237171486020088,
      "num_tokens": 126328988.0,
      "step": 72830
    },
    {
      "entropy": 5.945426368713379,
      "epoch": 5.666796343123906,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023053941500291014,
      "loss": 4.9848,
      "mean_token_accuracy": 0.21202347725629805,
      "num_tokens": 126339010.0,
      "step": 72835
    },
    {
      "entropy": 5.95056939125061,
      "epoch": 5.667185372495624,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023051224624277185,
      "loss": 4.8689,
      "mean_token_accuracy": 0.22332751452922822,
      "num_tokens": 126347249.0,
      "step": 72840
    },
    {
      "entropy": 6.002225923538208,
      "epoch": 5.667574401867341,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002304850781576644,
      "loss": 4.9945,
      "mean_token_accuracy": 0.21101197004318237,
      "num_tokens": 126355082.0,
      "step": 72845
    },
    {
      "entropy": 5.985752534866333,
      "epoch": 5.667963431239059,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00023045791074800004,
      "loss": 4.9607,
      "mean_token_accuracy": 0.2130718007683754,
      "num_tokens": 126363698.0,
      "step": 72850
    },
    {
      "entropy": 5.962158250808716,
      "epoch": 5.668352460610776,
      "grad_norm": 1.5625,
      "learning_rate": 0.000230430744014191,
      "loss": 4.9502,
      "mean_token_accuracy": 0.21430813223123552,
      "num_tokens": 126373615.0,
      "step": 72855
    },
    {
      "entropy": 5.898162698745727,
      "epoch": 5.6687414899824935,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023040357795664945,
      "loss": 4.8596,
      "mean_token_accuracy": 0.22666718512773515,
      "num_tokens": 126383311.0,
      "step": 72860
    },
    {
      "entropy": 5.990855884552002,
      "epoch": 5.669130519354211,
      "grad_norm": 1.703125,
      "learning_rate": 0.00023037641257578768,
      "loss": 4.9277,
      "mean_token_accuracy": 0.2133368134498596,
      "num_tokens": 126391228.0,
      "step": 72865
    },
    {
      "entropy": 5.9140480041503904,
      "epoch": 5.669519548725929,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023034924787201778,
      "loss": 4.9465,
      "mean_token_accuracy": 0.21369985044002532,
      "num_tokens": 126400123.0,
      "step": 72870
    },
    {
      "entropy": 5.9018913269042965,
      "epoch": 5.669908578097647,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023032208384575188,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22617561370134354,
      "num_tokens": 126409410.0,
      "step": 72875
    },
    {
      "entropy": 5.882086181640625,
      "epoch": 5.670297607469364,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023029492049740232,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22622902691364288,
      "num_tokens": 126418906.0,
      "step": 72880
    },
    {
      "entropy": 5.964866828918457,
      "epoch": 5.670686636841081,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00023026775782738123,
      "loss": 4.9697,
      "mean_token_accuracy": 0.21480847299098968,
      "num_tokens": 126426780.0,
      "step": 72885
    },
    {
      "entropy": 5.980423498153686,
      "epoch": 5.671075666212799,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002302405958361007,
      "loss": 5.0332,
      "mean_token_accuracy": 0.2138255313038826,
      "num_tokens": 126435313.0,
      "step": 72890
    },
    {
      "entropy": 5.871739387512207,
      "epoch": 5.6714646955845165,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023021343452397287,
      "loss": 4.8267,
      "mean_token_accuracy": 0.2279508590698242,
      "num_tokens": 126443921.0,
      "step": 72895
    },
    {
      "entropy": 5.934856843948364,
      "epoch": 5.671853724956234,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023018627389140989,
      "loss": 4.9407,
      "mean_token_accuracy": 0.22341656386852266,
      "num_tokens": 126452988.0,
      "step": 72900
    },
    {
      "entropy": 6.001782464981079,
      "epoch": 5.672242754327952,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002301591139388239,
      "loss": 4.9684,
      "mean_token_accuracy": 0.22213578522205352,
      "num_tokens": 126461552.0,
      "step": 72905
    },
    {
      "entropy": 5.966126871109009,
      "epoch": 5.67263178369967,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00023013195466662696,
      "loss": 4.9481,
      "mean_token_accuracy": 0.21851833313703536,
      "num_tokens": 126470406.0,
      "step": 72910
    },
    {
      "entropy": 5.980513954162598,
      "epoch": 5.673020813071387,
      "grad_norm": 1.53125,
      "learning_rate": 0.00023010479607523123,
      "loss": 4.9721,
      "mean_token_accuracy": 0.21908336579799653,
      "num_tokens": 126478082.0,
      "step": 72915
    },
    {
      "entropy": 5.959233188629151,
      "epoch": 5.673409842443104,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00023007763816504866,
      "loss": 4.9583,
      "mean_token_accuracy": 0.21539176404476165,
      "num_tokens": 126487298.0,
      "step": 72920
    },
    {
      "entropy": 5.896697235107422,
      "epoch": 5.673798871814822,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023005048093649157,
      "loss": 4.8513,
      "mean_token_accuracy": 0.2270795598626137,
      "num_tokens": 126495977.0,
      "step": 72925
    },
    {
      "entropy": 5.9429123878479,
      "epoch": 5.6741879011865395,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00023002332438997174,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21777993589639663,
      "num_tokens": 126504497.0,
      "step": 72930
    },
    {
      "entropy": 5.904701614379883,
      "epoch": 5.674576930558257,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022999616852590148,
      "loss": 4.8181,
      "mean_token_accuracy": 0.23265775591135024,
      "num_tokens": 126513563.0,
      "step": 72935
    },
    {
      "entropy": 5.970194578170776,
      "epoch": 5.674965959929975,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022996901334469268,
      "loss": 4.8649,
      "mean_token_accuracy": 0.21541497409343718,
      "num_tokens": 126521668.0,
      "step": 72940
    },
    {
      "entropy": 5.960848045349121,
      "epoch": 5.675354989301693,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022994185884675738,
      "loss": 4.9657,
      "mean_token_accuracy": 0.225208979845047,
      "num_tokens": 126530452.0,
      "step": 72945
    },
    {
      "entropy": 5.974166536331177,
      "epoch": 5.675744018673409,
      "grad_norm": 1.5,
      "learning_rate": 0.00022991470503250778,
      "loss": 4.9628,
      "mean_token_accuracy": 0.21377215087413787,
      "num_tokens": 126539409.0,
      "step": 72950
    },
    {
      "entropy": 5.962796926498413,
      "epoch": 5.676133048045127,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022988755190235573,
      "loss": 4.8707,
      "mean_token_accuracy": 0.2269235610961914,
      "num_tokens": 126547620.0,
      "step": 72955
    },
    {
      "entropy": 5.977900695800781,
      "epoch": 5.676522077416845,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022986039945671323,
      "loss": 4.9807,
      "mean_token_accuracy": 0.2160336822271347,
      "num_tokens": 126556141.0,
      "step": 72960
    },
    {
      "entropy": 5.966115188598633,
      "epoch": 5.6769111067885625,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022983324769599239,
      "loss": 4.976,
      "mean_token_accuracy": 0.22178489118814468,
      "num_tokens": 126564687.0,
      "step": 72965
    },
    {
      "entropy": 5.961482381820678,
      "epoch": 5.67730013616028,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002298060966206051,
      "loss": 5.0168,
      "mean_token_accuracy": 0.2112610563635826,
      "num_tokens": 126573227.0,
      "step": 72970
    },
    {
      "entropy": 6.022966146469116,
      "epoch": 5.677689165531998,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022977894623096336,
      "loss": 5.0309,
      "mean_token_accuracy": 0.20791575461626052,
      "num_tokens": 126581939.0,
      "step": 72975
    },
    {
      "entropy": 6.043413686752319,
      "epoch": 5.678078194903716,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022975179652747908,
      "loss": 5.0642,
      "mean_token_accuracy": 0.20750347375869752,
      "num_tokens": 126591341.0,
      "step": 72980
    },
    {
      "entropy": 6.003195476531983,
      "epoch": 5.678467224275433,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022972464751056432,
      "loss": 5.028,
      "mean_token_accuracy": 0.21449422389268874,
      "num_tokens": 126600791.0,
      "step": 72985
    },
    {
      "entropy": 5.928032732009887,
      "epoch": 5.67885625364715,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022969749918063092,
      "loss": 4.9311,
      "mean_token_accuracy": 0.21313513517379762,
      "num_tokens": 126610725.0,
      "step": 72990
    },
    {
      "entropy": 5.964128065109253,
      "epoch": 5.679245283018868,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022967035153809097,
      "loss": 4.917,
      "mean_token_accuracy": 0.21796328574419022,
      "num_tokens": 126619478.0,
      "step": 72995
    },
    {
      "entropy": 6.014876508712769,
      "epoch": 5.6796343123905855,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002296432045833562,
      "loss": 4.9096,
      "mean_token_accuracy": 0.22569421380758287,
      "num_tokens": 126627304.0,
      "step": 73000
    },
    {
      "entropy": 5.965623569488526,
      "epoch": 5.680023341762303,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022961605831683856,
      "loss": 4.9555,
      "mean_token_accuracy": 0.2218638077378273,
      "num_tokens": 126635965.0,
      "step": 73005
    },
    {
      "entropy": 5.993376111984253,
      "epoch": 5.680412371134021,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022958891273895004,
      "loss": 5.0607,
      "mean_token_accuracy": 0.2095496326684952,
      "num_tokens": 126645452.0,
      "step": 73010
    },
    {
      "entropy": 6.004300212860107,
      "epoch": 5.680801400505739,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022956176785010252,
      "loss": 4.9425,
      "mean_token_accuracy": 0.22099160701036452,
      "num_tokens": 126654032.0,
      "step": 73015
    },
    {
      "entropy": 6.013660669326782,
      "epoch": 5.681190429877455,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002295346236507079,
      "loss": 5.0768,
      "mean_token_accuracy": 0.21135346740484237,
      "num_tokens": 126662479.0,
      "step": 73020
    },
    {
      "entropy": 5.909350919723511,
      "epoch": 5.681579459249173,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002295074801411779,
      "loss": 4.8355,
      "mean_token_accuracy": 0.2201574131846428,
      "num_tokens": 126670218.0,
      "step": 73025
    },
    {
      "entropy": 5.976257228851319,
      "epoch": 5.681968488620891,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022948033732192452,
      "loss": 4.9807,
      "mean_token_accuracy": 0.21574137806892396,
      "num_tokens": 126678852.0,
      "step": 73030
    },
    {
      "entropy": 6.0373852252960205,
      "epoch": 5.6823575179926085,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022945319519335962,
      "loss": 5.0207,
      "mean_token_accuracy": 0.21247072368860245,
      "num_tokens": 126687909.0,
      "step": 73035
    },
    {
      "entropy": 5.9868968486785885,
      "epoch": 5.682746547364326,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002294260537558949,
      "loss": 5.0004,
      "mean_token_accuracy": 0.21822952926158906,
      "num_tokens": 126696476.0,
      "step": 73040
    },
    {
      "entropy": 6.05645260810852,
      "epoch": 5.683135576736044,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022939891300994238,
      "loss": 5.0287,
      "mean_token_accuracy": 0.2070160076022148,
      "num_tokens": 126705302.0,
      "step": 73045
    },
    {
      "entropy": 6.037787771224975,
      "epoch": 5.683524606107762,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002293717729559137,
      "loss": 4.9794,
      "mean_token_accuracy": 0.21440818756818772,
      "num_tokens": 126714201.0,
      "step": 73050
    },
    {
      "entropy": 6.099712991714478,
      "epoch": 5.683913635479478,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022934463359422074,
      "loss": 5.1083,
      "mean_token_accuracy": 0.20904088020324707,
      "num_tokens": 126722118.0,
      "step": 73055
    },
    {
      "entropy": 5.959718465805054,
      "epoch": 5.684302664851196,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002293174949252753,
      "loss": 4.9153,
      "mean_token_accuracy": 0.22269941866397858,
      "num_tokens": 126730894.0,
      "step": 73060
    },
    {
      "entropy": 5.983993339538574,
      "epoch": 5.684691694222914,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002292903569494892,
      "loss": 4.9546,
      "mean_token_accuracy": 0.2162748247385025,
      "num_tokens": 126739255.0,
      "step": 73065
    },
    {
      "entropy": 5.962177371978759,
      "epoch": 5.6850807235946315,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022926321966727404,
      "loss": 4.953,
      "mean_token_accuracy": 0.21684591025114058,
      "num_tokens": 126747978.0,
      "step": 73070
    },
    {
      "entropy": 5.937758255004883,
      "epoch": 5.685469752966349,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022923608307904192,
      "loss": 4.8708,
      "mean_token_accuracy": 0.22621162831783295,
      "num_tokens": 126755677.0,
      "step": 73075
    },
    {
      "entropy": 5.961876964569091,
      "epoch": 5.685858782338067,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002292089471852043,
      "loss": 4.9695,
      "mean_token_accuracy": 0.22240551561117172,
      "num_tokens": 126764185.0,
      "step": 73080
    },
    {
      "entropy": 5.954472351074219,
      "epoch": 5.686247811709784,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022918181198617306,
      "loss": 5.0049,
      "mean_token_accuracy": 0.21714649349451065,
      "num_tokens": 126773895.0,
      "step": 73085
    },
    {
      "entropy": 6.056010055541992,
      "epoch": 5.686636841081501,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022915467748235984,
      "loss": 5.0255,
      "mean_token_accuracy": 0.21674968898296357,
      "num_tokens": 126782336.0,
      "step": 73090
    },
    {
      "entropy": 6.136869955062866,
      "epoch": 5.687025870453219,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022912754367417648,
      "loss": 5.133,
      "mean_token_accuracy": 0.19777119606733323,
      "num_tokens": 126790050.0,
      "step": 73095
    },
    {
      "entropy": 6.092506837844849,
      "epoch": 5.687414899824937,
      "grad_norm": 1.59375,
      "learning_rate": 0.00022910041056203458,
      "loss": 4.9572,
      "mean_token_accuracy": 0.21926980167627336,
      "num_tokens": 126797819.0,
      "step": 73100
    },
    {
      "entropy": 5.990675830841065,
      "epoch": 5.6878039291966545,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022907327814634593,
      "loss": 4.9768,
      "mean_token_accuracy": 0.22129994481801987,
      "num_tokens": 126806620.0,
      "step": 73105
    },
    {
      "entropy": 5.9328569889068605,
      "epoch": 5.688192958568372,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022904614642752218,
      "loss": 5.0313,
      "mean_token_accuracy": 0.2160789266228676,
      "num_tokens": 126815524.0,
      "step": 73110
    },
    {
      "entropy": 5.916598987579346,
      "epoch": 5.68858198794009,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022901901540597498,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22628319561481475,
      "num_tokens": 126824184.0,
      "step": 73115
    },
    {
      "entropy": 5.9809126377105715,
      "epoch": 5.688971017311807,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002289918850821161,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22222674787044525,
      "num_tokens": 126832455.0,
      "step": 73120
    },
    {
      "entropy": 6.012555742263794,
      "epoch": 5.689360046683524,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022896475545635704,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2232738807797432,
      "num_tokens": 126841816.0,
      "step": 73125
    },
    {
      "entropy": 5.9670123100280765,
      "epoch": 5.689749076055242,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00022893762652910965,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23527687042951584,
      "num_tokens": 126850237.0,
      "step": 73130
    },
    {
      "entropy": 5.909859037399292,
      "epoch": 5.69013810542696,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022891049830078543,
      "loss": 4.8728,
      "mean_token_accuracy": 0.217604660987854,
      "num_tokens": 126858995.0,
      "step": 73135
    },
    {
      "entropy": 5.9720470905303955,
      "epoch": 5.6905271347986774,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022888337077179595,
      "loss": 5.0006,
      "mean_token_accuracy": 0.21161896884441375,
      "num_tokens": 126867942.0,
      "step": 73140
    },
    {
      "entropy": 5.91645393371582,
      "epoch": 5.690916164170395,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00022885624394255294,
      "loss": 4.8881,
      "mean_token_accuracy": 0.23157857358455658,
      "num_tokens": 126877266.0,
      "step": 73145
    },
    {
      "entropy": 6.023869180679322,
      "epoch": 5.691305193542112,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022882911781346805,
      "loss": 5.0166,
      "mean_token_accuracy": 0.2152101367712021,
      "num_tokens": 126885967.0,
      "step": 73150
    },
    {
      "entropy": 5.969755697250366,
      "epoch": 5.69169422291383,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002288019923849528,
      "loss": 4.8543,
      "mean_token_accuracy": 0.22965194433927535,
      "num_tokens": 126893971.0,
      "step": 73155
    },
    {
      "entropy": 5.989588689804077,
      "epoch": 5.692083252285547,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002287748676574188,
      "loss": 4.959,
      "mean_token_accuracy": 0.22332651764154435,
      "num_tokens": 126902343.0,
      "step": 73160
    },
    {
      "entropy": 5.96540207862854,
      "epoch": 5.692472281657265,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002287477436312776,
      "loss": 4.9299,
      "mean_token_accuracy": 0.2194982185959816,
      "num_tokens": 126910757.0,
      "step": 73165
    },
    {
      "entropy": 6.010799551010132,
      "epoch": 5.692861311028983,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022872062030694073,
      "loss": 5.0959,
      "mean_token_accuracy": 0.20197096914052964,
      "num_tokens": 126919393.0,
      "step": 73170
    },
    {
      "entropy": 6.001668787002563,
      "epoch": 5.6932503404007,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002286934976848198,
      "loss": 4.9091,
      "mean_token_accuracy": 0.22211263179779053,
      "num_tokens": 126927867.0,
      "step": 73175
    },
    {
      "entropy": 5.983290433883667,
      "epoch": 5.693639369772418,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022866637576532633,
      "loss": 4.9039,
      "mean_token_accuracy": 0.22412295192480086,
      "num_tokens": 126936280.0,
      "step": 73180
    },
    {
      "entropy": 5.95237455368042,
      "epoch": 5.694028399144136,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002286392545488719,
      "loss": 4.9219,
      "mean_token_accuracy": 0.22013919055461884,
      "num_tokens": 126945248.0,
      "step": 73185
    },
    {
      "entropy": 5.89198637008667,
      "epoch": 5.694417428515853,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022861213403586802,
      "loss": 4.9025,
      "mean_token_accuracy": 0.2262927323579788,
      "num_tokens": 126953876.0,
      "step": 73190
    },
    {
      "entropy": 6.022714281082154,
      "epoch": 5.69480645788757,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022858501422672606,
      "loss": 5.034,
      "mean_token_accuracy": 0.21607715785503387,
      "num_tokens": 126962267.0,
      "step": 73195
    },
    {
      "entropy": 6.000125598907471,
      "epoch": 5.695195487259288,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002285578951218577,
      "loss": 4.9331,
      "mean_token_accuracy": 0.22253677994012833,
      "num_tokens": 126970530.0,
      "step": 73200
    },
    {
      "entropy": 5.917522811889649,
      "epoch": 5.695584516631006,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00022853077672167432,
      "loss": 4.8926,
      "mean_token_accuracy": 0.2170576795935631,
      "num_tokens": 126978772.0,
      "step": 73205
    },
    {
      "entropy": 5.963950300216675,
      "epoch": 5.695973546002723,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002285036590265875,
      "loss": 4.9425,
      "mean_token_accuracy": 0.22397357672452928,
      "num_tokens": 126986540.0,
      "step": 73210
    },
    {
      "entropy": 5.971228551864624,
      "epoch": 5.696362575374441,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022847654203700857,
      "loss": 5.0257,
      "mean_token_accuracy": 0.2144510805606842,
      "num_tokens": 126995010.0,
      "step": 73215
    },
    {
      "entropy": 5.931143808364868,
      "epoch": 5.696751604746158,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022844942575334914,
      "loss": 4.8907,
      "mean_token_accuracy": 0.22319256961345674,
      "num_tokens": 127003366.0,
      "step": 73220
    },
    {
      "entropy": 6.003247022628784,
      "epoch": 5.697140634117876,
      "grad_norm": 1.671875,
      "learning_rate": 0.00022842231017602055,
      "loss": 4.9277,
      "mean_token_accuracy": 0.22666798681020736,
      "num_tokens": 127012593.0,
      "step": 73225
    },
    {
      "entropy": 5.947619771957397,
      "epoch": 5.697529663489593,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022839519530543423,
      "loss": 4.949,
      "mean_token_accuracy": 0.21606281846761705,
      "num_tokens": 127021023.0,
      "step": 73230
    },
    {
      "entropy": 5.972163963317871,
      "epoch": 5.697918692861311,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0002283680811420017,
      "loss": 4.9533,
      "mean_token_accuracy": 0.22022514492273332,
      "num_tokens": 127028958.0,
      "step": 73235
    },
    {
      "entropy": 5.950520753860474,
      "epoch": 5.698307722233029,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022834096768613437,
      "loss": 4.8658,
      "mean_token_accuracy": 0.2287825360894203,
      "num_tokens": 127038041.0,
      "step": 73240
    },
    {
      "entropy": 6.025143718719482,
      "epoch": 5.698696751604746,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022831385493824353,
      "loss": 5.0295,
      "mean_token_accuracy": 0.21622103452682495,
      "num_tokens": 127046612.0,
      "step": 73245
    },
    {
      "entropy": 6.051503610610962,
      "epoch": 5.699085780976464,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00022828674289874064,
      "loss": 5.1075,
      "mean_token_accuracy": 0.2133649617433548,
      "num_tokens": 127055019.0,
      "step": 73250
    },
    {
      "entropy": 6.030028915405273,
      "epoch": 5.699474810348181,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00022825963156803704,
      "loss": 5.0403,
      "mean_token_accuracy": 0.20574882626533508,
      "num_tokens": 127063108.0,
      "step": 73255
    },
    {
      "entropy": 5.993446493148804,
      "epoch": 5.699863839719899,
      "grad_norm": 1.59375,
      "learning_rate": 0.00022823252094654413,
      "loss": 4.913,
      "mean_token_accuracy": 0.21809035241603852,
      "num_tokens": 127071385.0,
      "step": 73260
    },
    {
      "entropy": 5.931715202331543,
      "epoch": 5.700252869091616,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002282054110346733,
      "loss": 4.8404,
      "mean_token_accuracy": 0.22022452652454377,
      "num_tokens": 127080144.0,
      "step": 73265
    },
    {
      "entropy": 6.001713228225708,
      "epoch": 5.700641898463334,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022817830183283582,
      "loss": 4.9586,
      "mean_token_accuracy": 0.22673921585083007,
      "num_tokens": 127089261.0,
      "step": 73270
    },
    {
      "entropy": 5.918449258804321,
      "epoch": 5.701030927835052,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022815119334144312,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22626321911811828,
      "num_tokens": 127097673.0,
      "step": 73275
    },
    {
      "entropy": 6.013857030868531,
      "epoch": 5.701419957206769,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022812408556090657,
      "loss": 4.9986,
      "mean_token_accuracy": 0.21233246326446534,
      "num_tokens": 127105774.0,
      "step": 73280
    },
    {
      "entropy": 6.051039505004883,
      "epoch": 5.701808986578486,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022809697849163736,
      "loss": 5.0519,
      "mean_token_accuracy": 0.20555254817008972,
      "num_tokens": 127114254.0,
      "step": 73285
    },
    {
      "entropy": 6.007740592956543,
      "epoch": 5.702198015950204,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022806987213404685,
      "loss": 4.9174,
      "mean_token_accuracy": 0.2264646425843239,
      "num_tokens": 127123228.0,
      "step": 73290
    },
    {
      "entropy": 5.893957090377808,
      "epoch": 5.702587045321922,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022804276648854632,
      "loss": 4.8508,
      "mean_token_accuracy": 0.2234017878770828,
      "num_tokens": 127132325.0,
      "step": 73295
    },
    {
      "entropy": 6.041404294967651,
      "epoch": 5.702976074693639,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00022801566155554708,
      "loss": 4.9854,
      "mean_token_accuracy": 0.21549068838357927,
      "num_tokens": 127141594.0,
      "step": 73300
    },
    {
      "entropy": 6.03462700843811,
      "epoch": 5.703365104065357,
      "grad_norm": 1.703125,
      "learning_rate": 0.00022798855733546038,
      "loss": 4.9949,
      "mean_token_accuracy": 0.21758174002170563,
      "num_tokens": 127150099.0,
      "step": 73305
    },
    {
      "entropy": 5.993129110336303,
      "epoch": 5.703754133437075,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002279614538286975,
      "loss": 5.0228,
      "mean_token_accuracy": 0.2110902860760689,
      "num_tokens": 127158569.0,
      "step": 73310
    },
    {
      "entropy": 6.029888916015625,
      "epoch": 5.704143162808792,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022793435103566972,
      "loss": 5.0375,
      "mean_token_accuracy": 0.21542368829250336,
      "num_tokens": 127167514.0,
      "step": 73315
    },
    {
      "entropy": 5.949106073379516,
      "epoch": 5.70453219218051,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022790724895678822,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22586736381053923,
      "num_tokens": 127176036.0,
      "step": 73320
    },
    {
      "entropy": 5.969253158569336,
      "epoch": 5.704921221552227,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022788014759246429,
      "loss": 4.9552,
      "mean_token_accuracy": 0.21692299842834473,
      "num_tokens": 127184347.0,
      "step": 73325
    },
    {
      "entropy": 6.040313720703125,
      "epoch": 5.705310250923945,
      "grad_norm": 1.625,
      "learning_rate": 0.0002278530469431091,
      "loss": 5.0855,
      "mean_token_accuracy": 0.21209137737751008,
      "num_tokens": 127193350.0,
      "step": 73330
    },
    {
      "entropy": 5.973772478103638,
      "epoch": 5.705699280295662,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002278259470091339,
      "loss": 4.9326,
      "mean_token_accuracy": 0.2164492830634117,
      "num_tokens": 127201910.0,
      "step": 73335
    },
    {
      "entropy": 5.915620565414429,
      "epoch": 5.70608830966738,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022779884779094972,
      "loss": 4.8812,
      "mean_token_accuracy": 0.22569292336702346,
      "num_tokens": 127210528.0,
      "step": 73340
    },
    {
      "entropy": 5.96608247756958,
      "epoch": 5.706477339039098,
      "grad_norm": 1.609375,
      "learning_rate": 0.00022777174928896804,
      "loss": 4.9621,
      "mean_token_accuracy": 0.21429431736469268,
      "num_tokens": 127218671.0,
      "step": 73345
    },
    {
      "entropy": 5.990810108184815,
      "epoch": 5.706866368410815,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002277446515035999,
      "loss": 5.0361,
      "mean_token_accuracy": 0.21298310905694962,
      "num_tokens": 127228003.0,
      "step": 73350
    },
    {
      "entropy": 5.980877637863159,
      "epoch": 5.707255397782532,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022771755443525644,
      "loss": 4.9244,
      "mean_token_accuracy": 0.223445363342762,
      "num_tokens": 127236114.0,
      "step": 73355
    },
    {
      "entropy": 5.971400213241577,
      "epoch": 5.70764442715425,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022769045808434884,
      "loss": 4.9369,
      "mean_token_accuracy": 0.21867581456899643,
      "num_tokens": 127245427.0,
      "step": 73360
    },
    {
      "entropy": 5.912933778762818,
      "epoch": 5.708033456525968,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002276633624512882,
      "loss": 4.8748,
      "mean_token_accuracy": 0.22943075150251388,
      "num_tokens": 127254732.0,
      "step": 73365
    },
    {
      "entropy": 5.903292989730835,
      "epoch": 5.708422485897685,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022763626753648575,
      "loss": 4.8514,
      "mean_token_accuracy": 0.22833894491195678,
      "num_tokens": 127262851.0,
      "step": 73370
    },
    {
      "entropy": 6.022132015228271,
      "epoch": 5.708811515269403,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00022760917334035252,
      "loss": 5.0391,
      "mean_token_accuracy": 0.20980738699436188,
      "num_tokens": 127270633.0,
      "step": 73375
    },
    {
      "entropy": 5.950638484954834,
      "epoch": 5.709200544641121,
      "grad_norm": 1.640625,
      "learning_rate": 0.00022758207986329964,
      "loss": 4.9174,
      "mean_token_accuracy": 0.2243496894836426,
      "num_tokens": 127278922.0,
      "step": 73380
    },
    {
      "entropy": 5.979461908340454,
      "epoch": 5.709589574012838,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022755498710573825,
      "loss": 4.979,
      "mean_token_accuracy": 0.2126516342163086,
      "num_tokens": 127287843.0,
      "step": 73385
    },
    {
      "entropy": 5.952922534942627,
      "epoch": 5.709978603384555,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022752789506807935,
      "loss": 4.9283,
      "mean_token_accuracy": 0.22086714655160905,
      "num_tokens": 127296999.0,
      "step": 73390
    },
    {
      "entropy": 6.054407930374145,
      "epoch": 5.710367632756273,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022750080375073418,
      "loss": 5.0284,
      "mean_token_accuracy": 0.22336628139019013,
      "num_tokens": 127305387.0,
      "step": 73395
    },
    {
      "entropy": 5.985918664932251,
      "epoch": 5.710756662127991,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022747371315411368,
      "loss": 4.8908,
      "mean_token_accuracy": 0.22526371628046035,
      "num_tokens": 127313467.0,
      "step": 73400
    },
    {
      "entropy": 5.953846645355225,
      "epoch": 5.711145691499708,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022744662327862887,
      "loss": 4.916,
      "mean_token_accuracy": 0.22159930914640427,
      "num_tokens": 127322381.0,
      "step": 73405
    },
    {
      "entropy": 5.978432846069336,
      "epoch": 5.711534720871426,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002274195341246909,
      "loss": 4.9469,
      "mean_token_accuracy": 0.2099955439567566,
      "num_tokens": 127330656.0,
      "step": 73410
    },
    {
      "entropy": 5.953506708145142,
      "epoch": 5.711923750243144,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022739244569271077,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2216631755232811,
      "num_tokens": 127339056.0,
      "step": 73415
    },
    {
      "entropy": 5.929978942871093,
      "epoch": 5.7123127796148605,
      "grad_norm": 1.609375,
      "learning_rate": 0.00022736535798309953,
      "loss": 4.879,
      "mean_token_accuracy": 0.22452974319458008,
      "num_tokens": 127346911.0,
      "step": 73420
    },
    {
      "entropy": 5.981953954696655,
      "epoch": 5.712701808986578,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022733827099626815,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21798523962497712,
      "num_tokens": 127354910.0,
      "step": 73425
    },
    {
      "entropy": 6.037444591522217,
      "epoch": 5.713090838358296,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002273111847326277,
      "loss": 5.0303,
      "mean_token_accuracy": 0.2170851856470108,
      "num_tokens": 127363658.0,
      "step": 73430
    },
    {
      "entropy": 5.902577924728393,
      "epoch": 5.713479867730014,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022728409919258908,
      "loss": 4.7888,
      "mean_token_accuracy": 0.2286820724606514,
      "num_tokens": 127371749.0,
      "step": 73435
    },
    {
      "entropy": 5.916402339935303,
      "epoch": 5.713868897101731,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022725701437656328,
      "loss": 4.8927,
      "mean_token_accuracy": 0.2284694105386734,
      "num_tokens": 127380414.0,
      "step": 73440
    },
    {
      "entropy": 5.992641258239746,
      "epoch": 5.714257926473449,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002272299302849613,
      "loss": 4.9812,
      "mean_token_accuracy": 0.21842539608478545,
      "num_tokens": 127388672.0,
      "step": 73445
    },
    {
      "entropy": 5.915173721313477,
      "epoch": 5.714646955845167,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002272028469181941,
      "loss": 4.8968,
      "mean_token_accuracy": 0.21677886694669724,
      "num_tokens": 127397313.0,
      "step": 73450
    },
    {
      "entropy": 5.9890586853027346,
      "epoch": 5.7150359852168835,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022717576427667258,
      "loss": 4.9166,
      "mean_token_accuracy": 0.2207922488451004,
      "num_tokens": 127404953.0,
      "step": 73455
    },
    {
      "entropy": 5.899847555160522,
      "epoch": 5.715425014588601,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022714868236080776,
      "loss": 4.8539,
      "mean_token_accuracy": 0.23402039110660552,
      "num_tokens": 127413750.0,
      "step": 73460
    },
    {
      "entropy": 5.9386186599731445,
      "epoch": 5.715814043960319,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022712160117101054,
      "loss": 4.9699,
      "mean_token_accuracy": 0.22194555252790452,
      "num_tokens": 127422667.0,
      "step": 73465
    },
    {
      "entropy": 5.9725341796875,
      "epoch": 5.716203073332037,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022709452070769176,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22151690423488618,
      "num_tokens": 127431627.0,
      "step": 73470
    },
    {
      "entropy": 6.029455709457397,
      "epoch": 5.716592102703754,
      "grad_norm": 1.640625,
      "learning_rate": 0.00022706744097126242,
      "loss": 4.9766,
      "mean_token_accuracy": 0.21266457289457322,
      "num_tokens": 127439867.0,
      "step": 73475
    },
    {
      "entropy": 5.934326887130737,
      "epoch": 5.716981132075472,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022704036196213336,
      "loss": 4.8194,
      "mean_token_accuracy": 0.2300341919064522,
      "num_tokens": 127448239.0,
      "step": 73480
    },
    {
      "entropy": 5.944498062133789,
      "epoch": 5.717370161447189,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022701328368071545,
      "loss": 5.0026,
      "mean_token_accuracy": 0.20820536613464355,
      "num_tokens": 127458518.0,
      "step": 73485
    },
    {
      "entropy": 6.006759357452393,
      "epoch": 5.7177591908189065,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022698620612741962,
      "loss": 4.9914,
      "mean_token_accuracy": 0.20946191996335983,
      "num_tokens": 127467640.0,
      "step": 73490
    },
    {
      "entropy": 5.93070502281189,
      "epoch": 5.718148220190624,
      "grad_norm": 1.5,
      "learning_rate": 0.00022695912930265666,
      "loss": 4.8178,
      "mean_token_accuracy": 0.2310593232512474,
      "num_tokens": 127475835.0,
      "step": 73495
    },
    {
      "entropy": 5.904641056060791,
      "epoch": 5.718537249562342,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002269320532068375,
      "loss": 4.9045,
      "mean_token_accuracy": 0.22180160284042358,
      "num_tokens": 127485081.0,
      "step": 73500
    },
    {
      "entropy": 5.965151453018189,
      "epoch": 5.71892627893406,
      "grad_norm": 1.6875,
      "learning_rate": 0.0002269049778403729,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22713935375213623,
      "num_tokens": 127493380.0,
      "step": 73505
    },
    {
      "entropy": 6.013972473144531,
      "epoch": 5.719315308305777,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022687790320367375,
      "loss": 5.0125,
      "mean_token_accuracy": 0.212660513818264,
      "num_tokens": 127502330.0,
      "step": 73510
    },
    {
      "entropy": 5.956238031387329,
      "epoch": 5.719704337677495,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022685082929715074,
      "loss": 4.9055,
      "mean_token_accuracy": 0.22114844620227814,
      "num_tokens": 127510516.0,
      "step": 73515
    },
    {
      "entropy": 5.974145078659058,
      "epoch": 5.720093367049213,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022682375612121481,
      "loss": 4.96,
      "mean_token_accuracy": 0.21204971969127656,
      "num_tokens": 127519192.0,
      "step": 73520
    },
    {
      "entropy": 5.985685539245606,
      "epoch": 5.7204823964209295,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002267966836762767,
      "loss": 4.9164,
      "mean_token_accuracy": 0.21958349496126175,
      "num_tokens": 127527795.0,
      "step": 73525
    },
    {
      "entropy": 5.9342468738555905,
      "epoch": 5.720871425792647,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022676961196274714,
      "loss": 4.9713,
      "mean_token_accuracy": 0.22401099950075148,
      "num_tokens": 127536132.0,
      "step": 73530
    },
    {
      "entropy": 6.050785923004151,
      "epoch": 5.721260455164365,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022674254098103692,
      "loss": 5.0066,
      "mean_token_accuracy": 0.21433781683444977,
      "num_tokens": 127544785.0,
      "step": 73535
    },
    {
      "entropy": 5.960423469543457,
      "epoch": 5.721649484536083,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002267154707315569,
      "loss": 4.864,
      "mean_token_accuracy": 0.21525000184774398,
      "num_tokens": 127553442.0,
      "step": 73540
    },
    {
      "entropy": 6.071310710906983,
      "epoch": 5.7220385139078,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002266884012147178,
      "loss": 5.1023,
      "mean_token_accuracy": 0.20675372779369355,
      "num_tokens": 127561539.0,
      "step": 73545
    },
    {
      "entropy": 6.010825347900391,
      "epoch": 5.722427543279518,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002266613324309303,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21991277039051055,
      "num_tokens": 127569936.0,
      "step": 73550
    },
    {
      "entropy": 5.9777521133422855,
      "epoch": 5.722816572651235,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022663426438060508,
      "loss": 4.9561,
      "mean_token_accuracy": 0.22072144001722335,
      "num_tokens": 127578337.0,
      "step": 73555
    },
    {
      "entropy": 5.985173749923706,
      "epoch": 5.7232056020229525,
      "grad_norm": 1.59375,
      "learning_rate": 0.00022660719706415295,
      "loss": 4.9848,
      "mean_token_accuracy": 0.20939017683267594,
      "num_tokens": 127586780.0,
      "step": 73560
    },
    {
      "entropy": 5.920226144790649,
      "epoch": 5.72359463139467,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022658013048198455,
      "loss": 4.8725,
      "mean_token_accuracy": 0.23122124820947648,
      "num_tokens": 127595605.0,
      "step": 73565
    },
    {
      "entropy": 6.0729865550994875,
      "epoch": 5.723983660766388,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022655306463451063,
      "loss": 5.0846,
      "mean_token_accuracy": 0.21082287430763244,
      "num_tokens": 127604695.0,
      "step": 73570
    },
    {
      "entropy": 6.07193775177002,
      "epoch": 5.724372690138106,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022652599952214186,
      "loss": 5.0418,
      "mean_token_accuracy": 0.21565143764019012,
      "num_tokens": 127613666.0,
      "step": 73575
    },
    {
      "entropy": 5.9728750705719,
      "epoch": 5.724761719509823,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022649893514528892,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21677229404449463,
      "num_tokens": 127623067.0,
      "step": 73580
    },
    {
      "entropy": 6.002970457077026,
      "epoch": 5.725150748881541,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002264718715043624,
      "loss": 4.9426,
      "mean_token_accuracy": 0.21113581359386444,
      "num_tokens": 127631020.0,
      "step": 73585
    },
    {
      "entropy": 5.929935503005981,
      "epoch": 5.725539778253258,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000226444808599773,
      "loss": 4.9563,
      "mean_token_accuracy": 0.21889320611953736,
      "num_tokens": 127639606.0,
      "step": 73590
    },
    {
      "entropy": 5.941722345352173,
      "epoch": 5.7259288076249755,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022641774643193137,
      "loss": 4.892,
      "mean_token_accuracy": 0.22205786257982255,
      "num_tokens": 127648123.0,
      "step": 73595
    },
    {
      "entropy": 5.986019134521484,
      "epoch": 5.726317836996693,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022639068500124803,
      "loss": 5.0287,
      "mean_token_accuracy": 0.21759441792964934,
      "num_tokens": 127657210.0,
      "step": 73600
    },
    {
      "entropy": 5.970897579193116,
      "epoch": 5.726706866368411,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022636362430813363,
      "loss": 4.869,
      "mean_token_accuracy": 0.22656756937503814,
      "num_tokens": 127666617.0,
      "step": 73605
    },
    {
      "entropy": 6.023809719085693,
      "epoch": 5.727095895740129,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022633656435299893,
      "loss": 5.0513,
      "mean_token_accuracy": 0.2163427785038948,
      "num_tokens": 127675555.0,
      "step": 73610
    },
    {
      "entropy": 5.945064067840576,
      "epoch": 5.727484925111846,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022630950513625436,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22436068952083588,
      "num_tokens": 127684454.0,
      "step": 73615
    },
    {
      "entropy": 5.940009355545044,
      "epoch": 5.727873954483563,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022628244665831067,
      "loss": 4.9113,
      "mean_token_accuracy": 0.22099453955888748,
      "num_tokens": 127693286.0,
      "step": 73620
    },
    {
      "entropy": 5.997782611846924,
      "epoch": 5.728262983855281,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002262553889195782,
      "loss": 5.0433,
      "mean_token_accuracy": 0.2175013765692711,
      "num_tokens": 127701923.0,
      "step": 73625
    },
    {
      "entropy": 5.996645307540893,
      "epoch": 5.7286520132269985,
      "grad_norm": 1.640625,
      "learning_rate": 0.00022622833192046765,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2260258361697197,
      "num_tokens": 127709687.0,
      "step": 73630
    },
    {
      "entropy": 5.935972118377686,
      "epoch": 5.729041042598716,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022620127566138948,
      "loss": 4.9644,
      "mean_token_accuracy": 0.213890440762043,
      "num_tokens": 127718651.0,
      "step": 73635
    },
    {
      "entropy": 5.924044132232666,
      "epoch": 5.729430071970434,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022617422014275436,
      "loss": 4.8803,
      "mean_token_accuracy": 0.2313445121049881,
      "num_tokens": 127727563.0,
      "step": 73640
    },
    {
      "entropy": 5.934467554092407,
      "epoch": 5.7298191013421516,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002261471653649727,
      "loss": 4.937,
      "mean_token_accuracy": 0.2124100521206856,
      "num_tokens": 127736289.0,
      "step": 73645
    },
    {
      "entropy": 5.8813577651977536,
      "epoch": 5.730208130713869,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022612011132845506,
      "loss": 4.8212,
      "mean_token_accuracy": 0.22848794013261794,
      "num_tokens": 127745404.0,
      "step": 73650
    },
    {
      "entropy": 5.936109495162964,
      "epoch": 5.730597160085587,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002260930580336119,
      "loss": 4.9133,
      "mean_token_accuracy": 0.22635226994752883,
      "num_tokens": 127754631.0,
      "step": 73655
    },
    {
      "entropy": 6.009025144577026,
      "epoch": 5.730986189457304,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00022606600548085375,
      "loss": 4.9502,
      "mean_token_accuracy": 0.22071287631988526,
      "num_tokens": 127763564.0,
      "step": 73660
    },
    {
      "entropy": 5.899959707260132,
      "epoch": 5.7313752188290215,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00022603895367059109,
      "loss": 4.8344,
      "mean_token_accuracy": 0.22856004536151886,
      "num_tokens": 127772132.0,
      "step": 73665
    },
    {
      "entropy": 5.977571105957031,
      "epoch": 5.731764248200739,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022601190260323434,
      "loss": 5.0193,
      "mean_token_accuracy": 0.21808982342481614,
      "num_tokens": 127781582.0,
      "step": 73670
    },
    {
      "entropy": 5.984799194335937,
      "epoch": 5.732153277572457,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000225984852279194,
      "loss": 4.9669,
      "mean_token_accuracy": 0.21805085390806198,
      "num_tokens": 127789504.0,
      "step": 73675
    },
    {
      "entropy": 5.9849841594696045,
      "epoch": 5.7325423069441745,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022595780269888055,
      "loss": 4.9472,
      "mean_token_accuracy": 0.2196095511317253,
      "num_tokens": 127799268.0,
      "step": 73680
    },
    {
      "entropy": 5.947044038772583,
      "epoch": 5.732931336315891,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00022593075386270438,
      "loss": 4.9077,
      "mean_token_accuracy": 0.22274007946252822,
      "num_tokens": 127808209.0,
      "step": 73685
    },
    {
      "entropy": 5.866025447845459,
      "epoch": 5.733320365687609,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022590370577107595,
      "loss": 4.8685,
      "mean_token_accuracy": 0.2232494667172432,
      "num_tokens": 127816470.0,
      "step": 73690
    },
    {
      "entropy": 5.972497034072876,
      "epoch": 5.733709395059327,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002258766584244056,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2214958995580673,
      "num_tokens": 127824451.0,
      "step": 73695
    },
    {
      "entropy": 5.908875226974487,
      "epoch": 5.7340984244310444,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022584961182310382,
      "loss": 4.8779,
      "mean_token_accuracy": 0.23376343548297882,
      "num_tokens": 127833654.0,
      "step": 73700
    },
    {
      "entropy": 5.953571844100952,
      "epoch": 5.734487453802762,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022582256596758094,
      "loss": 4.8791,
      "mean_token_accuracy": 0.2238616392016411,
      "num_tokens": 127842569.0,
      "step": 73705
    },
    {
      "entropy": 6.003153085708618,
      "epoch": 5.73487648317448,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022579552085824728,
      "loss": 4.9676,
      "mean_token_accuracy": 0.22277337461709976,
      "num_tokens": 127851267.0,
      "step": 73710
    },
    {
      "entropy": 5.966017532348633,
      "epoch": 5.7352655125461975,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002257684764955133,
      "loss": 4.9606,
      "mean_token_accuracy": 0.22171594947576523,
      "num_tokens": 127860225.0,
      "step": 73715
    },
    {
      "entropy": 5.946017694473267,
      "epoch": 5.735654541917915,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022574143287978937,
      "loss": 4.938,
      "mean_token_accuracy": 0.2143661543726921,
      "num_tokens": 127868470.0,
      "step": 73720
    },
    {
      "entropy": 5.894034099578858,
      "epoch": 5.736043571289632,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022571439001148574,
      "loss": 4.8738,
      "mean_token_accuracy": 0.22755395025014877,
      "num_tokens": 127877141.0,
      "step": 73725
    },
    {
      "entropy": 5.957109689712524,
      "epoch": 5.73643260066135,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022568734789101287,
      "loss": 4.9474,
      "mean_token_accuracy": 0.22278209775686264,
      "num_tokens": 127885900.0,
      "step": 73730
    },
    {
      "entropy": 5.908208560943604,
      "epoch": 5.736821630033067,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002256603065187809,
      "loss": 4.8798,
      "mean_token_accuracy": 0.22201057076454161,
      "num_tokens": 127894938.0,
      "step": 73735
    },
    {
      "entropy": 5.979791355133057,
      "epoch": 5.737210659404785,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022563326589520027,
      "loss": 4.893,
      "mean_token_accuracy": 0.21238429844379425,
      "num_tokens": 127903846.0,
      "step": 73740
    },
    {
      "entropy": 5.9115869998931885,
      "epoch": 5.737599688776503,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022560622602068125,
      "loss": 4.8454,
      "mean_token_accuracy": 0.22681476920843124,
      "num_tokens": 127912479.0,
      "step": 73745
    },
    {
      "entropy": 5.895171117782593,
      "epoch": 5.7379887181482205,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022557918689563423,
      "loss": 4.9333,
      "mean_token_accuracy": 0.2277906745672226,
      "num_tokens": 127920773.0,
      "step": 73750
    },
    {
      "entropy": 5.996135997772217,
      "epoch": 5.738377747519937,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002255521485204693,
      "loss": 5.0049,
      "mean_token_accuracy": 0.21437996178865432,
      "num_tokens": 127929481.0,
      "step": 73755
    },
    {
      "entropy": 5.94037389755249,
      "epoch": 5.738766776891655,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022552511089559684,
      "loss": 4.8978,
      "mean_token_accuracy": 0.22356765121221542,
      "num_tokens": 127938270.0,
      "step": 73760
    },
    {
      "entropy": 5.907193756103515,
      "epoch": 5.739155806263373,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002254980740214271,
      "loss": 4.8961,
      "mean_token_accuracy": 0.22225691080093385,
      "num_tokens": 127946824.0,
      "step": 73765
    },
    {
      "entropy": 5.9058263301849365,
      "epoch": 5.73954483563509,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002254710378983702,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22207724452018737,
      "num_tokens": 127955935.0,
      "step": 73770
    },
    {
      "entropy": 5.994569873809814,
      "epoch": 5.739933865006808,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022544400252683656,
      "loss": 4.9902,
      "mean_token_accuracy": 0.21830691397190094,
      "num_tokens": 127965445.0,
      "step": 73775
    },
    {
      "entropy": 5.946188449859619,
      "epoch": 5.740322894378526,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002254169679072363,
      "loss": 4.8437,
      "mean_token_accuracy": 0.22741733342409134,
      "num_tokens": 127973391.0,
      "step": 73780
    },
    {
      "entropy": 6.002416944503784,
      "epoch": 5.7407119237502435,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00022538993403997966,
      "loss": 4.985,
      "mean_token_accuracy": 0.22188954651355744,
      "num_tokens": 127981492.0,
      "step": 73785
    },
    {
      "entropy": 5.927619504928589,
      "epoch": 5.74110095312196,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022536290092547672,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21692586392164231,
      "num_tokens": 127989457.0,
      "step": 73790
    },
    {
      "entropy": 5.9815528869628904,
      "epoch": 5.741489982493678,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002253358685641378,
      "loss": 5.0156,
      "mean_token_accuracy": 0.2105855405330658,
      "num_tokens": 127997652.0,
      "step": 73795
    },
    {
      "entropy": 5.937820768356323,
      "epoch": 5.741879011865396,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000225308836956373,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21981030702590942,
      "num_tokens": 128007530.0,
      "step": 73800
    },
    {
      "entropy": 5.9655677318573,
      "epoch": 5.742268041237113,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022528180610259247,
      "loss": 4.9488,
      "mean_token_accuracy": 0.22177362591028213,
      "num_tokens": 128016671.0,
      "step": 73805
    },
    {
      "entropy": 5.992573642730713,
      "epoch": 5.742657070608831,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002252547760032065,
      "loss": 4.9941,
      "mean_token_accuracy": 0.21629100143909455,
      "num_tokens": 128025928.0,
      "step": 73810
    },
    {
      "entropy": 6.009360456466675,
      "epoch": 5.743046099980549,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022522774665862506,
      "loss": 4.975,
      "mean_token_accuracy": 0.2168005645275116,
      "num_tokens": 128034750.0,
      "step": 73815
    },
    {
      "entropy": 5.945447826385498,
      "epoch": 5.743435129352266,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002252007180692584,
      "loss": 4.9115,
      "mean_token_accuracy": 0.22193935215473176,
      "num_tokens": 128042556.0,
      "step": 73820
    },
    {
      "entropy": 5.977887153625488,
      "epoch": 5.743824158723983,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022517369023551654,
      "loss": 5.0272,
      "mean_token_accuracy": 0.2116738438606262,
      "num_tokens": 128051394.0,
      "step": 73825
    },
    {
      "entropy": 5.938125658035278,
      "epoch": 5.744213188095701,
      "grad_norm": 1.609375,
      "learning_rate": 0.00022514666315780963,
      "loss": 4.9043,
      "mean_token_accuracy": 0.21459803879261016,
      "num_tokens": 128059765.0,
      "step": 73830
    },
    {
      "entropy": 6.005118894577026,
      "epoch": 5.744602217467419,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022511963683654774,
      "loss": 4.9195,
      "mean_token_accuracy": 0.22322259694337845,
      "num_tokens": 128068787.0,
      "step": 73835
    },
    {
      "entropy": 5.9656342506408695,
      "epoch": 5.744991246839136,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022509261127214098,
      "loss": 4.8749,
      "mean_token_accuracy": 0.21611988246440889,
      "num_tokens": 128077754.0,
      "step": 73840
    },
    {
      "entropy": 6.006128931045533,
      "epoch": 5.745380276210854,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002250655864649994,
      "loss": 4.9678,
      "mean_token_accuracy": 0.2157450407743454,
      "num_tokens": 128086798.0,
      "step": 73845
    },
    {
      "entropy": 5.921226072311401,
      "epoch": 5.745769305582572,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002250385624155331,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2248624712228775,
      "num_tokens": 128095349.0,
      "step": 73850
    },
    {
      "entropy": 5.932105255126953,
      "epoch": 5.7461583349542895,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022501153912415206,
      "loss": 4.9215,
      "mean_token_accuracy": 0.223323193192482,
      "num_tokens": 128104133.0,
      "step": 73855
    },
    {
      "entropy": 5.907842206954956,
      "epoch": 5.746547364326006,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022498451659126639,
      "loss": 4.8295,
      "mean_token_accuracy": 0.22730852663516998,
      "num_tokens": 128112914.0,
      "step": 73860
    },
    {
      "entropy": 5.992288637161255,
      "epoch": 5.746936393697724,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022495749481728604,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21637393087148665,
      "num_tokens": 128121296.0,
      "step": 73865
    },
    {
      "entropy": 5.981282424926758,
      "epoch": 5.747325423069442,
      "grad_norm": 1.515625,
      "learning_rate": 0.000224930473802621,
      "loss": 4.9126,
      "mean_token_accuracy": 0.21965936273336412,
      "num_tokens": 128130187.0,
      "step": 73870
    },
    {
      "entropy": 5.983903551101685,
      "epoch": 5.747714452441159,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022490345354768138,
      "loss": 4.9357,
      "mean_token_accuracy": 0.21568517237901688,
      "num_tokens": 128139802.0,
      "step": 73875
    },
    {
      "entropy": 5.956787729263306,
      "epoch": 5.748103481812877,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0002248764340528771,
      "loss": 4.8542,
      "mean_token_accuracy": 0.23148174285888673,
      "num_tokens": 128147902.0,
      "step": 73880
    },
    {
      "entropy": 6.013383245468139,
      "epoch": 5.748492511184595,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002248494153186182,
      "loss": 5.0319,
      "mean_token_accuracy": 0.21431788057088852,
      "num_tokens": 128156268.0,
      "step": 73885
    },
    {
      "entropy": 6.011148262023926,
      "epoch": 5.748881540556312,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022482239734531462,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21912100464105605,
      "num_tokens": 128164751.0,
      "step": 73890
    },
    {
      "entropy": 6.02691683769226,
      "epoch": 5.749270569928029,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022479538013337624,
      "loss": 4.9978,
      "mean_token_accuracy": 0.2184231996536255,
      "num_tokens": 128173782.0,
      "step": 73895
    },
    {
      "entropy": 6.062984418869019,
      "epoch": 5.749659599299747,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022476836368321302,
      "loss": 5.1419,
      "mean_token_accuracy": 0.20311890095472335,
      "num_tokens": 128182891.0,
      "step": 73900
    },
    {
      "entropy": 6.028166675567627,
      "epoch": 5.750048628671465,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022474134799523499,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21794721037149428,
      "num_tokens": 128191157.0,
      "step": 73905
    },
    {
      "entropy": 5.967519092559814,
      "epoch": 5.750437658043182,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022471433306985202,
      "loss": 4.9464,
      "mean_token_accuracy": 0.2215065613389015,
      "num_tokens": 128199904.0,
      "step": 73910
    },
    {
      "entropy": 5.932801961898804,
      "epoch": 5.7508266874149,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022468731890747397,
      "loss": 4.948,
      "mean_token_accuracy": 0.21921778321266175,
      "num_tokens": 128208190.0,
      "step": 73915
    },
    {
      "entropy": 5.9668622493743895,
      "epoch": 5.751215716786618,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002246603055085108,
      "loss": 4.9372,
      "mean_token_accuracy": 0.21764402985572814,
      "num_tokens": 128217375.0,
      "step": 73920
    },
    {
      "entropy": 6.030282068252563,
      "epoch": 5.751604746158335,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022463329287337235,
      "loss": 5.0262,
      "mean_token_accuracy": 0.21233670711517333,
      "num_tokens": 128227250.0,
      "step": 73925
    },
    {
      "entropy": 5.986766624450683,
      "epoch": 5.751993775530052,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002246062810024685,
      "loss": 4.9065,
      "mean_token_accuracy": 0.22450969219207764,
      "num_tokens": 128235080.0,
      "step": 73930
    },
    {
      "entropy": 5.9093952655792235,
      "epoch": 5.75238280490177,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00022457926989620915,
      "loss": 4.7909,
      "mean_token_accuracy": 0.2274740844964981,
      "num_tokens": 128243978.0,
      "step": 73935
    },
    {
      "entropy": 5.929203987121582,
      "epoch": 5.752771834273488,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022455225955500408,
      "loss": 4.9502,
      "mean_token_accuracy": 0.22190335541963577,
      "num_tokens": 128253222.0,
      "step": 73940
    },
    {
      "entropy": 5.988745880126953,
      "epoch": 5.753160863645205,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022452524997926322,
      "loss": 4.9424,
      "mean_token_accuracy": 0.21600847244262694,
      "num_tokens": 128262209.0,
      "step": 73945
    },
    {
      "entropy": 6.0106785774230955,
      "epoch": 5.753549893016923,
      "grad_norm": 1.5,
      "learning_rate": 0.0002244982411693964,
      "loss": 4.9414,
      "mean_token_accuracy": 0.21476107388734816,
      "num_tokens": 128270507.0,
      "step": 73950
    },
    {
      "entropy": 5.951693916320801,
      "epoch": 5.75393892238864,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002244712331258133,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21896999031305314,
      "num_tokens": 128279333.0,
      "step": 73955
    },
    {
      "entropy": 5.9365170955657955,
      "epoch": 5.754327951760358,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002244442258489238,
      "loss": 4.9952,
      "mean_token_accuracy": 0.2174038916826248,
      "num_tokens": 128288206.0,
      "step": 73960
    },
    {
      "entropy": 5.926932287216187,
      "epoch": 5.754716981132075,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022441721933913778,
      "loss": 4.9255,
      "mean_token_accuracy": 0.22017727941274642,
      "num_tokens": 128297047.0,
      "step": 73965
    },
    {
      "entropy": 5.99899001121521,
      "epoch": 5.755106010503793,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000224390213596865,
      "loss": 4.976,
      "mean_token_accuracy": 0.229835844039917,
      "num_tokens": 128306072.0,
      "step": 73970
    },
    {
      "entropy": 5.970637512207031,
      "epoch": 5.755495039875511,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002243632086225151,
      "loss": 4.9532,
      "mean_token_accuracy": 0.22001931816339493,
      "num_tokens": 128314004.0,
      "step": 73975
    },
    {
      "entropy": 6.012783670425415,
      "epoch": 5.755884069247228,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002243362044164979,
      "loss": 4.9623,
      "mean_token_accuracy": 0.21459968984127045,
      "num_tokens": 128322889.0,
      "step": 73980
    },
    {
      "entropy": 6.000055932998658,
      "epoch": 5.756273098618946,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022430920097922313,
      "loss": 4.9894,
      "mean_token_accuracy": 0.21614200323820115,
      "num_tokens": 128331929.0,
      "step": 73985
    },
    {
      "entropy": 6.044557332992554,
      "epoch": 5.756662127990664,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022428219831110063,
      "loss": 4.9377,
      "mean_token_accuracy": 0.22137764394283294,
      "num_tokens": 128340398.0,
      "step": 73990
    },
    {
      "entropy": 5.8930902004241945,
      "epoch": 5.757051157362381,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022425519641254,
      "loss": 4.9028,
      "mean_token_accuracy": 0.22109356373548508,
      "num_tokens": 128349195.0,
      "step": 73995
    },
    {
      "entropy": 5.952122545242309,
      "epoch": 5.757440186734098,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022422819528395094,
      "loss": 4.8931,
      "mean_token_accuracy": 0.22573360204696655,
      "num_tokens": 128357695.0,
      "step": 74000
    },
    {
      "entropy": 5.878929376602173,
      "epoch": 5.757829216105816,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022420119492574326,
      "loss": 4.8481,
      "mean_token_accuracy": 0.2337367594242096,
      "num_tokens": 128366959.0,
      "step": 74005
    },
    {
      "entropy": 6.002831649780274,
      "epoch": 5.758218245477534,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022417419533832662,
      "loss": 4.9705,
      "mean_token_accuracy": 0.21589026749134063,
      "num_tokens": 128376065.0,
      "step": 74010
    },
    {
      "entropy": 6.042919301986695,
      "epoch": 5.758607274849251,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022414719652211063,
      "loss": 5.0122,
      "mean_token_accuracy": 0.2203315779566765,
      "num_tokens": 128384974.0,
      "step": 74015
    },
    {
      "entropy": 5.938033771514893,
      "epoch": 5.758996304220968,
      "grad_norm": 1.5,
      "learning_rate": 0.00022412019847750498,
      "loss": 4.8922,
      "mean_token_accuracy": 0.2303795337677002,
      "num_tokens": 128393290.0,
      "step": 74020
    },
    {
      "entropy": 5.859355878829956,
      "epoch": 5.759385333592686,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002240932012049194,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22061227709054948,
      "num_tokens": 128401392.0,
      "step": 74025
    },
    {
      "entropy": 5.94991250038147,
      "epoch": 5.759774362964404,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0002240662047047634,
      "loss": 4.9303,
      "mean_token_accuracy": 0.22780146896839143,
      "num_tokens": 128409053.0,
      "step": 74030
    },
    {
      "entropy": 5.926258277893067,
      "epoch": 5.760163392336121,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022403920897744672,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22381449341773987,
      "num_tokens": 128417451.0,
      "step": 74035
    },
    {
      "entropy": 5.970511627197266,
      "epoch": 5.760552421707839,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002240122140233789,
      "loss": 4.9255,
      "mean_token_accuracy": 0.22673526555299758,
      "num_tokens": 128426053.0,
      "step": 74040
    },
    {
      "entropy": 5.955502557754516,
      "epoch": 5.760941451079557,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002239852198429696,
      "loss": 4.9463,
      "mean_token_accuracy": 0.22056055963039398,
      "num_tokens": 128434949.0,
      "step": 74045
    },
    {
      "entropy": 5.97566499710083,
      "epoch": 5.761330480451274,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022395822643662844,
      "loss": 5.0013,
      "mean_token_accuracy": 0.21395210921764374,
      "num_tokens": 128443830.0,
      "step": 74050
    },
    {
      "entropy": 6.018917226791382,
      "epoch": 5.761719509822992,
      "grad_norm": 1.59375,
      "learning_rate": 0.00022393123380476488,
      "loss": 4.9728,
      "mean_token_accuracy": 0.2152975931763649,
      "num_tokens": 128452055.0,
      "step": 74055
    },
    {
      "entropy": 6.004529190063477,
      "epoch": 5.762108539194709,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00022390424194778858,
      "loss": 4.9567,
      "mean_token_accuracy": 0.21951660066843032,
      "num_tokens": 128460890.0,
      "step": 74060
    },
    {
      "entropy": 5.930108547210693,
      "epoch": 5.762497568566427,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022387725086610904,
      "loss": 4.9612,
      "mean_token_accuracy": 0.2253092870116234,
      "num_tokens": 128469474.0,
      "step": 74065
    },
    {
      "entropy": 6.038225507736206,
      "epoch": 5.762886597938144,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002238502605601358,
      "loss": 5.0244,
      "mean_token_accuracy": 0.21642083376646043,
      "num_tokens": 128478217.0,
      "step": 74070
    },
    {
      "entropy": 5.935362529754639,
      "epoch": 5.763275627309862,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022382327103027854,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22184652835130692,
      "num_tokens": 128486640.0,
      "step": 74075
    },
    {
      "entropy": 5.9567564010620115,
      "epoch": 5.76366465668158,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022379628227694675,
      "loss": 4.9586,
      "mean_token_accuracy": 0.21578085124492646,
      "num_tokens": 128495460.0,
      "step": 74080
    },
    {
      "entropy": 5.99856276512146,
      "epoch": 5.764053686053297,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022376929430054982,
      "loss": 4.9821,
      "mean_token_accuracy": 0.21449815928936006,
      "num_tokens": 128504193.0,
      "step": 74085
    },
    {
      "entropy": 5.984014940261841,
      "epoch": 5.764442715425014,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00022374230710149723,
      "loss": 4.9403,
      "mean_token_accuracy": 0.22657763063907624,
      "num_tokens": 128513775.0,
      "step": 74090
    },
    {
      "entropy": 6.015690517425537,
      "epoch": 5.764831744796732,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022371532068019863,
      "loss": 4.9521,
      "mean_token_accuracy": 0.22351000756025313,
      "num_tokens": 128523382.0,
      "step": 74095
    },
    {
      "entropy": 6.060959672927856,
      "epoch": 5.76522077416845,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022368833503706338,
      "loss": 5.0276,
      "mean_token_accuracy": 0.22141036689281463,
      "num_tokens": 128531951.0,
      "step": 74100
    },
    {
      "entropy": 6.008744478225708,
      "epoch": 5.765609803540167,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000223661350172501,
      "loss": 5.0392,
      "mean_token_accuracy": 0.21202799677848816,
      "num_tokens": 128540847.0,
      "step": 74105
    },
    {
      "entropy": 5.909643745422363,
      "epoch": 5.765998832911885,
      "grad_norm": 1.65625,
      "learning_rate": 0.00022363436608692085,
      "loss": 4.8756,
      "mean_token_accuracy": 0.22227570563554763,
      "num_tokens": 128549479.0,
      "step": 74110
    },
    {
      "entropy": 5.899044227600098,
      "epoch": 5.766387862283603,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002236073827807325,
      "loss": 4.8055,
      "mean_token_accuracy": 0.2288249284029007,
      "num_tokens": 128558203.0,
      "step": 74115
    },
    {
      "entropy": 5.977974557876587,
      "epoch": 5.76677689165532,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002235804002543453,
      "loss": 4.9157,
      "mean_token_accuracy": 0.2306768536567688,
      "num_tokens": 128567189.0,
      "step": 74120
    },
    {
      "entropy": 6.029678773880005,
      "epoch": 5.767165921027037,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002235534185081687,
      "loss": 5.0153,
      "mean_token_accuracy": 0.2117398649454117,
      "num_tokens": 128576159.0,
      "step": 74125
    },
    {
      "entropy": 6.025912570953369,
      "epoch": 5.767554950398755,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002235264375426121,
      "loss": 5.0105,
      "mean_token_accuracy": 0.21395154744386674,
      "num_tokens": 128583893.0,
      "step": 74130
    },
    {
      "entropy": 5.972445297241211,
      "epoch": 5.767943979770473,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022349945735808475,
      "loss": 4.8987,
      "mean_token_accuracy": 0.23024458438158035,
      "num_tokens": 128592455.0,
      "step": 74135
    },
    {
      "entropy": 5.900569486618042,
      "epoch": 5.76833300914219,
      "grad_norm": 1.640625,
      "learning_rate": 0.00022347247795499625,
      "loss": 4.8778,
      "mean_token_accuracy": 0.23239547610282899,
      "num_tokens": 128601754.0,
      "step": 74140
    },
    {
      "entropy": 5.987294101715088,
      "epoch": 5.768722038513908,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022344549933375597,
      "loss": 4.8811,
      "mean_token_accuracy": 0.22763680964708327,
      "num_tokens": 128609498.0,
      "step": 74145
    },
    {
      "entropy": 5.911344194412232,
      "epoch": 5.769111067885626,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022341852149477309,
      "loss": 4.892,
      "mean_token_accuracy": 0.22397129088640214,
      "num_tokens": 128617906.0,
      "step": 74150
    },
    {
      "entropy": 5.9225921630859375,
      "epoch": 5.7695000972573425,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022339154443845705,
      "loss": 4.9394,
      "mean_token_accuracy": 0.21992535889148712,
      "num_tokens": 128626375.0,
      "step": 74155
    },
    {
      "entropy": 5.957703113555908,
      "epoch": 5.76988912662906,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022336456816521722,
      "loss": 4.9077,
      "mean_token_accuracy": 0.22036875337362288,
      "num_tokens": 128635009.0,
      "step": 74160
    },
    {
      "entropy": 5.996530628204345,
      "epoch": 5.770278156000778,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002233375926754629,
      "loss": 4.9128,
      "mean_token_accuracy": 0.21568868905305863,
      "num_tokens": 128643895.0,
      "step": 74165
    },
    {
      "entropy": 6.087753868103027,
      "epoch": 5.770667185372496,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022331061796960328,
      "loss": 5.1089,
      "mean_token_accuracy": 0.20633645951747895,
      "num_tokens": 128652271.0,
      "step": 74170
    },
    {
      "entropy": 5.9094460010528564,
      "epoch": 5.771056214744213,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022328364404804784,
      "loss": 4.8733,
      "mean_token_accuracy": 0.2230027586221695,
      "num_tokens": 128661366.0,
      "step": 74175
    },
    {
      "entropy": 5.920444536209106,
      "epoch": 5.771445244115931,
      "grad_norm": 1.625,
      "learning_rate": 0.00022325667091120576,
      "loss": 4.9196,
      "mean_token_accuracy": 0.2127443879842758,
      "num_tokens": 128669517.0,
      "step": 74180
    },
    {
      "entropy": 5.954239988327027,
      "epoch": 5.771834273487649,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022322969855948638,
      "loss": 4.9547,
      "mean_token_accuracy": 0.22198956310749055,
      "num_tokens": 128679064.0,
      "step": 74185
    },
    {
      "entropy": 5.97049126625061,
      "epoch": 5.772223302859366,
      "grad_norm": 1.5,
      "learning_rate": 0.00022320272699329885,
      "loss": 4.956,
      "mean_token_accuracy": 0.22241855561733245,
      "num_tokens": 128687547.0,
      "step": 74190
    },
    {
      "entropy": 5.97144193649292,
      "epoch": 5.772612332231083,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00022317575621305254,
      "loss": 4.991,
      "mean_token_accuracy": 0.2125887766480446,
      "num_tokens": 128696519.0,
      "step": 74195
    },
    {
      "entropy": 6.0282642364501955,
      "epoch": 5.773001361602801,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022314878621915656,
      "loss": 4.9808,
      "mean_token_accuracy": 0.22029686570167542,
      "num_tokens": 128705093.0,
      "step": 74200
    },
    {
      "entropy": 6.009080839157105,
      "epoch": 5.773390390974519,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022312181701202033,
      "loss": 4.9164,
      "mean_token_accuracy": 0.22363196462392806,
      "num_tokens": 128715094.0,
      "step": 74205
    },
    {
      "entropy": 6.012739515304565,
      "epoch": 5.773779420346236,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022309484859205293,
      "loss": 4.9982,
      "mean_token_accuracy": 0.21884559392929076,
      "num_tokens": 128724202.0,
      "step": 74210
    },
    {
      "entropy": 5.953884315490723,
      "epoch": 5.774168449717954,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022306788095966356,
      "loss": 4.8644,
      "mean_token_accuracy": 0.22896973490715028,
      "num_tokens": 128732793.0,
      "step": 74215
    },
    {
      "entropy": 6.004562330245972,
      "epoch": 5.774557479089672,
      "grad_norm": 1.765625,
      "learning_rate": 0.00022304091411526144,
      "loss": 5.0081,
      "mean_token_accuracy": 0.21928359419107438,
      "num_tokens": 128742114.0,
      "step": 74220
    },
    {
      "entropy": 5.955711698532104,
      "epoch": 5.7749465084613885,
      "grad_norm": 1.5,
      "learning_rate": 0.0002230139480592558,
      "loss": 4.9448,
      "mean_token_accuracy": 0.21268683671951294,
      "num_tokens": 128751129.0,
      "step": 74225
    },
    {
      "entropy": 6.019234466552734,
      "epoch": 5.775335537833106,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00022298698279205571,
      "loss": 5.0379,
      "mean_token_accuracy": 0.2128247231245041,
      "num_tokens": 128759929.0,
      "step": 74230
    },
    {
      "entropy": 6.046534776687622,
      "epoch": 5.775724567204824,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022296001831407036,
      "loss": 4.9797,
      "mean_token_accuracy": 0.21412660777568818,
      "num_tokens": 128768069.0,
      "step": 74235
    },
    {
      "entropy": 5.95254225730896,
      "epoch": 5.7761135965765416,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022293305462570896,
      "loss": 4.9105,
      "mean_token_accuracy": 0.2199403628706932,
      "num_tokens": 128776590.0,
      "step": 74240
    },
    {
      "entropy": 5.979356050491333,
      "epoch": 5.776502625948259,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022290609172738058,
      "loss": 4.9234,
      "mean_token_accuracy": 0.2163197547197342,
      "num_tokens": 128785407.0,
      "step": 74245
    },
    {
      "entropy": 5.977344512939453,
      "epoch": 5.776891655319977,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022287912961949425,
      "loss": 5.0382,
      "mean_token_accuracy": 0.21667649894952773,
      "num_tokens": 128794357.0,
      "step": 74250
    },
    {
      "entropy": 5.979211664199829,
      "epoch": 5.777280684691695,
      "grad_norm": 1.5,
      "learning_rate": 0.00022285216830245924,
      "loss": 4.9693,
      "mean_token_accuracy": 0.21672483533620834,
      "num_tokens": 128803435.0,
      "step": 74255
    },
    {
      "entropy": 5.901760005950928,
      "epoch": 5.7776697140634115,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002228252077766846,
      "loss": 4.8221,
      "mean_token_accuracy": 0.2268456280231476,
      "num_tokens": 128811898.0,
      "step": 74260
    },
    {
      "entropy": 6.054072761535645,
      "epoch": 5.778058743435129,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002227982480425793,
      "loss": 5.0136,
      "mean_token_accuracy": 0.21808566749095917,
      "num_tokens": 128821607.0,
      "step": 74265
    },
    {
      "entropy": 5.9550083637237545,
      "epoch": 5.778447772806847,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002227712891005525,
      "loss": 4.9007,
      "mean_token_accuracy": 0.22599513828754425,
      "num_tokens": 128830520.0,
      "step": 74270
    },
    {
      "entropy": 5.948575210571289,
      "epoch": 5.7788368021785645,
      "grad_norm": 1.640625,
      "learning_rate": 0.00022274433095101332,
      "loss": 4.8819,
      "mean_token_accuracy": 0.22165126204490662,
      "num_tokens": 128838963.0,
      "step": 74275
    },
    {
      "entropy": 5.961292695999146,
      "epoch": 5.779225831550282,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022271737359437073,
      "loss": 4.9163,
      "mean_token_accuracy": 0.22529024332761766,
      "num_tokens": 128848398.0,
      "step": 74280
    },
    {
      "entropy": 6.02721471786499,
      "epoch": 5.779614860922,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022269041703103375,
      "loss": 4.9627,
      "mean_token_accuracy": 0.22163235247135163,
      "num_tokens": 128857030.0,
      "step": 74285
    },
    {
      "entropy": 5.881620073318482,
      "epoch": 5.780003890293717,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022266346126141145,
      "loss": 4.8527,
      "mean_token_accuracy": 0.23045167177915574,
      "num_tokens": 128866097.0,
      "step": 74290
    },
    {
      "entropy": 5.910781764984131,
      "epoch": 5.7803929196654344,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002226365062859128,
      "loss": 4.7693,
      "mean_token_accuracy": 0.23547940701246262,
      "num_tokens": 128874320.0,
      "step": 74295
    },
    {
      "entropy": 5.915812253952026,
      "epoch": 5.780781949037152,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022260955210494688,
      "loss": 4.862,
      "mean_token_accuracy": 0.22532907724380494,
      "num_tokens": 128882309.0,
      "step": 74300
    },
    {
      "entropy": 5.9534656524658205,
      "epoch": 5.78117097840887,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00022258259871892256,
      "loss": 4.9971,
      "mean_token_accuracy": 0.21490557491779327,
      "num_tokens": 128891313.0,
      "step": 74305
    },
    {
      "entropy": 5.988818883895874,
      "epoch": 5.7815600077805875,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002225556461282489,
      "loss": 4.96,
      "mean_token_accuracy": 0.21611177325248718,
      "num_tokens": 128899935.0,
      "step": 74310
    },
    {
      "entropy": 6.042578983306885,
      "epoch": 5.781949037152305,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022252869433333483,
      "loss": 5.0399,
      "mean_token_accuracy": 0.21017520129680634,
      "num_tokens": 128909048.0,
      "step": 74315
    },
    {
      "entropy": 5.974864339828491,
      "epoch": 5.782338066524023,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002225017433345894,
      "loss": 4.9587,
      "mean_token_accuracy": 0.21120716482400895,
      "num_tokens": 128918350.0,
      "step": 74320
    },
    {
      "entropy": 6.004842472076416,
      "epoch": 5.78272709589574,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022247479313242135,
      "loss": 4.9389,
      "mean_token_accuracy": 0.2227520987391472,
      "num_tokens": 128926579.0,
      "step": 74325
    },
    {
      "entropy": 5.956453514099121,
      "epoch": 5.783116125267457,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022244784372723976,
      "loss": 4.8808,
      "mean_token_accuracy": 0.227315816283226,
      "num_tokens": 128935117.0,
      "step": 74330
    },
    {
      "entropy": 5.895151424407959,
      "epoch": 5.783505154639175,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022242089511945336,
      "loss": 4.8401,
      "mean_token_accuracy": 0.22679676562547685,
      "num_tokens": 128943639.0,
      "step": 74335
    },
    {
      "entropy": 5.986280012130737,
      "epoch": 5.783894184010893,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022239394730947127,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21992597579956055,
      "num_tokens": 128952163.0,
      "step": 74340
    },
    {
      "entropy": 5.906219244003296,
      "epoch": 5.7842832133826105,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022236700029770235,
      "loss": 4.8696,
      "mean_token_accuracy": 0.22599049359560014,
      "num_tokens": 128960705.0,
      "step": 74345
    },
    {
      "entropy": 5.962341737747193,
      "epoch": 5.784672242754328,
      "grad_norm": 1.609375,
      "learning_rate": 0.00022234005408455543,
      "loss": 4.91,
      "mean_token_accuracy": 0.22650137841701506,
      "num_tokens": 128971127.0,
      "step": 74350
    },
    {
      "entropy": 6.00708794593811,
      "epoch": 5.785061272126045,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00022231310867043934,
      "loss": 5.0603,
      "mean_token_accuracy": 0.21395664811134338,
      "num_tokens": 128981181.0,
      "step": 74355
    },
    {
      "entropy": 5.989402532577515,
      "epoch": 5.785450301497763,
      "grad_norm": 1.5,
      "learning_rate": 0.00022228616405576296,
      "loss": 4.9285,
      "mean_token_accuracy": 0.22056756764650345,
      "num_tokens": 128989136.0,
      "step": 74360
    },
    {
      "entropy": 6.0663048267364506,
      "epoch": 5.78583933086948,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022225922024093515,
      "loss": 4.9935,
      "mean_token_accuracy": 0.21976594477891923,
      "num_tokens": 128997866.0,
      "step": 74365
    },
    {
      "entropy": 6.040359973907471,
      "epoch": 5.786228360241198,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022223227722636474,
      "loss": 4.9283,
      "mean_token_accuracy": 0.22389977425336838,
      "num_tokens": 129006332.0,
      "step": 74370
    },
    {
      "entropy": 5.957034969329834,
      "epoch": 5.786617389612916,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00022220533501246049,
      "loss": 4.8926,
      "mean_token_accuracy": 0.22728656083345414,
      "num_tokens": 129014604.0,
      "step": 74375
    },
    {
      "entropy": 5.845887136459351,
      "epoch": 5.7870064189846335,
      "grad_norm": 1.625,
      "learning_rate": 0.00022217839359963133,
      "loss": 4.866,
      "mean_token_accuracy": 0.22101878821849824,
      "num_tokens": 129022754.0,
      "step": 74380
    },
    {
      "entropy": 5.914460277557373,
      "epoch": 5.787395448356351,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022215145298828587,
      "loss": 4.8693,
      "mean_token_accuracy": 0.22014544606208802,
      "num_tokens": 129031617.0,
      "step": 74385
    },
    {
      "entropy": 5.9735533714294435,
      "epoch": 5.787784477728069,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002221245131788331,
      "loss": 4.8313,
      "mean_token_accuracy": 0.22528285086154937,
      "num_tokens": 129040197.0,
      "step": 74390
    },
    {
      "entropy": 5.976630067825317,
      "epoch": 5.788173507099786,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002220975741716816,
      "loss": 4.9154,
      "mean_token_accuracy": 0.22415462136268616,
      "num_tokens": 129048975.0,
      "step": 74395
    },
    {
      "entropy": 5.967725944519043,
      "epoch": 5.788562536471503,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002220706359672402,
      "loss": 4.8491,
      "mean_token_accuracy": 0.22340276688337327,
      "num_tokens": 129056804.0,
      "step": 74400
    },
    {
      "entropy": 5.973757028579712,
      "epoch": 5.788951565843221,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022204369856591765,
      "loss": 4.9832,
      "mean_token_accuracy": 0.21980056911706924,
      "num_tokens": 129065426.0,
      "step": 74405
    },
    {
      "entropy": 5.988022089004517,
      "epoch": 5.789340595214939,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022201676196812275,
      "loss": 4.9987,
      "mean_token_accuracy": 0.21796843111515046,
      "num_tokens": 129073935.0,
      "step": 74410
    },
    {
      "entropy": 5.927433776855469,
      "epoch": 5.7897296245866565,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022198982617426417,
      "loss": 4.9454,
      "mean_token_accuracy": 0.21734785586595534,
      "num_tokens": 129082823.0,
      "step": 74415
    },
    {
      "entropy": 6.050788307189942,
      "epoch": 5.790118653958374,
      "grad_norm": 1.75,
      "learning_rate": 0.00022196289118475056,
      "loss": 4.9938,
      "mean_token_accuracy": 0.21520411372184753,
      "num_tokens": 129092309.0,
      "step": 74420
    },
    {
      "entropy": 5.938680648803711,
      "epoch": 5.790507683330091,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022193595699999063,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22673131227493287,
      "num_tokens": 129100892.0,
      "step": 74425
    },
    {
      "entropy": 6.017592716217041,
      "epoch": 5.790896712701809,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022190902362039316,
      "loss": 5.0243,
      "mean_token_accuracy": 0.21443244814872742,
      "num_tokens": 129110331.0,
      "step": 74430
    },
    {
      "entropy": 5.98145170211792,
      "epoch": 5.791285742073526,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022188209104636675,
      "loss": 5.0287,
      "mean_token_accuracy": 0.20809046030044556,
      "num_tokens": 129119066.0,
      "step": 74435
    },
    {
      "entropy": 5.9320793628692625,
      "epoch": 5.791674771445244,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022185515927832,
      "loss": 4.9011,
      "mean_token_accuracy": 0.2259317457675934,
      "num_tokens": 129127612.0,
      "step": 74440
    },
    {
      "entropy": 5.989129161834716,
      "epoch": 5.792063800816962,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022182822831666167,
      "loss": 4.9698,
      "mean_token_accuracy": 0.2177804470062256,
      "num_tokens": 129136385.0,
      "step": 74445
    },
    {
      "entropy": 5.984185743331909,
      "epoch": 5.7924528301886795,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022180129816180033,
      "loss": 4.9121,
      "mean_token_accuracy": 0.2214762508869171,
      "num_tokens": 129145918.0,
      "step": 74450
    },
    {
      "entropy": 5.952899503707886,
      "epoch": 5.792841859560397,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022177436881414458,
      "loss": 4.978,
      "mean_token_accuracy": 0.21036597490310668,
      "num_tokens": 129155234.0,
      "step": 74455
    },
    {
      "entropy": 6.007909202575684,
      "epoch": 5.793230888932114,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022174744027410315,
      "loss": 4.8813,
      "mean_token_accuracy": 0.23109251409769058,
      "num_tokens": 129163113.0,
      "step": 74460
    },
    {
      "entropy": 6.000672197341919,
      "epoch": 5.793619918303832,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022172051254208442,
      "loss": 4.9663,
      "mean_token_accuracy": 0.22267835587263107,
      "num_tokens": 129172020.0,
      "step": 74465
    },
    {
      "entropy": 5.9690320014953615,
      "epoch": 5.794008947675549,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022169358561849724,
      "loss": 4.8817,
      "mean_token_accuracy": 0.2194139301776886,
      "num_tokens": 129180869.0,
      "step": 74470
    },
    {
      "entropy": 6.025505542755127,
      "epoch": 5.794397977047267,
      "grad_norm": 1.640625,
      "learning_rate": 0.00022166665950374998,
      "loss": 5.074,
      "mean_token_accuracy": 0.21208372265100478,
      "num_tokens": 129189557.0,
      "step": 74475
    },
    {
      "entropy": 5.971070241928101,
      "epoch": 5.794787006418985,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022163973419825128,
      "loss": 4.9716,
      "mean_token_accuracy": 0.21518829166889192,
      "num_tokens": 129198524.0,
      "step": 74480
    },
    {
      "entropy": 6.012900972366333,
      "epoch": 5.7951760357907025,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002216128097024097,
      "loss": 4.9832,
      "mean_token_accuracy": 0.21429242491722106,
      "num_tokens": 129207119.0,
      "step": 74485
    },
    {
      "entropy": 5.93060827255249,
      "epoch": 5.795565065162419,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022158588601663376,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22371262907981873,
      "num_tokens": 129214886.0,
      "step": 74490
    },
    {
      "entropy": 5.922150182723999,
      "epoch": 5.795954094534137,
      "grad_norm": 1.5,
      "learning_rate": 0.00022155896314133194,
      "loss": 4.8942,
      "mean_token_accuracy": 0.21472203582525254,
      "num_tokens": 129223638.0,
      "step": 74495
    },
    {
      "entropy": 5.911415004730225,
      "epoch": 5.796343123905855,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022153204107691287,
      "loss": 4.9324,
      "mean_token_accuracy": 0.2171951174736023,
      "num_tokens": 129232157.0,
      "step": 74500
    },
    {
      "entropy": 5.9966139793396,
      "epoch": 5.796732153277572,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022150511982378491,
      "loss": 5.0686,
      "mean_token_accuracy": 0.2092237040400505,
      "num_tokens": 129241067.0,
      "step": 74505
    },
    {
      "entropy": 6.022087669372558,
      "epoch": 5.79712118264929,
      "grad_norm": 1.5,
      "learning_rate": 0.0002214781993823567,
      "loss": 4.9621,
      "mean_token_accuracy": 0.2258890151977539,
      "num_tokens": 129249222.0,
      "step": 74510
    },
    {
      "entropy": 5.973256778717041,
      "epoch": 5.797510212021008,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022145127975303658,
      "loss": 4.8931,
      "mean_token_accuracy": 0.2184516504406929,
      "num_tokens": 129258103.0,
      "step": 74515
    },
    {
      "entropy": 5.986192083358764,
      "epoch": 5.7978992413927255,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022142436093623304,
      "loss": 4.8993,
      "mean_token_accuracy": 0.2264567032456398,
      "num_tokens": 129265651.0,
      "step": 74520
    },
    {
      "entropy": 5.963683605194092,
      "epoch": 5.798288270764443,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022139744293235453,
      "loss": 5.0253,
      "mean_token_accuracy": 0.21307792961597444,
      "num_tokens": 129274608.0,
      "step": 74525
    },
    {
      "entropy": 5.960446834564209,
      "epoch": 5.79867730013616,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022137052574180955,
      "loss": 4.9307,
      "mean_token_accuracy": 0.21911882013082504,
      "num_tokens": 129283123.0,
      "step": 74530
    },
    {
      "entropy": 5.961977863311768,
      "epoch": 5.799066329507878,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022134360936500636,
      "loss": 4.8515,
      "mean_token_accuracy": 0.22168502658605577,
      "num_tokens": 129291663.0,
      "step": 74535
    },
    {
      "entropy": 5.9627532958984375,
      "epoch": 5.799455358879595,
      "grad_norm": 1.609375,
      "learning_rate": 0.00022131669380235364,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21777227967977525,
      "num_tokens": 129300213.0,
      "step": 74540
    },
    {
      "entropy": 6.0190352439880375,
      "epoch": 5.799844388251313,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022128977905425955,
      "loss": 5.0507,
      "mean_token_accuracy": 0.21483851075172425,
      "num_tokens": 129308128.0,
      "step": 74545
    },
    {
      "entropy": 5.935274982452393,
      "epoch": 5.800233417623031,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022126286512113265,
      "loss": 4.8399,
      "mean_token_accuracy": 0.22432650476694108,
      "num_tokens": 129316121.0,
      "step": 74550
    },
    {
      "entropy": 5.970142412185669,
      "epoch": 5.800622446994748,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002212359520033812,
      "loss": 4.9402,
      "mean_token_accuracy": 0.2170947641134262,
      "num_tokens": 129324474.0,
      "step": 74555
    },
    {
      "entropy": 5.9086161136627195,
      "epoch": 5.801011476366465,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022120903970141364,
      "loss": 4.9128,
      "mean_token_accuracy": 0.22303399592638015,
      "num_tokens": 129332808.0,
      "step": 74560
    },
    {
      "entropy": 5.992447566986084,
      "epoch": 5.801400505738183,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00022118212821563827,
      "loss": 5.0102,
      "mean_token_accuracy": 0.2226722940802574,
      "num_tokens": 129340668.0,
      "step": 74565
    },
    {
      "entropy": 5.958662986755371,
      "epoch": 5.801789535109901,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022115521754646346,
      "loss": 5.0065,
      "mean_token_accuracy": 0.21348714977502822,
      "num_tokens": 129348557.0,
      "step": 74570
    },
    {
      "entropy": 5.9620596408844,
      "epoch": 5.802178564481618,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022112830769429744,
      "loss": 4.8813,
      "mean_token_accuracy": 0.2182161405682564,
      "num_tokens": 129357190.0,
      "step": 74575
    },
    {
      "entropy": 5.983040618896484,
      "epoch": 5.802567593853336,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022110139865954865,
      "loss": 4.938,
      "mean_token_accuracy": 0.21999209523200988,
      "num_tokens": 129365180.0,
      "step": 74580
    },
    {
      "entropy": 6.048008441925049,
      "epoch": 5.802956623225054,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002210744904426254,
      "loss": 4.9882,
      "mean_token_accuracy": 0.21842924505472183,
      "num_tokens": 129374038.0,
      "step": 74585
    },
    {
      "entropy": 5.991805171966552,
      "epoch": 5.8033456525967715,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002210475830439359,
      "loss": 4.9205,
      "mean_token_accuracy": 0.22405851930379866,
      "num_tokens": 129382511.0,
      "step": 74590
    },
    {
      "entropy": 5.959285497665405,
      "epoch": 5.803734681968488,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00022102067646388845,
      "loss": 4.9438,
      "mean_token_accuracy": 0.21658312976360322,
      "num_tokens": 129391309.0,
      "step": 74595
    },
    {
      "entropy": 5.992723846435547,
      "epoch": 5.804123711340206,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022099377070289124,
      "loss": 5.0672,
      "mean_token_accuracy": 0.20857247859239578,
      "num_tokens": 129401090.0,
      "step": 74600
    },
    {
      "entropy": 6.005412864685058,
      "epoch": 5.804512740711924,
      "grad_norm": 1.5,
      "learning_rate": 0.00022096686576135268,
      "loss": 4.9172,
      "mean_token_accuracy": 0.222334323823452,
      "num_tokens": 129409890.0,
      "step": 74605
    },
    {
      "entropy": 6.039430284500122,
      "epoch": 5.804901770083641,
      "grad_norm": 1.5,
      "learning_rate": 0.00022093996163968088,
      "loss": 5.0355,
      "mean_token_accuracy": 0.2135229155421257,
      "num_tokens": 129418440.0,
      "step": 74610
    },
    {
      "entropy": 6.070140790939331,
      "epoch": 5.805290799455359,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022091305833828412,
      "loss": 5.0526,
      "mean_token_accuracy": 0.21326230615377426,
      "num_tokens": 129427518.0,
      "step": 74615
    },
    {
      "entropy": 5.894359302520752,
      "epoch": 5.805679828827077,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022088615585757072,
      "loss": 4.8637,
      "mean_token_accuracy": 0.2253113105893135,
      "num_tokens": 129435720.0,
      "step": 74620
    },
    {
      "entropy": 5.961225461959839,
      "epoch": 5.806068858198794,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022085925419794872,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22265677154064178,
      "num_tokens": 129443900.0,
      "step": 74625
    },
    {
      "entropy": 5.9214503288269045,
      "epoch": 5.806457887570511,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022083235335982633,
      "loss": 4.887,
      "mean_token_accuracy": 0.2305159643292427,
      "num_tokens": 129452626.0,
      "step": 74630
    },
    {
      "entropy": 5.9452403545379635,
      "epoch": 5.806846916942229,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022080545334361175,
      "loss": 4.9358,
      "mean_token_accuracy": 0.22646720707416534,
      "num_tokens": 129461672.0,
      "step": 74635
    },
    {
      "entropy": 5.977540063858032,
      "epoch": 5.807235946313947,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002207785541497132,
      "loss": 4.9935,
      "mean_token_accuracy": 0.21779861450195312,
      "num_tokens": 129470647.0,
      "step": 74640
    },
    {
      "entropy": 6.000776863098144,
      "epoch": 5.807624975685664,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022075165577853872,
      "loss": 4.9614,
      "mean_token_accuracy": 0.21805160492658615,
      "num_tokens": 129479501.0,
      "step": 74645
    },
    {
      "entropy": 6.004046058654785,
      "epoch": 5.808014005057382,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002207247582304966,
      "loss": 4.9448,
      "mean_token_accuracy": 0.22690936625003816,
      "num_tokens": 129487999.0,
      "step": 74650
    },
    {
      "entropy": 6.000285005569458,
      "epoch": 5.8084030344291,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022069786150599486,
      "loss": 4.9703,
      "mean_token_accuracy": 0.21706788390874862,
      "num_tokens": 129496334.0,
      "step": 74655
    },
    {
      "entropy": 5.987415170669555,
      "epoch": 5.808792063800817,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022067096560544165,
      "loss": 4.983,
      "mean_token_accuracy": 0.20947948545217515,
      "num_tokens": 129505042.0,
      "step": 74660
    },
    {
      "entropy": 5.96248984336853,
      "epoch": 5.809181093172534,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022064407052924494,
      "loss": 4.9111,
      "mean_token_accuracy": 0.2145792618393898,
      "num_tokens": 129514228.0,
      "step": 74665
    },
    {
      "entropy": 5.90454626083374,
      "epoch": 5.809570122544252,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022061717627781308,
      "loss": 4.8735,
      "mean_token_accuracy": 0.2274075046181679,
      "num_tokens": 129522346.0,
      "step": 74670
    },
    {
      "entropy": 5.940846633911133,
      "epoch": 5.80995915191597,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022059028285155397,
      "loss": 4.9042,
      "mean_token_accuracy": 0.22727851569652557,
      "num_tokens": 129531160.0,
      "step": 74675
    },
    {
      "entropy": 6.022503471374511,
      "epoch": 5.810348181287687,
      "grad_norm": 1.625,
      "learning_rate": 0.00022056339025087573,
      "loss": 4.9828,
      "mean_token_accuracy": 0.21539098024368286,
      "num_tokens": 129539809.0,
      "step": 74680
    },
    {
      "entropy": 6.0160809516906735,
      "epoch": 5.810737210659405,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002205364984761864,
      "loss": 4.986,
      "mean_token_accuracy": 0.21632646173238754,
      "num_tokens": 129548645.0,
      "step": 74685
    },
    {
      "entropy": 5.937717771530151,
      "epoch": 5.811126240031122,
      "grad_norm": 1.515625,
      "learning_rate": 0.000220509607527894,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22783505618572236,
      "num_tokens": 129557648.0,
      "step": 74690
    },
    {
      "entropy": 5.951441097259521,
      "epoch": 5.81151526940284,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002204827174064066,
      "loss": 4.9575,
      "mean_token_accuracy": 0.21695807427167893,
      "num_tokens": 129567171.0,
      "step": 74695
    },
    {
      "entropy": 5.980539846420288,
      "epoch": 5.811904298774557,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022045582811213221,
      "loss": 5.0013,
      "mean_token_accuracy": 0.21406418979167938,
      "num_tokens": 129575942.0,
      "step": 74700
    },
    {
      "entropy": 5.97846736907959,
      "epoch": 5.812293328146275,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022042893964547877,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21719594597816466,
      "num_tokens": 129584404.0,
      "step": 74705
    },
    {
      "entropy": 5.991814422607422,
      "epoch": 5.812682357517993,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002204020520068543,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21474804878234863,
      "num_tokens": 129593714.0,
      "step": 74710
    },
    {
      "entropy": 5.912239694595337,
      "epoch": 5.81307138688971,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022037516519666684,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22384639829397202,
      "num_tokens": 129602307.0,
      "step": 74715
    },
    {
      "entropy": 5.956178474426269,
      "epoch": 5.813460416261428,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002203482792153243,
      "loss": 4.9721,
      "mean_token_accuracy": 0.21578607708215714,
      "num_tokens": 129611193.0,
      "step": 74720
    },
    {
      "entropy": 5.982153606414795,
      "epoch": 5.813849445633146,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002203213940632346,
      "loss": 4.9838,
      "mean_token_accuracy": 0.2152566209435463,
      "num_tokens": 129619976.0,
      "step": 74725
    },
    {
      "entropy": 5.971900177001953,
      "epoch": 5.814238475004863,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022029450974080572,
      "loss": 4.8968,
      "mean_token_accuracy": 0.21895099431276321,
      "num_tokens": 129629177.0,
      "step": 74730
    },
    {
      "entropy": 5.9613972187042235,
      "epoch": 5.81462750437658,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00022026762624844555,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21793029457330704,
      "num_tokens": 129637295.0,
      "step": 74735
    },
    {
      "entropy": 5.972036647796631,
      "epoch": 5.815016533748298,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022024074358656205,
      "loss": 4.9745,
      "mean_token_accuracy": 0.21856921762228013,
      "num_tokens": 129646050.0,
      "step": 74740
    },
    {
      "entropy": 6.001359748840332,
      "epoch": 5.815405563120016,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022021386175556318,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21905555278062822,
      "num_tokens": 129654982.0,
      "step": 74745
    },
    {
      "entropy": 5.957924270629883,
      "epoch": 5.815794592491733,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002201869807558567,
      "loss": 4.9399,
      "mean_token_accuracy": 0.22014482468366622,
      "num_tokens": 129663611.0,
      "step": 74750
    },
    {
      "entropy": 6.043409061431885,
      "epoch": 5.816183621863451,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022016010058785053,
      "loss": 5.0688,
      "mean_token_accuracy": 0.21099483370780944,
      "num_tokens": 129672076.0,
      "step": 74755
    },
    {
      "entropy": 5.960317611694336,
      "epoch": 5.816572651235168,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022013322125195256,
      "loss": 4.9653,
      "mean_token_accuracy": 0.22158870249986648,
      "num_tokens": 129681234.0,
      "step": 74760
    },
    {
      "entropy": 6.025253200531006,
      "epoch": 5.816961680606886,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002201063427485706,
      "loss": 4.9594,
      "mean_token_accuracy": 0.21399749368429183,
      "num_tokens": 129689602.0,
      "step": 74765
    },
    {
      "entropy": 5.995745086669922,
      "epoch": 5.817350709978603,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002200794650781126,
      "loss": 4.8994,
      "mean_token_accuracy": 0.2258068934082985,
      "num_tokens": 129698614.0,
      "step": 74770
    },
    {
      "entropy": 6.042575836181641,
      "epoch": 5.817739739350321,
      "grad_norm": 1.5625,
      "learning_rate": 0.00022005258824098624,
      "loss": 5.0112,
      "mean_token_accuracy": 0.2115415647625923,
      "num_tokens": 129707167.0,
      "step": 74775
    },
    {
      "entropy": 6.032680416107178,
      "epoch": 5.818128768722039,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002200257122375995,
      "loss": 4.9978,
      "mean_token_accuracy": 0.21499050110578538,
      "num_tokens": 129715257.0,
      "step": 74780
    },
    {
      "entropy": 5.912034225463867,
      "epoch": 5.818517798093756,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021999883706835994,
      "loss": 4.8439,
      "mean_token_accuracy": 0.22459196895360947,
      "num_tokens": 129724593.0,
      "step": 74785
    },
    {
      "entropy": 5.920084285736084,
      "epoch": 5.818906827465474,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021997196273367553,
      "loss": 4.9023,
      "mean_token_accuracy": 0.22555151730775833,
      "num_tokens": 129733255.0,
      "step": 74790
    },
    {
      "entropy": 5.92941107749939,
      "epoch": 5.819295856837191,
      "grad_norm": 1.484375,
      "learning_rate": 0.000219945089233954,
      "loss": 4.8622,
      "mean_token_accuracy": 0.2250741496682167,
      "num_tokens": 129740828.0,
      "step": 74795
    },
    {
      "entropy": 5.947973775863647,
      "epoch": 5.819684886208909,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021991821656960303,
      "loss": 4.9372,
      "mean_token_accuracy": 0.21692323982715606,
      "num_tokens": 129748999.0,
      "step": 74800
    },
    {
      "entropy": 5.978154134750366,
      "epoch": 5.820073915580626,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021989134474103056,
      "loss": 4.9117,
      "mean_token_accuracy": 0.2206864356994629,
      "num_tokens": 129757278.0,
      "step": 74805
    },
    {
      "entropy": 5.947690629959107,
      "epoch": 5.820462944952344,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021986447374864422,
      "loss": 4.9589,
      "mean_token_accuracy": 0.20802993029356004,
      "num_tokens": 129765644.0,
      "step": 74810
    },
    {
      "entropy": 6.005950403213501,
      "epoch": 5.820851974324062,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021983760359285171,
      "loss": 4.9866,
      "mean_token_accuracy": 0.21875055730342866,
      "num_tokens": 129774394.0,
      "step": 74815
    },
    {
      "entropy": 5.958064556121826,
      "epoch": 5.821241003695779,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002198107342740608,
      "loss": 4.9045,
      "mean_token_accuracy": 0.22646512538194657,
      "num_tokens": 129782448.0,
      "step": 74820
    },
    {
      "entropy": 5.953708124160767,
      "epoch": 5.821630033067496,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021978386579267907,
      "loss": 4.7913,
      "mean_token_accuracy": 0.2305470660328865,
      "num_tokens": 129790557.0,
      "step": 74825
    },
    {
      "entropy": 5.944882249832153,
      "epoch": 5.822019062439214,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021975699814911432,
      "loss": 4.9084,
      "mean_token_accuracy": 0.22396039813756943,
      "num_tokens": 129798651.0,
      "step": 74830
    },
    {
      "entropy": 5.893306016921997,
      "epoch": 5.8224080918109316,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021973013134377422,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23473196625709533,
      "num_tokens": 129807665.0,
      "step": 74835
    },
    {
      "entropy": 5.9664756774902346,
      "epoch": 5.822797121182649,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00021970326537706643,
      "loss": 4.8576,
      "mean_token_accuracy": 0.2306991085410118,
      "num_tokens": 129815927.0,
      "step": 74840
    },
    {
      "entropy": 5.952864408493042,
      "epoch": 5.823186150554367,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021967640024939851,
      "loss": 4.9217,
      "mean_token_accuracy": 0.22236475348472595,
      "num_tokens": 129823840.0,
      "step": 74845
    },
    {
      "entropy": 5.981057214736938,
      "epoch": 5.823575179926085,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002196495359611782,
      "loss": 4.9717,
      "mean_token_accuracy": 0.21021681725978852,
      "num_tokens": 129832238.0,
      "step": 74850
    },
    {
      "entropy": 6.023973417282105,
      "epoch": 5.823964209297802,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002196226725128131,
      "loss": 5.0461,
      "mean_token_accuracy": 0.2133082702755928,
      "num_tokens": 129841072.0,
      "step": 74855
    },
    {
      "entropy": 5.929681873321533,
      "epoch": 5.82435323866952,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021959580990471073,
      "loss": 4.9407,
      "mean_token_accuracy": 0.2181203246116638,
      "num_tokens": 129850255.0,
      "step": 74860
    },
    {
      "entropy": 5.935883188247681,
      "epoch": 5.824742268041237,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002195689481372788,
      "loss": 4.926,
      "mean_token_accuracy": 0.21653397977352143,
      "num_tokens": 129859201.0,
      "step": 74865
    },
    {
      "entropy": 5.9956306457519535,
      "epoch": 5.8251312974129545,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021954208721092482,
      "loss": 4.9234,
      "mean_token_accuracy": 0.2257247507572174,
      "num_tokens": 129867760.0,
      "step": 74870
    },
    {
      "entropy": 5.948914957046509,
      "epoch": 5.825520326784672,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021951522712605648,
      "loss": 4.9135,
      "mean_token_accuracy": 0.21794894486665725,
      "num_tokens": 129875852.0,
      "step": 74875
    },
    {
      "entropy": 6.033364009857178,
      "epoch": 5.82590935615639,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021948836788308117,
      "loss": 5.0033,
      "mean_token_accuracy": 0.2119821012020111,
      "num_tokens": 129884629.0,
      "step": 74880
    },
    {
      "entropy": 6.04655647277832,
      "epoch": 5.826298385528108,
      "grad_norm": 1.5,
      "learning_rate": 0.00021946150948240657,
      "loss": 5.0176,
      "mean_token_accuracy": 0.21859383434057236,
      "num_tokens": 129893354.0,
      "step": 74885
    },
    {
      "entropy": 5.9582772731781,
      "epoch": 5.8266874148998244,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021943465192444023,
      "loss": 4.9063,
      "mean_token_accuracy": 0.2189235880970955,
      "num_tokens": 129902048.0,
      "step": 74890
    },
    {
      "entropy": 5.974492835998535,
      "epoch": 5.827076444271542,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021940779520958952,
      "loss": 4.9215,
      "mean_token_accuracy": 0.22404073178768158,
      "num_tokens": 129910346.0,
      "step": 74895
    },
    {
      "entropy": 6.043318748474121,
      "epoch": 5.82746547364326,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002193809393382621,
      "loss": 5.1016,
      "mean_token_accuracy": 0.2042091280221939,
      "num_tokens": 129918971.0,
      "step": 74900
    },
    {
      "entropy": 5.994036340713501,
      "epoch": 5.8278545030149775,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021935408431086528,
      "loss": 4.9457,
      "mean_token_accuracy": 0.2123458355665207,
      "num_tokens": 129928356.0,
      "step": 74905
    },
    {
      "entropy": 5.989865827560425,
      "epoch": 5.828243532386695,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021932723012780677,
      "loss": 4.8869,
      "mean_token_accuracy": 0.2295105993747711,
      "num_tokens": 129937205.0,
      "step": 74910
    },
    {
      "entropy": 5.9744339942932125,
      "epoch": 5.828632561758413,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021930037678949388,
      "loss": 4.9638,
      "mean_token_accuracy": 0.21859457492828369,
      "num_tokens": 129945438.0,
      "step": 74915
    },
    {
      "entropy": 6.003981065750122,
      "epoch": 5.829021591130131,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021927352429633413,
      "loss": 4.9583,
      "mean_token_accuracy": 0.21607041954994202,
      "num_tokens": 129953565.0,
      "step": 74920
    },
    {
      "entropy": 5.941140651702881,
      "epoch": 5.829410620501848,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021924667264873492,
      "loss": 4.9517,
      "mean_token_accuracy": 0.21605134010314941,
      "num_tokens": 129963015.0,
      "step": 74925
    },
    {
      "entropy": 5.973437547683716,
      "epoch": 5.829799649873565,
      "grad_norm": 1.625,
      "learning_rate": 0.0002192198218471037,
      "loss": 4.9841,
      "mean_token_accuracy": 0.21653540432453156,
      "num_tokens": 129972077.0,
      "step": 74930
    },
    {
      "entropy": 6.0400285720825195,
      "epoch": 5.830188679245283,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00021919297189184795,
      "loss": 5.0388,
      "mean_token_accuracy": 0.2149857461452484,
      "num_tokens": 129981535.0,
      "step": 74935
    },
    {
      "entropy": 6.049007940292358,
      "epoch": 5.8305777086170005,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000219166122783375,
      "loss": 4.9694,
      "mean_token_accuracy": 0.2202465608716011,
      "num_tokens": 129988859.0,
      "step": 74940
    },
    {
      "entropy": 5.981409120559692,
      "epoch": 5.830966737988718,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021913927452209232,
      "loss": 4.9386,
      "mean_token_accuracy": 0.21949045211076737,
      "num_tokens": 129997030.0,
      "step": 74945
    },
    {
      "entropy": 5.841466808319092,
      "epoch": 5.831355767360436,
      "grad_norm": 1.671875,
      "learning_rate": 0.00021911242710840717,
      "loss": 4.7505,
      "mean_token_accuracy": 0.2314741790294647,
      "num_tokens": 130005515.0,
      "step": 74950
    },
    {
      "entropy": 5.85128870010376,
      "epoch": 5.831744796732154,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021908558054272703,
      "loss": 4.8209,
      "mean_token_accuracy": 0.22870866507291793,
      "num_tokens": 130014848.0,
      "step": 74955
    },
    {
      "entropy": 5.899602699279785,
      "epoch": 5.83213382610387,
      "grad_norm": 1.5,
      "learning_rate": 0.0002190587348254592,
      "loss": 4.8597,
      "mean_token_accuracy": 0.223554590344429,
      "num_tokens": 130023341.0,
      "step": 74960
    },
    {
      "entropy": 5.986833810806274,
      "epoch": 5.832522855475588,
      "grad_norm": 1.640625,
      "learning_rate": 0.000219031889957011,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21719675809144973,
      "num_tokens": 130031617.0,
      "step": 74965
    },
    {
      "entropy": 5.996923160552979,
      "epoch": 5.832911884847306,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021900504593778985,
      "loss": 4.9328,
      "mean_token_accuracy": 0.21896204948425294,
      "num_tokens": 130040192.0,
      "step": 74970
    },
    {
      "entropy": 6.026058387756348,
      "epoch": 5.8333009142190235,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00021897820276820296,
      "loss": 4.9996,
      "mean_token_accuracy": 0.21317307353019715,
      "num_tokens": 130048707.0,
      "step": 74975
    },
    {
      "entropy": 5.968855953216552,
      "epoch": 5.833689943590741,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002189513604486577,
      "loss": 4.9555,
      "mean_token_accuracy": 0.2164836809039116,
      "num_tokens": 130056971.0,
      "step": 74980
    },
    {
      "entropy": 5.950008535385132,
      "epoch": 5.834078972962459,
      "grad_norm": 1.375,
      "learning_rate": 0.0002189245189795613,
      "loss": 4.9039,
      "mean_token_accuracy": 0.21826159209012985,
      "num_tokens": 130066151.0,
      "step": 74985
    },
    {
      "entropy": 5.942103052139283,
      "epoch": 5.834468002334177,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021889767836132107,
      "loss": 5.005,
      "mean_token_accuracy": 0.21520727574825288,
      "num_tokens": 130075236.0,
      "step": 74990
    },
    {
      "entropy": 5.956946086883545,
      "epoch": 5.834857031705893,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021887083859434433,
      "loss": 4.9585,
      "mean_token_accuracy": 0.22355642467737197,
      "num_tokens": 130084154.0,
      "step": 74995
    },
    {
      "entropy": 6.012527418136597,
      "epoch": 5.835246061077611,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021884399967903818,
      "loss": 4.9624,
      "mean_token_accuracy": 0.2186752215027809,
      "num_tokens": 130093046.0,
      "step": 75000
    },
    {
      "epoch": 5.835246061077611,
      "eval_entropy": 5.701734090608628,
      "eval_loss": 5.04495906829834,
      "eval_mean_token_accuracy": 0.22350261892345344,
      "eval_num_tokens": 130093046.0,
      "eval_runtime": 21.6427,
      "eval_samples_per_second": 1920.188,
      "eval_steps_per_second": 240.035,
      "step": 75000
    },
    {
      "entropy": 5.955037450790405,
      "epoch": 5.835635090449329,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021881716161581,
      "loss": 4.9309,
      "mean_token_accuracy": 0.21489462107419968,
      "num_tokens": 130101403.0,
      "step": 75005
    },
    {
      "entropy": 5.902156209945678,
      "epoch": 5.8360241198210465,
      "grad_norm": 1.75,
      "learning_rate": 0.00021879032440506697,
      "loss": 4.8546,
      "mean_token_accuracy": 0.221111299097538,
      "num_tokens": 130110051.0,
      "step": 75010
    },
    {
      "entropy": 5.990875196456909,
      "epoch": 5.836413149192764,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002187634880472163,
      "loss": 4.9415,
      "mean_token_accuracy": 0.21849344968795775,
      "num_tokens": 130118598.0,
      "step": 75015
    },
    {
      "entropy": 5.996807956695557,
      "epoch": 5.836802178564482,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00021873665254266517,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21732147783041,
      "num_tokens": 130126784.0,
      "step": 75020
    },
    {
      "entropy": 5.908301544189453,
      "epoch": 5.837191207936199,
      "grad_norm": 1.625,
      "learning_rate": 0.00021870981789182082,
      "loss": 4.919,
      "mean_token_accuracy": 0.22357892245054245,
      "num_tokens": 130135493.0,
      "step": 75025
    },
    {
      "entropy": 5.919711399078369,
      "epoch": 5.837580237307916,
      "grad_norm": 1.640625,
      "learning_rate": 0.0002186829840950904,
      "loss": 4.93,
      "mean_token_accuracy": 0.2198515623807907,
      "num_tokens": 130144871.0,
      "step": 75030
    },
    {
      "entropy": 6.035551595687866,
      "epoch": 5.837969266679634,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000218656151152881,
      "loss": 4.9601,
      "mean_token_accuracy": 0.21848211586475372,
      "num_tokens": 130153917.0,
      "step": 75035
    },
    {
      "entropy": 6.029910326004028,
      "epoch": 5.838358296051352,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021862931906559992,
      "loss": 5.0072,
      "mean_token_accuracy": 0.21773739904165268,
      "num_tokens": 130161873.0,
      "step": 75040
    },
    {
      "entropy": 5.920537376403809,
      "epoch": 5.8387473254230695,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021860248783365412,
      "loss": 4.797,
      "mean_token_accuracy": 0.23541744202375411,
      "num_tokens": 130170844.0,
      "step": 75045
    },
    {
      "entropy": 5.938541746139526,
      "epoch": 5.839136354794787,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021857565745745083,
      "loss": 4.9135,
      "mean_token_accuracy": 0.2217307820916176,
      "num_tokens": 130179076.0,
      "step": 75050
    },
    {
      "entropy": 6.055966854095459,
      "epoch": 5.839525384166505,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021854882793739716,
      "loss": 5.0421,
      "mean_token_accuracy": 0.21184190064668657,
      "num_tokens": 130187992.0,
      "step": 75055
    },
    {
      "entropy": 5.992700624465942,
      "epoch": 5.839914413538223,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021852199927390015,
      "loss": 5.007,
      "mean_token_accuracy": 0.21636045277118682,
      "num_tokens": 130196642.0,
      "step": 75060
    },
    {
      "entropy": 5.878013849258423,
      "epoch": 5.840303442909939,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002184951714673668,
      "loss": 4.7668,
      "mean_token_accuracy": 0.23662480860948562,
      "num_tokens": 130205536.0,
      "step": 75065
    },
    {
      "entropy": 5.9727105617523195,
      "epoch": 5.840692472281657,
      "grad_norm": 1.5,
      "learning_rate": 0.00021846834451820437,
      "loss": 5.0138,
      "mean_token_accuracy": 0.21317314654588698,
      "num_tokens": 130214153.0,
      "step": 75070
    },
    {
      "entropy": 5.989931058883667,
      "epoch": 5.841081501653375,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021844151842681986,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22068050503730774,
      "num_tokens": 130222437.0,
      "step": 75075
    },
    {
      "entropy": 5.891117334365845,
      "epoch": 5.8414705310250925,
      "grad_norm": 1.5,
      "learning_rate": 0.00021841469319362034,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22799665629863738,
      "num_tokens": 130230920.0,
      "step": 75080
    },
    {
      "entropy": 5.978481912612915,
      "epoch": 5.84185956039681,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021838786881901267,
      "loss": 4.9311,
      "mean_token_accuracy": 0.21898679286241532,
      "num_tokens": 130239374.0,
      "step": 75085
    },
    {
      "entropy": 5.967590951919556,
      "epoch": 5.842248589768528,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00021836104530340407,
      "loss": 4.9194,
      "mean_token_accuracy": 0.21591199487447738,
      "num_tokens": 130249070.0,
      "step": 75090
    },
    {
      "entropy": 6.022045755386353,
      "epoch": 5.842637619140245,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021833422264720136,
      "loss": 5.008,
      "mean_token_accuracy": 0.2155309408903122,
      "num_tokens": 130258087.0,
      "step": 75095
    },
    {
      "entropy": 5.998011255264283,
      "epoch": 5.843026648511962,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021830740085081169,
      "loss": 4.9902,
      "mean_token_accuracy": 0.20982946455478668,
      "num_tokens": 130266039.0,
      "step": 75100
    },
    {
      "entropy": 5.946198320388794,
      "epoch": 5.84341567788368,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000218280579914642,
      "loss": 4.8978,
      "mean_token_accuracy": 0.2221483513712883,
      "num_tokens": 130274787.0,
      "step": 75105
    },
    {
      "entropy": 5.972231340408325,
      "epoch": 5.843804707255398,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021825375983909918,
      "loss": 4.938,
      "mean_token_accuracy": 0.21653356403112411,
      "num_tokens": 130284013.0,
      "step": 75110
    },
    {
      "entropy": 5.981378650665283,
      "epoch": 5.8441937366271155,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002182269406245903,
      "loss": 4.8897,
      "mean_token_accuracy": 0.23564158231019974,
      "num_tokens": 130292167.0,
      "step": 75115
    },
    {
      "entropy": 5.974547910690307,
      "epoch": 5.844582765998833,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021820012227152214,
      "loss": 4.9189,
      "mean_token_accuracy": 0.2205956682562828,
      "num_tokens": 130300614.0,
      "step": 75120
    },
    {
      "entropy": 5.956206321716309,
      "epoch": 5.844971795370551,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021817330478030177,
      "loss": 4.8784,
      "mean_token_accuracy": 0.22766678631305695,
      "num_tokens": 130309683.0,
      "step": 75125
    },
    {
      "entropy": 6.008305931091309,
      "epoch": 5.845360824742268,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021814648815133604,
      "loss": 4.9826,
      "mean_token_accuracy": 0.2131974294781685,
      "num_tokens": 130318771.0,
      "step": 75130
    },
    {
      "entropy": 6.017253017425537,
      "epoch": 5.845749854113985,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021811967238503184,
      "loss": 5.0363,
      "mean_token_accuracy": 0.21643770784139632,
      "num_tokens": 130327359.0,
      "step": 75135
    },
    {
      "entropy": 6.0009904384613035,
      "epoch": 5.846138883485703,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021809285748179612,
      "loss": 5.047,
      "mean_token_accuracy": 0.20678585916757583,
      "num_tokens": 130336690.0,
      "step": 75140
    },
    {
      "entropy": 5.96063494682312,
      "epoch": 5.846527912857421,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002180660434420357,
      "loss": 4.9429,
      "mean_token_accuracy": 0.2206372380256653,
      "num_tokens": 130345391.0,
      "step": 75145
    },
    {
      "entropy": 6.031637382507324,
      "epoch": 5.8469169422291385,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021803923026615746,
      "loss": 5.0111,
      "mean_token_accuracy": 0.2200985997915268,
      "num_tokens": 130354134.0,
      "step": 75150
    },
    {
      "entropy": 6.0123405933380125,
      "epoch": 5.847305971600856,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002180124179545682,
      "loss": 4.9381,
      "mean_token_accuracy": 0.21608905792236327,
      "num_tokens": 130362633.0,
      "step": 75155
    },
    {
      "entropy": 6.04129548072815,
      "epoch": 5.847695000972573,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021798560650767486,
      "loss": 5.0677,
      "mean_token_accuracy": 0.2096573978662491,
      "num_tokens": 130371110.0,
      "step": 75160
    },
    {
      "entropy": 5.92115626335144,
      "epoch": 5.848084030344291,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021795879592588418,
      "loss": 4.9254,
      "mean_token_accuracy": 0.2140694960951805,
      "num_tokens": 130379754.0,
      "step": 75165
    },
    {
      "entropy": 5.932467269897461,
      "epoch": 5.848473059716008,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021793198620960293,
      "loss": 4.926,
      "mean_token_accuracy": 0.21644344478845595,
      "num_tokens": 130388352.0,
      "step": 75170
    },
    {
      "entropy": 6.022741317749023,
      "epoch": 5.848862089087726,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021790517735923798,
      "loss": 4.9871,
      "mean_token_accuracy": 0.21522610187530516,
      "num_tokens": 130397190.0,
      "step": 75175
    },
    {
      "entropy": 6.017928743362427,
      "epoch": 5.849251118459444,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002178783693751961,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21671112775802612,
      "num_tokens": 130405955.0,
      "step": 75180
    },
    {
      "entropy": 6.002825403213501,
      "epoch": 5.8496401478311615,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000217851562257884,
      "loss": 4.983,
      "mean_token_accuracy": 0.2171285182237625,
      "num_tokens": 130414366.0,
      "step": 75185
    },
    {
      "entropy": 5.948372459411621,
      "epoch": 5.850029177202879,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021782475600770856,
      "loss": 4.848,
      "mean_token_accuracy": 0.22881193608045577,
      "num_tokens": 130422600.0,
      "step": 75190
    },
    {
      "entropy": 5.934473085403442,
      "epoch": 5.850418206574597,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021779795062507634,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21951285749673843,
      "num_tokens": 130431208.0,
      "step": 75195
    },
    {
      "entropy": 5.970952892303467,
      "epoch": 5.850807235946314,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021777114611039422,
      "loss": 4.9857,
      "mean_token_accuracy": 0.2151118740439415,
      "num_tokens": 130440181.0,
      "step": 75200
    },
    {
      "entropy": 5.966131401062012,
      "epoch": 5.851196265318031,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021774434246406887,
      "loss": 4.936,
      "mean_token_accuracy": 0.21687786430120468,
      "num_tokens": 130448555.0,
      "step": 75205
    },
    {
      "entropy": 6.060975170135498,
      "epoch": 5.851585294689749,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021771753968650705,
      "loss": 5.0702,
      "mean_token_accuracy": 0.207884281873703,
      "num_tokens": 130457262.0,
      "step": 75210
    },
    {
      "entropy": 5.9914445877075195,
      "epoch": 5.851974324061467,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021769073777811533,
      "loss": 4.9102,
      "mean_token_accuracy": 0.2216745287179947,
      "num_tokens": 130465257.0,
      "step": 75215
    },
    {
      "entropy": 5.934783315658569,
      "epoch": 5.8523633534331845,
      "grad_norm": 1.640625,
      "learning_rate": 0.00021766393673930046,
      "loss": 4.8442,
      "mean_token_accuracy": 0.21795590668916703,
      "num_tokens": 130473441.0,
      "step": 75220
    },
    {
      "entropy": 5.924726676940918,
      "epoch": 5.852752382804901,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021763713657046906,
      "loss": 4.9074,
      "mean_token_accuracy": 0.22204250395298003,
      "num_tokens": 130482758.0,
      "step": 75225
    },
    {
      "entropy": 5.98883810043335,
      "epoch": 5.853141412176619,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021761033727202782,
      "loss": 5.0377,
      "mean_token_accuracy": 0.2088179662823677,
      "num_tokens": 130491826.0,
      "step": 75230
    },
    {
      "entropy": 6.004818153381348,
      "epoch": 5.853530441548337,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021758353884438343,
      "loss": 5.0035,
      "mean_token_accuracy": 0.21052451878786088,
      "num_tokens": 130501275.0,
      "step": 75235
    },
    {
      "entropy": 5.947307825088501,
      "epoch": 5.853919470920054,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0002175567412879424,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22054249793291092,
      "num_tokens": 130511127.0,
      "step": 75240
    },
    {
      "entropy": 5.944646739959717,
      "epoch": 5.854308500291772,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021752994460311144,
      "loss": 4.9262,
      "mean_token_accuracy": 0.22301099598407745,
      "num_tokens": 130520104.0,
      "step": 75245
    },
    {
      "entropy": 6.019086837768555,
      "epoch": 5.85469752966349,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021750314879029708,
      "loss": 4.9843,
      "mean_token_accuracy": 0.22124726325273514,
      "num_tokens": 130528657.0,
      "step": 75250
    },
    {
      "entropy": 5.9754256248474125,
      "epoch": 5.8550865590352075,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002174763538499059,
      "loss": 4.9792,
      "mean_token_accuracy": 0.2157418742775917,
      "num_tokens": 130537286.0,
      "step": 75255
    },
    {
      "entropy": 6.04588475227356,
      "epoch": 5.855475588406925,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021744955978234449,
      "loss": 5.0582,
      "mean_token_accuracy": 0.21473521888256072,
      "num_tokens": 130546220.0,
      "step": 75260
    },
    {
      "entropy": 5.989775228500366,
      "epoch": 5.855864617778642,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002174227665880193,
      "loss": 4.9446,
      "mean_token_accuracy": 0.2144552245736122,
      "num_tokens": 130555223.0,
      "step": 75265
    },
    {
      "entropy": 6.043210840225219,
      "epoch": 5.85625364715036,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021739597426733714,
      "loss": 4.9752,
      "mean_token_accuracy": 0.21351270228624344,
      "num_tokens": 130563930.0,
      "step": 75270
    },
    {
      "entropy": 6.011846399307251,
      "epoch": 5.856642676522077,
      "grad_norm": 1.640625,
      "learning_rate": 0.00021736918282070435,
      "loss": 5.0101,
      "mean_token_accuracy": 0.22058479189872743,
      "num_tokens": 130572757.0,
      "step": 75275
    },
    {
      "entropy": 5.954722881317139,
      "epoch": 5.857031705893795,
      "grad_norm": 1.5,
      "learning_rate": 0.00021734239224852742,
      "loss": 4.9184,
      "mean_token_accuracy": 0.21994494497776032,
      "num_tokens": 130581585.0,
      "step": 75280
    },
    {
      "entropy": 5.9526691913604735,
      "epoch": 5.857420735265513,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021731560255121296,
      "loss": 4.9157,
      "mean_token_accuracy": 0.221183080971241,
      "num_tokens": 130590040.0,
      "step": 75285
    },
    {
      "entropy": 5.927372312545776,
      "epoch": 5.8578097646372305,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021728881372916743,
      "loss": 4.8363,
      "mean_token_accuracy": 0.22593884468078612,
      "num_tokens": 130599623.0,
      "step": 75290
    },
    {
      "entropy": 5.894705200195313,
      "epoch": 5.858198794008947,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002172620257827973,
      "loss": 4.8469,
      "mean_token_accuracy": 0.223989999294281,
      "num_tokens": 130608092.0,
      "step": 75295
    },
    {
      "entropy": 5.951822090148926,
      "epoch": 5.858587823380665,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00021723523871250905,
      "loss": 4.9101,
      "mean_token_accuracy": 0.22703285664319992,
      "num_tokens": 130616274.0,
      "step": 75300
    },
    {
      "entropy": 5.955262756347656,
      "epoch": 5.858976852752383,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021720845251870907,
      "loss": 4.9494,
      "mean_token_accuracy": 0.210658498108387,
      "num_tokens": 130625591.0,
      "step": 75305
    },
    {
      "entropy": 6.022744655609131,
      "epoch": 5.8593658821241,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021718166720180384,
      "loss": 4.9582,
      "mean_token_accuracy": 0.21697202771902085,
      "num_tokens": 130634754.0,
      "step": 75310
    },
    {
      "entropy": 5.9204576969146725,
      "epoch": 5.859754911495818,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002171548827621998,
      "loss": 4.8631,
      "mean_token_accuracy": 0.23571909964084625,
      "num_tokens": 130643538.0,
      "step": 75315
    },
    {
      "entropy": 5.921241760253906,
      "epoch": 5.860143940867536,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002171280992003033,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22502767592668532,
      "num_tokens": 130653277.0,
      "step": 75320
    },
    {
      "entropy": 5.942056179046631,
      "epoch": 5.8605329702392535,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021710131651652087,
      "loss": 4.9246,
      "mean_token_accuracy": 0.22603473961353301,
      "num_tokens": 130663009.0,
      "step": 75325
    },
    {
      "entropy": 5.9027002334594725,
      "epoch": 5.86092199961097,
      "grad_norm": 1.65625,
      "learning_rate": 0.00021707453471125866,
      "loss": 4.8977,
      "mean_token_accuracy": 0.21811441332101822,
      "num_tokens": 130672246.0,
      "step": 75330
    },
    {
      "entropy": 5.958973121643067,
      "epoch": 5.861311028982688,
      "grad_norm": 1.734375,
      "learning_rate": 0.00021704775378492326,
      "loss": 4.8622,
      "mean_token_accuracy": 0.22785282284021377,
      "num_tokens": 130680948.0,
      "step": 75335
    },
    {
      "entropy": 5.983928632736206,
      "epoch": 5.861700058354406,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00021702097373792095,
      "loss": 4.9251,
      "mean_token_accuracy": 0.22308722734451295,
      "num_tokens": 130689755.0,
      "step": 75340
    },
    {
      "entropy": 5.97366852760315,
      "epoch": 5.862089087726123,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021699419457065812,
      "loss": 4.8969,
      "mean_token_accuracy": 0.2192154422402382,
      "num_tokens": 130698429.0,
      "step": 75345
    },
    {
      "entropy": 6.0411553382873535,
      "epoch": 5.862478117097841,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021696741628354104,
      "loss": 5.0245,
      "mean_token_accuracy": 0.2171551451086998,
      "num_tokens": 130706811.0,
      "step": 75350
    },
    {
      "entropy": 5.911888217926025,
      "epoch": 5.862867146469559,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021694063887697614,
      "loss": 4.8303,
      "mean_token_accuracy": 0.22512923330068588,
      "num_tokens": 130714909.0,
      "step": 75355
    },
    {
      "entropy": 5.93531060218811,
      "epoch": 5.863256175841276,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021691386235136951,
      "loss": 4.8465,
      "mean_token_accuracy": 0.2270304426550865,
      "num_tokens": 130723006.0,
      "step": 75360
    },
    {
      "entropy": 5.962205696105957,
      "epoch": 5.863645205212993,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00021688708670712764,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21693944483995437,
      "num_tokens": 130731615.0,
      "step": 75365
    },
    {
      "entropy": 6.006429672241211,
      "epoch": 5.864034234584711,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021686031194465666,
      "loss": 4.998,
      "mean_token_accuracy": 0.2137029156088829,
      "num_tokens": 130740387.0,
      "step": 75370
    },
    {
      "entropy": 5.983992671966552,
      "epoch": 5.864423263956429,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021683353806436296,
      "loss": 4.8565,
      "mean_token_accuracy": 0.2252255544066429,
      "num_tokens": 130748420.0,
      "step": 75375
    },
    {
      "entropy": 5.977304410934448,
      "epoch": 5.864812293328146,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002168067650666527,
      "loss": 4.9407,
      "mean_token_accuracy": 0.21890005022287368,
      "num_tokens": 130756907.0,
      "step": 75380
    },
    {
      "entropy": 6.026253414154053,
      "epoch": 5.865201322699864,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00021677999295193224,
      "loss": 5.0743,
      "mean_token_accuracy": 0.20892684757709504,
      "num_tokens": 130765112.0,
      "step": 75385
    },
    {
      "entropy": 5.986418533325195,
      "epoch": 5.865590352071582,
      "grad_norm": 1.65625,
      "learning_rate": 0.00021675322172060764,
      "loss": 4.9168,
      "mean_token_accuracy": 0.2222522273659706,
      "num_tokens": 130773328.0,
      "step": 75390
    },
    {
      "entropy": 5.91234769821167,
      "epoch": 5.8659793814432994,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021672645137308516,
      "loss": 4.9506,
      "mean_token_accuracy": 0.21173347234725953,
      "num_tokens": 130782919.0,
      "step": 75395
    },
    {
      "entropy": 5.853524303436279,
      "epoch": 5.866368410815016,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021669968190977107,
      "loss": 4.8241,
      "mean_token_accuracy": 0.23128565698862075,
      "num_tokens": 130791021.0,
      "step": 75400
    },
    {
      "entropy": 5.959136247634888,
      "epoch": 5.866757440186734,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021667291333107153,
      "loss": 4.9498,
      "mean_token_accuracy": 0.2261599987745285,
      "num_tokens": 130799784.0,
      "step": 75405
    },
    {
      "entropy": 5.965036869049072,
      "epoch": 5.867146469558452,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021664614563739266,
      "loss": 4.8648,
      "mean_token_accuracy": 0.23410189598798753,
      "num_tokens": 130808259.0,
      "step": 75410
    },
    {
      "entropy": 5.97726583480835,
      "epoch": 5.867535498930169,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021661937882914067,
      "loss": 4.9727,
      "mean_token_accuracy": 0.21823120564222337,
      "num_tokens": 130817389.0,
      "step": 75415
    },
    {
      "entropy": 6.015581703186035,
      "epoch": 5.867924528301887,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021659261290672168,
      "loss": 4.971,
      "mean_token_accuracy": 0.21386676728725434,
      "num_tokens": 130826595.0,
      "step": 75420
    },
    {
      "entropy": 6.027264595031738,
      "epoch": 5.868313557673605,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00021656584787054184,
      "loss": 5.048,
      "mean_token_accuracy": 0.2160298928618431,
      "num_tokens": 130835723.0,
      "step": 75425
    },
    {
      "entropy": 6.023162984848023,
      "epoch": 5.8687025870453216,
      "grad_norm": 1.6875,
      "learning_rate": 0.00021653908372100728,
      "loss": 5.028,
      "mean_token_accuracy": 0.21050597429275514,
      "num_tokens": 130844830.0,
      "step": 75430
    },
    {
      "entropy": 6.018211936950683,
      "epoch": 5.869091616417039,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021651232045852403,
      "loss": 4.9642,
      "mean_token_accuracy": 0.21456619650125502,
      "num_tokens": 130853673.0,
      "step": 75435
    },
    {
      "entropy": 5.9396641731262205,
      "epoch": 5.869480645788757,
      "grad_norm": 1.5,
      "learning_rate": 0.0002164855580834983,
      "loss": 4.8173,
      "mean_token_accuracy": 0.22934497147798538,
      "num_tokens": 130862712.0,
      "step": 75440
    },
    {
      "entropy": 5.954195928573609,
      "epoch": 5.869869675160475,
      "grad_norm": 1.546875,
      "learning_rate": 0.000216458796596336,
      "loss": 4.9126,
      "mean_token_accuracy": 0.2219357505440712,
      "num_tokens": 130871260.0,
      "step": 75445
    },
    {
      "entropy": 5.9409260749816895,
      "epoch": 5.870258704532192,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021643203599744327,
      "loss": 4.8967,
      "mean_token_accuracy": 0.22026266008615494,
      "num_tokens": 130879349.0,
      "step": 75450
    },
    {
      "entropy": 5.95132737159729,
      "epoch": 5.87064773390391,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021640527628722622,
      "loss": 4.9123,
      "mean_token_accuracy": 0.2238190531730652,
      "num_tokens": 130887375.0,
      "step": 75455
    },
    {
      "entropy": 5.952106475830078,
      "epoch": 5.871036763275628,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021637851746609082,
      "loss": 4.8991,
      "mean_token_accuracy": 0.22524582892656325,
      "num_tokens": 130895803.0,
      "step": 75460
    },
    {
      "entropy": 5.937530755996704,
      "epoch": 5.8714257926473445,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021635175953444314,
      "loss": 4.9148,
      "mean_token_accuracy": 0.22142511159181594,
      "num_tokens": 130903960.0,
      "step": 75465
    },
    {
      "entropy": 5.9152813911437985,
      "epoch": 5.871814822019062,
      "grad_norm": 1.625,
      "learning_rate": 0.0002163250024926891,
      "loss": 4.8366,
      "mean_token_accuracy": 0.22593443244695663,
      "num_tokens": 130912244.0,
      "step": 75470
    },
    {
      "entropy": 5.87206449508667,
      "epoch": 5.87220385139078,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021629824634123479,
      "loss": 4.8541,
      "mean_token_accuracy": 0.23321328461170196,
      "num_tokens": 130921443.0,
      "step": 75475
    },
    {
      "entropy": 6.015424013137817,
      "epoch": 5.872592880762498,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021627149108048615,
      "loss": 4.9985,
      "mean_token_accuracy": 0.2202403575181961,
      "num_tokens": 130929850.0,
      "step": 75480
    },
    {
      "entropy": 6.000602960586548,
      "epoch": 5.872981910134215,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002162447367108492,
      "loss": 5.0084,
      "mean_token_accuracy": 0.21271976381540297,
      "num_tokens": 130938171.0,
      "step": 75485
    },
    {
      "entropy": 5.95771017074585,
      "epoch": 5.873370939505933,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021621798323272985,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22501126080751419,
      "num_tokens": 130946063.0,
      "step": 75490
    },
    {
      "entropy": 5.89755802154541,
      "epoch": 5.87375996887765,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000216191230646534,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22036340534687043,
      "num_tokens": 130953644.0,
      "step": 75495
    },
    {
      "entropy": 5.945853042602539,
      "epoch": 5.8741489982493675,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021616447895266762,
      "loss": 4.8879,
      "mean_token_accuracy": 0.22229375839233398,
      "num_tokens": 130961939.0,
      "step": 75500
    },
    {
      "entropy": 5.9587281227111815,
      "epoch": 5.874538027621085,
      "grad_norm": 1.5,
      "learning_rate": 0.00021613772815153666,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2280240088701248,
      "num_tokens": 130970513.0,
      "step": 75505
    },
    {
      "entropy": 5.903785562515258,
      "epoch": 5.874927056992803,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021611097824354697,
      "loss": 4.8377,
      "mean_token_accuracy": 0.2314414396882057,
      "num_tokens": 130978662.0,
      "step": 75510
    },
    {
      "entropy": 5.944231653213501,
      "epoch": 5.875316086364521,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002160842292291045,
      "loss": 4.9496,
      "mean_token_accuracy": 0.21866417229175567,
      "num_tokens": 130987226.0,
      "step": 75515
    },
    {
      "entropy": 5.965762567520142,
      "epoch": 5.875705115736238,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021605748110861507,
      "loss": 4.9469,
      "mean_token_accuracy": 0.22677277028560638,
      "num_tokens": 130996853.0,
      "step": 75520
    },
    {
      "entropy": 5.957023620605469,
      "epoch": 5.876094145107956,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021603073388248452,
      "loss": 4.9135,
      "mean_token_accuracy": 0.22655759006738663,
      "num_tokens": 131005582.0,
      "step": 75525
    },
    {
      "entropy": 5.971748018264771,
      "epoch": 5.876483174479673,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002160039875511186,
      "loss": 4.8913,
      "mean_token_accuracy": 0.22119861841201782,
      "num_tokens": 131014138.0,
      "step": 75530
    },
    {
      "entropy": 6.050435256958008,
      "epoch": 5.8768722038513905,
      "grad_norm": 1.6875,
      "learning_rate": 0.00021597724211492336,
      "loss": 5.0504,
      "mean_token_accuracy": 0.20884641408920288,
      "num_tokens": 131022144.0,
      "step": 75535
    },
    {
      "entropy": 6.003734064102173,
      "epoch": 5.877261233223108,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021595049757430458,
      "loss": 4.935,
      "mean_token_accuracy": 0.220840160548687,
      "num_tokens": 131031316.0,
      "step": 75540
    },
    {
      "entropy": 5.95683479309082,
      "epoch": 5.877650262594826,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021592375392966802,
      "loss": 4.9174,
      "mean_token_accuracy": 0.22201566100120546,
      "num_tokens": 131039621.0,
      "step": 75545
    },
    {
      "entropy": 5.948804903030395,
      "epoch": 5.878039291966544,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021589701118141942,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22742396146059035,
      "num_tokens": 131048279.0,
      "step": 75550
    },
    {
      "entropy": 5.918148231506348,
      "epoch": 5.878428321338261,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021587026932996456,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22408656775951385,
      "num_tokens": 131056895.0,
      "step": 75555
    },
    {
      "entropy": 5.9957688331604,
      "epoch": 5.878817350709978,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021584352837570936,
      "loss": 4.9947,
      "mean_token_accuracy": 0.21729426831007004,
      "num_tokens": 131066076.0,
      "step": 75560
    },
    {
      "entropy": 5.8761138916015625,
      "epoch": 5.879206380081696,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021581678831905939,
      "loss": 4.8314,
      "mean_token_accuracy": 0.22739178836345672,
      "num_tokens": 131074842.0,
      "step": 75565
    },
    {
      "entropy": 5.952015113830567,
      "epoch": 5.8795954094534135,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021579004916042044,
      "loss": 5.0239,
      "mean_token_accuracy": 0.20771122127771377,
      "num_tokens": 131083925.0,
      "step": 75570
    },
    {
      "entropy": 5.985554981231689,
      "epoch": 5.879984438825131,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00021576331090019825,
      "loss": 4.9374,
      "mean_token_accuracy": 0.21668141782283784,
      "num_tokens": 131092088.0,
      "step": 75575
    },
    {
      "entropy": 5.984002256393433,
      "epoch": 5.880373468196849,
      "grad_norm": 1.671875,
      "learning_rate": 0.00021573657353879855,
      "loss": 4.9691,
      "mean_token_accuracy": 0.21702119410037995,
      "num_tokens": 131100485.0,
      "step": 75580
    },
    {
      "entropy": 5.943580198287964,
      "epoch": 5.880762497568567,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021570983707662705,
      "loss": 4.8424,
      "mean_token_accuracy": 0.23992462158203126,
      "num_tokens": 131109431.0,
      "step": 75585
    },
    {
      "entropy": 5.955830240249634,
      "epoch": 5.881151526940284,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021568310151408939,
      "loss": 4.9192,
      "mean_token_accuracy": 0.22028935998678206,
      "num_tokens": 131117861.0,
      "step": 75590
    },
    {
      "entropy": 5.976267004013062,
      "epoch": 5.881540556312002,
      "grad_norm": 1.5,
      "learning_rate": 0.00021565636685159117,
      "loss": 4.976,
      "mean_token_accuracy": 0.219989512860775,
      "num_tokens": 131126578.0,
      "step": 75595
    },
    {
      "entropy": 5.950083589553833,
      "epoch": 5.881929585683719,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0002156296330895382,
      "loss": 4.9266,
      "mean_token_accuracy": 0.22697990387678146,
      "num_tokens": 131135517.0,
      "step": 75600
    },
    {
      "entropy": 6.036066198348999,
      "epoch": 5.8823186150554365,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021560290022833606,
      "loss": 5.0318,
      "mean_token_accuracy": 0.2147227331995964,
      "num_tokens": 131144211.0,
      "step": 75605
    },
    {
      "entropy": 5.97566442489624,
      "epoch": 5.882707644427154,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002155761682683904,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2219949707388878,
      "num_tokens": 131152937.0,
      "step": 75610
    },
    {
      "entropy": 6.014400243759155,
      "epoch": 5.883096673798872,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021554943721010673,
      "loss": 4.9274,
      "mean_token_accuracy": 0.2193492978811264,
      "num_tokens": 131161557.0,
      "step": 75615
    },
    {
      "entropy": 5.976441287994385,
      "epoch": 5.88348570317059,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002155227070538907,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2224429726600647,
      "num_tokens": 131170628.0,
      "step": 75620
    },
    {
      "entropy": 5.980197525024414,
      "epoch": 5.883874732542307,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021549597780014807,
      "loss": 4.9859,
      "mean_token_accuracy": 0.21812517642974855,
      "num_tokens": 131179047.0,
      "step": 75625
    },
    {
      "entropy": 6.012881422042847,
      "epoch": 5.884263761914024,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002154692494492842,
      "loss": 4.9725,
      "mean_token_accuracy": 0.2183452218770981,
      "num_tokens": 131187743.0,
      "step": 75630
    },
    {
      "entropy": 6.016065406799316,
      "epoch": 5.884652791285742,
      "grad_norm": 1.734375,
      "learning_rate": 0.00021544252200170465,
      "loss": 5.064,
      "mean_token_accuracy": 0.21276151090860368,
      "num_tokens": 131196572.0,
      "step": 75635
    },
    {
      "entropy": 5.961779594421387,
      "epoch": 5.8850418206574595,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002154157954578151,
      "loss": 4.8773,
      "mean_token_accuracy": 0.2208438739180565,
      "num_tokens": 131205332.0,
      "step": 75640
    },
    {
      "entropy": 5.927547359466553,
      "epoch": 5.885430850029177,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000215389069818021,
      "loss": 4.8432,
      "mean_token_accuracy": 0.2271549090743065,
      "num_tokens": 131214455.0,
      "step": 75645
    },
    {
      "entropy": 6.0099348545074465,
      "epoch": 5.885819879400895,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021536234508272783,
      "loss": 5.0655,
      "mean_token_accuracy": 0.2201812669634819,
      "num_tokens": 131222855.0,
      "step": 75650
    },
    {
      "entropy": 5.942979335784912,
      "epoch": 5.886208908772613,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021533562125234125,
      "loss": 4.8944,
      "mean_token_accuracy": 0.22255752682685853,
      "num_tokens": 131231899.0,
      "step": 75655
    },
    {
      "entropy": 5.9647557735443115,
      "epoch": 5.88659793814433,
      "grad_norm": 1.625,
      "learning_rate": 0.00021530889832726652,
      "loss": 4.8919,
      "mean_token_accuracy": 0.22657197266817092,
      "num_tokens": 131239696.0,
      "step": 75660
    },
    {
      "entropy": 5.964805841445923,
      "epoch": 5.886986967516047,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021528217630790937,
      "loss": 4.9454,
      "mean_token_accuracy": 0.22092599272727967,
      "num_tokens": 131248495.0,
      "step": 75665
    },
    {
      "entropy": 5.947979259490967,
      "epoch": 5.887375996887765,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021525545519467506,
      "loss": 4.8991,
      "mean_token_accuracy": 0.22491041719913482,
      "num_tokens": 131256662.0,
      "step": 75670
    },
    {
      "entropy": 6.076389455795288,
      "epoch": 5.8877650262594825,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021522873498796912,
      "loss": 5.1159,
      "mean_token_accuracy": 0.20593930780887604,
      "num_tokens": 131266424.0,
      "step": 75675
    },
    {
      "entropy": 5.988425159454346,
      "epoch": 5.8881540556312,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021520201568819708,
      "loss": 4.9523,
      "mean_token_accuracy": 0.21846496760845185,
      "num_tokens": 131275070.0,
      "step": 75680
    },
    {
      "entropy": 5.953958511352539,
      "epoch": 5.888543085002918,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00021517529729576425,
      "loss": 4.8626,
      "mean_token_accuracy": 0.22399222552776338,
      "num_tokens": 131283138.0,
      "step": 75685
    },
    {
      "entropy": 6.003613615036011,
      "epoch": 5.888932114374636,
      "grad_norm": 1.703125,
      "learning_rate": 0.00021514857981107604,
      "loss": 4.9059,
      "mean_token_accuracy": 0.22280484735965728,
      "num_tokens": 131291284.0,
      "step": 75690
    },
    {
      "entropy": 5.915501117706299,
      "epoch": 5.889321143746352,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021512186323453786,
      "loss": 4.875,
      "mean_token_accuracy": 0.2244104340672493,
      "num_tokens": 131300377.0,
      "step": 75695
    },
    {
      "entropy": 5.974612188339234,
      "epoch": 5.88971017311807,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021509514756655514,
      "loss": 4.9166,
      "mean_token_accuracy": 0.2197792962193489,
      "num_tokens": 131309074.0,
      "step": 75700
    },
    {
      "entropy": 5.906082582473755,
      "epoch": 5.890099202489788,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021506843280753318,
      "loss": 4.9107,
      "mean_token_accuracy": 0.21922635287046432,
      "num_tokens": 131318028.0,
      "step": 75705
    },
    {
      "entropy": 5.96537675857544,
      "epoch": 5.8904882318615055,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021504171895787738,
      "loss": 4.954,
      "mean_token_accuracy": 0.2240684524178505,
      "num_tokens": 131326141.0,
      "step": 75710
    },
    {
      "entropy": 5.978250026702881,
      "epoch": 5.890877261233223,
      "grad_norm": 1.46875,
      "learning_rate": 0.000215015006017993,
      "loss": 5.0073,
      "mean_token_accuracy": 0.21652478575706482,
      "num_tokens": 131334958.0,
      "step": 75715
    },
    {
      "entropy": 6.014389896392823,
      "epoch": 5.891266290604941,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002149882939882855,
      "loss": 5.0207,
      "mean_token_accuracy": 0.2175044909119606,
      "num_tokens": 131343752.0,
      "step": 75720
    },
    {
      "entropy": 5.9900815963745115,
      "epoch": 5.891655319976659,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021496158286916011,
      "loss": 4.9677,
      "mean_token_accuracy": 0.21860455423593522,
      "num_tokens": 131352429.0,
      "step": 75725
    },
    {
      "entropy": 5.982416868209839,
      "epoch": 5.892044349348376,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021493487266102202,
      "loss": 4.9676,
      "mean_token_accuracy": 0.21308455169200896,
      "num_tokens": 131360437.0,
      "step": 75730
    },
    {
      "entropy": 5.852781820297241,
      "epoch": 5.892433378720093,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021490816336427676,
      "loss": 4.8139,
      "mean_token_accuracy": 0.22976686507463456,
      "num_tokens": 131369126.0,
      "step": 75735
    },
    {
      "entropy": 6.021676445007325,
      "epoch": 5.892822408091811,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021488145497932949,
      "loss": 5.0262,
      "mean_token_accuracy": 0.21673137694597244,
      "num_tokens": 131378062.0,
      "step": 75740
    },
    {
      "entropy": 5.962700319290161,
      "epoch": 5.8932114374635285,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021485474750658536,
      "loss": 4.9495,
      "mean_token_accuracy": 0.22273496985435487,
      "num_tokens": 131386244.0,
      "step": 75745
    },
    {
      "entropy": 5.972991895675659,
      "epoch": 5.893600466835246,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021482804094644976,
      "loss": 4.9435,
      "mean_token_accuracy": 0.22451961040496826,
      "num_tokens": 131394850.0,
      "step": 75750
    },
    {
      "entropy": 6.018024539947509,
      "epoch": 5.893989496206964,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00021480133529932789,
      "loss": 4.943,
      "mean_token_accuracy": 0.2180756837129593,
      "num_tokens": 131402825.0,
      "step": 75755
    },
    {
      "entropy": 5.971045970916748,
      "epoch": 5.894378525578681,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002147746305656249,
      "loss": 4.9517,
      "mean_token_accuracy": 0.2148667871952057,
      "num_tokens": 131410839.0,
      "step": 75760
    },
    {
      "entropy": 5.9775135040283205,
      "epoch": 5.894767554950398,
      "grad_norm": 1.578125,
      "learning_rate": 0.000214747926745746,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21701246201992036,
      "num_tokens": 131418582.0,
      "step": 75765
    },
    {
      "entropy": 5.949578952789307,
      "epoch": 5.895156584322116,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021472122384009646,
      "loss": 4.8899,
      "mean_token_accuracy": 0.22478539496660233,
      "num_tokens": 131426813.0,
      "step": 75770
    },
    {
      "entropy": 5.9648108959198,
      "epoch": 5.895545613693834,
      "grad_norm": 1.625,
      "learning_rate": 0.0002146945218490814,
      "loss": 4.9678,
      "mean_token_accuracy": 0.21769788563251496,
      "num_tokens": 131436065.0,
      "step": 75775
    },
    {
      "entropy": 5.971624279022217,
      "epoch": 5.8959346430655515,
      "grad_norm": 1.6875,
      "learning_rate": 0.00021466782077310598,
      "loss": 4.9035,
      "mean_token_accuracy": 0.22813172936439513,
      "num_tokens": 131444547.0,
      "step": 75780
    },
    {
      "entropy": 6.011195516586303,
      "epoch": 5.896323672437269,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002146411206125754,
      "loss": 4.9152,
      "mean_token_accuracy": 0.22330242544412612,
      "num_tokens": 131452904.0,
      "step": 75785
    },
    {
      "entropy": 5.957762861251831,
      "epoch": 5.896712701808987,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021461442136789472,
      "loss": 4.9228,
      "mean_token_accuracy": 0.21901368647813796,
      "num_tokens": 131462119.0,
      "step": 75790
    },
    {
      "entropy": 5.897545289993286,
      "epoch": 5.897101731180705,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021458772303946895,
      "loss": 4.789,
      "mean_token_accuracy": 0.22688434422016143,
      "num_tokens": 131470916.0,
      "step": 75795
    },
    {
      "entropy": 5.844310235977173,
      "epoch": 5.897490760552421,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021456102562770346,
      "loss": 4.7558,
      "mean_token_accuracy": 0.23435195535421371,
      "num_tokens": 131479087.0,
      "step": 75800
    },
    {
      "entropy": 5.966272354125977,
      "epoch": 5.897879789924139,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002145343291330032,
      "loss": 4.964,
      "mean_token_accuracy": 0.21106974333524703,
      "num_tokens": 131487818.0,
      "step": 75805
    },
    {
      "entropy": 5.955660581588745,
      "epoch": 5.898268819295857,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021450763355577318,
      "loss": 4.9388,
      "mean_token_accuracy": 0.2234267547726631,
      "num_tokens": 131497464.0,
      "step": 75810
    },
    {
      "entropy": 5.986806726455688,
      "epoch": 5.8986578486675745,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021448093889641867,
      "loss": 4.9184,
      "mean_token_accuracy": 0.21886376142501832,
      "num_tokens": 131505909.0,
      "step": 75815
    },
    {
      "entropy": 5.963280248641968,
      "epoch": 5.899046878039292,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002144542451553445,
      "loss": 4.9467,
      "mean_token_accuracy": 0.22219013422727585,
      "num_tokens": 131515189.0,
      "step": 75820
    },
    {
      "entropy": 6.036799669265747,
      "epoch": 5.89943590741101,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002144275523329558,
      "loss": 4.9664,
      "mean_token_accuracy": 0.2268372133374214,
      "num_tokens": 131524076.0,
      "step": 75825
    },
    {
      "entropy": 5.994196081161499,
      "epoch": 5.899824936782727,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021440086042965762,
      "loss": 4.9433,
      "mean_token_accuracy": 0.22480585873126985,
      "num_tokens": 131532876.0,
      "step": 75830
    },
    {
      "entropy": 5.974919223785401,
      "epoch": 5.900213966154444,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021437416944585487,
      "loss": 4.9732,
      "mean_token_accuracy": 0.22314099818468094,
      "num_tokens": 131541669.0,
      "step": 75835
    },
    {
      "entropy": 5.964779567718506,
      "epoch": 5.900602995526162,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021434747938195266,
      "loss": 4.9791,
      "mean_token_accuracy": 0.21351983398199081,
      "num_tokens": 131550337.0,
      "step": 75840
    },
    {
      "entropy": 5.986673831939697,
      "epoch": 5.90099202489788,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021432079023835587,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21799728125333787,
      "num_tokens": 131559893.0,
      "step": 75845
    },
    {
      "entropy": 6.007574796676636,
      "epoch": 5.9013810542695975,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002142941020154695,
      "loss": 5.0267,
      "mean_token_accuracy": 0.21164468973875045,
      "num_tokens": 131568557.0,
      "step": 75850
    },
    {
      "entropy": 5.979750347137451,
      "epoch": 5.901770083641315,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00021426741471369855,
      "loss": 4.9651,
      "mean_token_accuracy": 0.22452736347913743,
      "num_tokens": 131577298.0,
      "step": 75855
    },
    {
      "entropy": 5.978922033309937,
      "epoch": 5.902159113013033,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00021424072833344781,
      "loss": 4.9546,
      "mean_token_accuracy": 0.21701303422451018,
      "num_tokens": 131585966.0,
      "step": 75860
    },
    {
      "entropy": 6.023138761520386,
      "epoch": 5.90254814238475,
      "grad_norm": 1.703125,
      "learning_rate": 0.0002142140428751224,
      "loss": 4.9299,
      "mean_token_accuracy": 0.21994149386882783,
      "num_tokens": 131594618.0,
      "step": 75865
    },
    {
      "entropy": 5.991744470596314,
      "epoch": 5.902937171756467,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002141873583391271,
      "loss": 4.9767,
      "mean_token_accuracy": 0.2223924919962883,
      "num_tokens": 131603634.0,
      "step": 75870
    },
    {
      "entropy": 5.926083517074585,
      "epoch": 5.903326201128185,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021416067472586692,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22827312499284744,
      "num_tokens": 131611710.0,
      "step": 75875
    },
    {
      "entropy": 5.898599147796631,
      "epoch": 5.903715230499903,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00021413399203574658,
      "loss": 4.9603,
      "mean_token_accuracy": 0.22203667908906938,
      "num_tokens": 131620636.0,
      "step": 75880
    },
    {
      "entropy": 5.914532995223999,
      "epoch": 5.9041042598716205,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021410731026917106,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22430076599121093,
      "num_tokens": 131629074.0,
      "step": 75885
    },
    {
      "entropy": 5.984271478652954,
      "epoch": 5.904493289243338,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002140806294265452,
      "loss": 4.8479,
      "mean_token_accuracy": 0.22373515963554383,
      "num_tokens": 131637398.0,
      "step": 75890
    },
    {
      "entropy": 5.954380130767822,
      "epoch": 5.904882318615055,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002140539495082739,
      "loss": 4.906,
      "mean_token_accuracy": 0.22087418884038926,
      "num_tokens": 131645732.0,
      "step": 75895
    },
    {
      "entropy": 5.9468159675598145,
      "epoch": 5.905271347986773,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021402727051476184,
      "loss": 4.9431,
      "mean_token_accuracy": 0.2222451537847519,
      "num_tokens": 131654519.0,
      "step": 75900
    },
    {
      "entropy": 6.002316236495972,
      "epoch": 5.90566037735849,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002140005924464139,
      "loss": 5.0414,
      "mean_token_accuracy": 0.20920560956001283,
      "num_tokens": 131663731.0,
      "step": 75905
    },
    {
      "entropy": 6.0468282222747805,
      "epoch": 5.906049406730208,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021397391530363487,
      "loss": 4.997,
      "mean_token_accuracy": 0.21522744446992875,
      "num_tokens": 131672659.0,
      "step": 75910
    },
    {
      "entropy": 5.945220851898194,
      "epoch": 5.906438436101926,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021394723908682955,
      "loss": 4.8218,
      "mean_token_accuracy": 0.22444464564323424,
      "num_tokens": 131680978.0,
      "step": 75915
    },
    {
      "entropy": 5.9075664520263675,
      "epoch": 5.9068274654736435,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00021392056379640267,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2243838056921959,
      "num_tokens": 131689491.0,
      "step": 75920
    },
    {
      "entropy": 5.933395576477051,
      "epoch": 5.907216494845361,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021389388943275907,
      "loss": 4.9745,
      "mean_token_accuracy": 0.21307849735021592,
      "num_tokens": 131698519.0,
      "step": 75925
    },
    {
      "entropy": 5.941667938232422,
      "epoch": 5.907605524217079,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00021386721599630333,
      "loss": 4.9167,
      "mean_token_accuracy": 0.22358036637306214,
      "num_tokens": 131707035.0,
      "step": 75930
    },
    {
      "entropy": 5.952823448181152,
      "epoch": 5.907994553588796,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021384054348744035,
      "loss": 4.929,
      "mean_token_accuracy": 0.22395991832017897,
      "num_tokens": 131715656.0,
      "step": 75935
    },
    {
      "entropy": 5.967580366134643,
      "epoch": 5.908383582960513,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002138138719065748,
      "loss": 4.9716,
      "mean_token_accuracy": 0.22600956559181212,
      "num_tokens": 131723916.0,
      "step": 75940
    },
    {
      "entropy": 5.977602481842041,
      "epoch": 5.908772612332231,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002137872012541114,
      "loss": 4.9865,
      "mean_token_accuracy": 0.21291892379522323,
      "num_tokens": 131732818.0,
      "step": 75945
    },
    {
      "entropy": 5.90695013999939,
      "epoch": 5.909161641703949,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021376053153045473,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22568605691194535,
      "num_tokens": 131741287.0,
      "step": 75950
    },
    {
      "entropy": 5.988749217987061,
      "epoch": 5.9095506710756665,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021373386273600953,
      "loss": 4.938,
      "mean_token_accuracy": 0.21693199574947358,
      "num_tokens": 131749167.0,
      "step": 75955
    },
    {
      "entropy": 5.954548263549805,
      "epoch": 5.909939700447384,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021370719487118046,
      "loss": 4.9452,
      "mean_token_accuracy": 0.22163229286670685,
      "num_tokens": 131757588.0,
      "step": 75960
    },
    {
      "entropy": 6.000315570831299,
      "epoch": 5.910328729819101,
      "grad_norm": 1.5,
      "learning_rate": 0.0002136805279363721,
      "loss": 4.9828,
      "mean_token_accuracy": 0.21620805710554122,
      "num_tokens": 131766128.0,
      "step": 75965
    },
    {
      "entropy": 6.072630119323731,
      "epoch": 5.910717759190819,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021365386193198917,
      "loss": 4.9935,
      "mean_token_accuracy": 0.21735645234584808,
      "num_tokens": 131773923.0,
      "step": 75970
    },
    {
      "entropy": 5.931922960281372,
      "epoch": 5.911106788562536,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002136271968584363,
      "loss": 4.8731,
      "mean_token_accuracy": 0.22980883568525315,
      "num_tokens": 131782359.0,
      "step": 75975
    },
    {
      "entropy": 5.96781587600708,
      "epoch": 5.911495817934254,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021360053271611795,
      "loss": 4.9267,
      "mean_token_accuracy": 0.2190510094165802,
      "num_tokens": 131790394.0,
      "step": 75980
    },
    {
      "entropy": 5.9741966247558596,
      "epoch": 5.911884847305972,
      "grad_norm": 1.59375,
      "learning_rate": 0.00021357386950543884,
      "loss": 5.0478,
      "mean_token_accuracy": 0.20757967829704285,
      "num_tokens": 131798812.0,
      "step": 75985
    },
    {
      "entropy": 5.903533887863159,
      "epoch": 5.912273876677689,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021354720722680347,
      "loss": 4.8859,
      "mean_token_accuracy": 0.2236032098531723,
      "num_tokens": 131806938.0,
      "step": 75990
    },
    {
      "entropy": 5.955878829956054,
      "epoch": 5.912662906049407,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021352054588061632,
      "loss": 4.9818,
      "mean_token_accuracy": 0.21862297207117082,
      "num_tokens": 131815995.0,
      "step": 75995
    },
    {
      "entropy": 5.9811145782470705,
      "epoch": 5.913051935421124,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002134938854672821,
      "loss": 4.8926,
      "mean_token_accuracy": 0.2151590868830681,
      "num_tokens": 131824258.0,
      "step": 76000
    },
    {
      "entropy": 5.994270086288452,
      "epoch": 5.913440964792842,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021346722598720536,
      "loss": 4.9781,
      "mean_token_accuracy": 0.217181196808815,
      "num_tokens": 131832410.0,
      "step": 76005
    },
    {
      "entropy": 5.981575441360474,
      "epoch": 5.913829994164559,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021344056744079044,
      "loss": 4.9068,
      "mean_token_accuracy": 0.22308890223503114,
      "num_tokens": 131840753.0,
      "step": 76010
    },
    {
      "entropy": 5.948577499389648,
      "epoch": 5.914219023536277,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021341390982844194,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21866722106933595,
      "num_tokens": 131848724.0,
      "step": 76015
    },
    {
      "entropy": 5.974097061157226,
      "epoch": 5.914608052907995,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021338725315056434,
      "loss": 5.0642,
      "mean_token_accuracy": 0.2120060384273529,
      "num_tokens": 131857393.0,
      "step": 76020
    },
    {
      "entropy": 6.040370273590088,
      "epoch": 5.914997082279712,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021336059740756214,
      "loss": 5.0559,
      "mean_token_accuracy": 0.20803145468235015,
      "num_tokens": 131867450.0,
      "step": 76025
    },
    {
      "entropy": 5.994358205795288,
      "epoch": 5.915386111651429,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021333394259983975,
      "loss": 4.9108,
      "mean_token_accuracy": 0.22007979452610016,
      "num_tokens": 131875494.0,
      "step": 76030
    },
    {
      "entropy": 5.965454530715943,
      "epoch": 5.915775141023147,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021330728872780158,
      "loss": 4.9792,
      "mean_token_accuracy": 0.21237826049327851,
      "num_tokens": 131884867.0,
      "step": 76035
    },
    {
      "entropy": 6.029264116287232,
      "epoch": 5.916164170394865,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021328063579185213,
      "loss": 5.0149,
      "mean_token_accuracy": 0.21859276294708252,
      "num_tokens": 131893173.0,
      "step": 76040
    },
    {
      "entropy": 5.925900077819824,
      "epoch": 5.916553199766582,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002132539837923958,
      "loss": 4.9276,
      "mean_token_accuracy": 0.22561073899269105,
      "num_tokens": 131901696.0,
      "step": 76045
    },
    {
      "entropy": 5.928811073303223,
      "epoch": 5.9169422291383,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021322733272983703,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22970204055309296,
      "num_tokens": 131910066.0,
      "step": 76050
    },
    {
      "entropy": 5.891610193252563,
      "epoch": 5.917331258510018,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021320068260458015,
      "loss": 4.8506,
      "mean_token_accuracy": 0.224024261534214,
      "num_tokens": 131918169.0,
      "step": 76055
    },
    {
      "entropy": 5.898797035217285,
      "epoch": 5.917720287881735,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021317403341702945,
      "loss": 4.9005,
      "mean_token_accuracy": 0.21719679832458497,
      "num_tokens": 131927746.0,
      "step": 76060
    },
    {
      "entropy": 6.022747659683228,
      "epoch": 5.918109317253453,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021314738516758946,
      "loss": 5.0116,
      "mean_token_accuracy": 0.2127477928996086,
      "num_tokens": 131937044.0,
      "step": 76065
    },
    {
      "entropy": 5.946790599822998,
      "epoch": 5.91849834662517,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002131207378566644,
      "loss": 4.9416,
      "mean_token_accuracy": 0.22318791896104812,
      "num_tokens": 131945990.0,
      "step": 76070
    },
    {
      "entropy": 5.931189107894897,
      "epoch": 5.918887375996888,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021309409148465876,
      "loss": 4.8261,
      "mean_token_accuracy": 0.22401141375303268,
      "num_tokens": 131954153.0,
      "step": 76075
    },
    {
      "entropy": 5.966889142990112,
      "epoch": 5.919276405368605,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002130674460519767,
      "loss": 4.9204,
      "mean_token_accuracy": 0.22252853214740753,
      "num_tokens": 131962628.0,
      "step": 76080
    },
    {
      "entropy": 5.965726661682129,
      "epoch": 5.919665434740323,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021304080155902256,
      "loss": 4.9884,
      "mean_token_accuracy": 0.21511397659778594,
      "num_tokens": 131971158.0,
      "step": 76085
    },
    {
      "entropy": 5.9635313034057615,
      "epoch": 5.920054464112041,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021301415800620066,
      "loss": 4.931,
      "mean_token_accuracy": 0.22572052329778672,
      "num_tokens": 131979854.0,
      "step": 76090
    },
    {
      "entropy": 6.051819372177124,
      "epoch": 5.9204434934837575,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021298751539391525,
      "loss": 5.0362,
      "mean_token_accuracy": 0.2150921791791916,
      "num_tokens": 131987271.0,
      "step": 76095
    },
    {
      "entropy": 6.021653890609741,
      "epoch": 5.920832522855475,
      "grad_norm": 1.640625,
      "learning_rate": 0.0002129608737225706,
      "loss": 5.0652,
      "mean_token_accuracy": 0.21265406161546707,
      "num_tokens": 131995897.0,
      "step": 76100
    },
    {
      "entropy": 5.9407209873199465,
      "epoch": 5.921221552227193,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021293423299257092,
      "loss": 4.8781,
      "mean_token_accuracy": 0.22856495380401612,
      "num_tokens": 132004723.0,
      "step": 76105
    },
    {
      "entropy": 6.022292995452881,
      "epoch": 5.921610581598911,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021290759320432046,
      "loss": 4.9769,
      "mean_token_accuracy": 0.21723985373973848,
      "num_tokens": 132013345.0,
      "step": 76110
    },
    {
      "entropy": 6.000735950469971,
      "epoch": 5.921999610970628,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002128809543582234,
      "loss": 5.0195,
      "mean_token_accuracy": 0.21672162264585496,
      "num_tokens": 132022206.0,
      "step": 76115
    },
    {
      "entropy": 5.993424129486084,
      "epoch": 5.922388640342346,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021285431645468405,
      "loss": 4.9622,
      "mean_token_accuracy": 0.21868382543325424,
      "num_tokens": 132031316.0,
      "step": 76120
    },
    {
      "entropy": 6.041905069351197,
      "epoch": 5.922777669714064,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021282767949410653,
      "loss": 4.9919,
      "mean_token_accuracy": 0.2187298357486725,
      "num_tokens": 132039836.0,
      "step": 76125
    },
    {
      "entropy": 6.018416547775269,
      "epoch": 5.923166699085781,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021280104347689495,
      "loss": 5.0108,
      "mean_token_accuracy": 0.20694769769906998,
      "num_tokens": 132048773.0,
      "step": 76130
    },
    {
      "entropy": 5.951013565063477,
      "epoch": 5.923555728457498,
      "grad_norm": 1.625,
      "learning_rate": 0.00021277440840345363,
      "loss": 4.882,
      "mean_token_accuracy": 0.22289179861545563,
      "num_tokens": 132056645.0,
      "step": 76135
    },
    {
      "entropy": 5.867643356323242,
      "epoch": 5.923944757829216,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021274777427418657,
      "loss": 4.8619,
      "mean_token_accuracy": 0.21571651548147203,
      "num_tokens": 132065400.0,
      "step": 76140
    },
    {
      "entropy": 5.929050016403198,
      "epoch": 5.924333787200934,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021272114108949797,
      "loss": 4.9772,
      "mean_token_accuracy": 0.21824582666158676,
      "num_tokens": 132073985.0,
      "step": 76145
    },
    {
      "entropy": 5.92439227104187,
      "epoch": 5.924722816572651,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002126945088497919,
      "loss": 4.8931,
      "mean_token_accuracy": 0.22408050000667573,
      "num_tokens": 132082711.0,
      "step": 76150
    },
    {
      "entropy": 5.917526769638061,
      "epoch": 5.925111845944369,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002126678775554725,
      "loss": 4.8491,
      "mean_token_accuracy": 0.22800713181495666,
      "num_tokens": 132091580.0,
      "step": 76155
    },
    {
      "entropy": 5.962390470504761,
      "epoch": 5.925500875316087,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002126412472069439,
      "loss": 4.9093,
      "mean_token_accuracy": 0.22120893895626068,
      "num_tokens": 132100183.0,
      "step": 76160
    },
    {
      "entropy": 5.968575477600098,
      "epoch": 5.9258899046878035,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021261461780461006,
      "loss": 4.9366,
      "mean_token_accuracy": 0.220784692466259,
      "num_tokens": 132109319.0,
      "step": 76165
    },
    {
      "entropy": 5.963030672073364,
      "epoch": 5.926278934059521,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021258798934887513,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21764845103025438,
      "num_tokens": 132118045.0,
      "step": 76170
    },
    {
      "entropy": 6.040059900283813,
      "epoch": 5.926667963431239,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021256136184014313,
      "loss": 5.0143,
      "mean_token_accuracy": 0.2187262937426567,
      "num_tokens": 132126714.0,
      "step": 76175
    },
    {
      "entropy": 5.939790105819702,
      "epoch": 5.927056992802957,
      "grad_norm": 1.5,
      "learning_rate": 0.00021253473527881806,
      "loss": 4.9163,
      "mean_token_accuracy": 0.22141444981098174,
      "num_tokens": 132135603.0,
      "step": 76180
    },
    {
      "entropy": 6.035864686965942,
      "epoch": 5.927446022174674,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021250810966530392,
      "loss": 4.9687,
      "mean_token_accuracy": 0.2188978001475334,
      "num_tokens": 132145306.0,
      "step": 76185
    },
    {
      "entropy": 5.990002679824829,
      "epoch": 5.927835051546392,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002124814850000048,
      "loss": 4.9587,
      "mean_token_accuracy": 0.21399222314357758,
      "num_tokens": 132153336.0,
      "step": 76190
    },
    {
      "entropy": 5.995144510269165,
      "epoch": 5.92822408091811,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021245486128332458,
      "loss": 4.9018,
      "mean_token_accuracy": 0.22614268064498902,
      "num_tokens": 132161649.0,
      "step": 76195
    },
    {
      "entropy": 5.962189674377441,
      "epoch": 5.9286131102898265,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021242823851566733,
      "loss": 4.893,
      "mean_token_accuracy": 0.22498476207256318,
      "num_tokens": 132170491.0,
      "step": 76200
    },
    {
      "entropy": 6.002477693557739,
      "epoch": 5.929002139661544,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021240161669743696,
      "loss": 5.0142,
      "mean_token_accuracy": 0.2159342497587204,
      "num_tokens": 132178635.0,
      "step": 76205
    },
    {
      "entropy": 5.905142402648925,
      "epoch": 5.929391169033262,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002123749958290374,
      "loss": 4.8375,
      "mean_token_accuracy": 0.22697214037179947,
      "num_tokens": 132188293.0,
      "step": 76210
    },
    {
      "entropy": 5.90825629234314,
      "epoch": 5.92978019840498,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002123483759108727,
      "loss": 4.8609,
      "mean_token_accuracy": 0.22335285395383836,
      "num_tokens": 132197184.0,
      "step": 76215
    },
    {
      "entropy": 5.984269094467163,
      "epoch": 5.930169227776697,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021232175694334653,
      "loss": 4.9772,
      "mean_token_accuracy": 0.22193194478750228,
      "num_tokens": 132205388.0,
      "step": 76220
    },
    {
      "entropy": 5.978462791442871,
      "epoch": 5.930558257148415,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021229513892686303,
      "loss": 4.9827,
      "mean_token_accuracy": 0.21611057370901107,
      "num_tokens": 132213711.0,
      "step": 76225
    },
    {
      "entropy": 6.034187889099121,
      "epoch": 5.930947286520132,
      "grad_norm": 1.59375,
      "learning_rate": 0.000212268521861826,
      "loss": 5.0246,
      "mean_token_accuracy": 0.21345863193273545,
      "num_tokens": 132222295.0,
      "step": 76230
    },
    {
      "entropy": 6.016640090942383,
      "epoch": 5.9313363158918495,
      "grad_norm": 1.703125,
      "learning_rate": 0.0002122419057486393,
      "loss": 4.9208,
      "mean_token_accuracy": 0.218939471244812,
      "num_tokens": 132230750.0,
      "step": 76235
    },
    {
      "entropy": 6.001356315612793,
      "epoch": 5.931725345263567,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021221529058770673,
      "loss": 4.9121,
      "mean_token_accuracy": 0.2259412318468094,
      "num_tokens": 132238385.0,
      "step": 76240
    },
    {
      "entropy": 5.951796007156372,
      "epoch": 5.932114374635285,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021218867637943224,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21261410564184188,
      "num_tokens": 132247702.0,
      "step": 76245
    },
    {
      "entropy": 6.01995816230774,
      "epoch": 5.932503404007003,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002121620631242196,
      "loss": 5.0102,
      "mean_token_accuracy": 0.21508917659521104,
      "num_tokens": 132256766.0,
      "step": 76250
    },
    {
      "entropy": 5.9945070266723635,
      "epoch": 5.93289243337872,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021213545082247266,
      "loss": 4.9246,
      "mean_token_accuracy": 0.22606377601623534,
      "num_tokens": 132265034.0,
      "step": 76255
    },
    {
      "entropy": 5.953047609329223,
      "epoch": 5.933281462750438,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021210883947459509,
      "loss": 4.8998,
      "mean_token_accuracy": 0.23012032210826874,
      "num_tokens": 132273721.0,
      "step": 76260
    },
    {
      "entropy": 5.9472228527069095,
      "epoch": 5.933670492122156,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021208222908099085,
      "loss": 4.9148,
      "mean_token_accuracy": 0.22382429838180543,
      "num_tokens": 132282890.0,
      "step": 76265
    },
    {
      "entropy": 5.977645397186279,
      "epoch": 5.9340595214938725,
      "grad_norm": 1.5,
      "learning_rate": 0.0002120556196420636,
      "loss": 4.9666,
      "mean_token_accuracy": 0.2156862735748291,
      "num_tokens": 132291509.0,
      "step": 76270
    },
    {
      "entropy": 5.93904480934143,
      "epoch": 5.93444855086559,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021202901115821728,
      "loss": 4.8251,
      "mean_token_accuracy": 0.2253840908408165,
      "num_tokens": 132299762.0,
      "step": 76275
    },
    {
      "entropy": 5.933605527877807,
      "epoch": 5.934837580237308,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002120024036298554,
      "loss": 4.9213,
      "mean_token_accuracy": 0.22036652714014054,
      "num_tokens": 132308573.0,
      "step": 76280
    },
    {
      "entropy": 5.9672215461730955,
      "epoch": 5.935226609609026,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021197579705738173,
      "loss": 4.9822,
      "mean_token_accuracy": 0.21302978843450546,
      "num_tokens": 132317211.0,
      "step": 76285
    },
    {
      "entropy": 5.965671253204346,
      "epoch": 5.935615638980743,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021194919144120006,
      "loss": 4.9905,
      "mean_token_accuracy": 0.21808684319257737,
      "num_tokens": 132325920.0,
      "step": 76290
    },
    {
      "entropy": 5.970162963867187,
      "epoch": 5.936004668352461,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021192258678171406,
      "loss": 4.9827,
      "mean_token_accuracy": 0.21107023507356643,
      "num_tokens": 132334886.0,
      "step": 76295
    },
    {
      "entropy": 5.979976034164428,
      "epoch": 5.936393697724178,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002118959830793274,
      "loss": 4.9101,
      "mean_token_accuracy": 0.21509451419115067,
      "num_tokens": 132344586.0,
      "step": 76300
    },
    {
      "entropy": 5.969796800613404,
      "epoch": 5.9367827270958955,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021186938033444382,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21757058501243592,
      "num_tokens": 132353716.0,
      "step": 76305
    },
    {
      "entropy": 5.9717775821685795,
      "epoch": 5.937171756467613,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021184277854746682,
      "loss": 4.9589,
      "mean_token_accuracy": 0.21845427751541138,
      "num_tokens": 132363213.0,
      "step": 76310
    },
    {
      "entropy": 5.995095252990723,
      "epoch": 5.937560785839331,
      "grad_norm": 1.703125,
      "learning_rate": 0.00021181617771880013,
      "loss": 4.9609,
      "mean_token_accuracy": 0.223108272254467,
      "num_tokens": 132372222.0,
      "step": 76315
    },
    {
      "entropy": 5.956013822555542,
      "epoch": 5.937949815211049,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021178957784884745,
      "loss": 4.9316,
      "mean_token_accuracy": 0.22291901260614394,
      "num_tokens": 132380886.0,
      "step": 76320
    },
    {
      "entropy": 5.97578797340393,
      "epoch": 5.938338844582766,
      "grad_norm": 1.734375,
      "learning_rate": 0.00021176297893801222,
      "loss": 4.898,
      "mean_token_accuracy": 0.22731840312480928,
      "num_tokens": 132389443.0,
      "step": 76325
    },
    {
      "entropy": 5.95592999458313,
      "epoch": 5.938727873954484,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021173638098669818,
      "loss": 4.8807,
      "mean_token_accuracy": 0.22008594125509262,
      "num_tokens": 132397700.0,
      "step": 76330
    },
    {
      "entropy": 5.95688419342041,
      "epoch": 5.939116903326201,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002117097839953089,
      "loss": 4.9352,
      "mean_token_accuracy": 0.22258417308330536,
      "num_tokens": 132406556.0,
      "step": 76335
    },
    {
      "entropy": 5.996276760101319,
      "epoch": 5.9395059326979185,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021168318796424784,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2200467362999916,
      "num_tokens": 132415062.0,
      "step": 76340
    },
    {
      "entropy": 5.983078813552856,
      "epoch": 5.939894962069636,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00021165659289391865,
      "loss": 4.903,
      "mean_token_accuracy": 0.2285816863179207,
      "num_tokens": 132423732.0,
      "step": 76345
    },
    {
      "entropy": 5.9513458728790285,
      "epoch": 5.940283991441354,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0002116299987847249,
      "loss": 4.9841,
      "mean_token_accuracy": 0.22003257274627686,
      "num_tokens": 132432524.0,
      "step": 76350
    },
    {
      "entropy": 5.9460986137390135,
      "epoch": 5.940673020813072,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021160340563707004,
      "loss": 4.877,
      "mean_token_accuracy": 0.23558858036994934,
      "num_tokens": 132442195.0,
      "step": 76355
    },
    {
      "entropy": 5.904898595809937,
      "epoch": 5.941062050184789,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021157681345135754,
      "loss": 4.8815,
      "mean_token_accuracy": 0.22475899010896683,
      "num_tokens": 132451488.0,
      "step": 76360
    },
    {
      "entropy": 5.887748289108276,
      "epoch": 5.941451079556506,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021155022222799097,
      "loss": 4.8686,
      "mean_token_accuracy": 0.22398218363523484,
      "num_tokens": 132460343.0,
      "step": 76365
    },
    {
      "entropy": 6.036226367950439,
      "epoch": 5.941840108928224,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021152363196737378,
      "loss": 5.0504,
      "mean_token_accuracy": 0.21844153255224227,
      "num_tokens": 132469700.0,
      "step": 76370
    },
    {
      "entropy": 5.946969652175904,
      "epoch": 5.9422291382999415,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002114970426699095,
      "loss": 4.8965,
      "mean_token_accuracy": 0.22064703553915024,
      "num_tokens": 132478181.0,
      "step": 76375
    },
    {
      "entropy": 5.98697190284729,
      "epoch": 5.942618167671659,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021147045433600142,
      "loss": 4.9197,
      "mean_token_accuracy": 0.21854668110609055,
      "num_tokens": 132487664.0,
      "step": 76380
    },
    {
      "entropy": 6.015982341766358,
      "epoch": 5.943007197043377,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021144386696605316,
      "loss": 5.0216,
      "mean_token_accuracy": 0.20751693546772004,
      "num_tokens": 132496657.0,
      "step": 76385
    },
    {
      "entropy": 6.005096054077148,
      "epoch": 5.943396226415095,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021141728056046804,
      "loss": 5.0417,
      "mean_token_accuracy": 0.21228197067975998,
      "num_tokens": 132506225.0,
      "step": 76390
    },
    {
      "entropy": 5.983226633071899,
      "epoch": 5.943785255786812,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021139069511964942,
      "loss": 4.9289,
      "mean_token_accuracy": 0.22022390216588975,
      "num_tokens": 132514922.0,
      "step": 76395
    },
    {
      "entropy": 6.009896516799927,
      "epoch": 5.944174285158529,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002113641106440008,
      "loss": 4.9531,
      "mean_token_accuracy": 0.21839309632778167,
      "num_tokens": 132523761.0,
      "step": 76400
    },
    {
      "entropy": 6.001777029037475,
      "epoch": 5.944563314530247,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002113375271339255,
      "loss": 4.9659,
      "mean_token_accuracy": 0.22089091688394547,
      "num_tokens": 132532483.0,
      "step": 76405
    },
    {
      "entropy": 5.975278234481811,
      "epoch": 5.9449523439019645,
      "grad_norm": 1.5,
      "learning_rate": 0.0002113109445898269,
      "loss": 4.8765,
      "mean_token_accuracy": 0.22130245268344878,
      "num_tokens": 132541177.0,
      "step": 76410
    },
    {
      "entropy": 5.982903671264649,
      "epoch": 5.945341373273682,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021128436301210842,
      "loss": 4.9991,
      "mean_token_accuracy": 0.21699151396751404,
      "num_tokens": 132549806.0,
      "step": 76415
    },
    {
      "entropy": 6.069178104400635,
      "epoch": 5.9457304026454,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00021125778240117327,
      "loss": 4.9811,
      "mean_token_accuracy": 0.22431721538305283,
      "num_tokens": 132557805.0,
      "step": 76420
    },
    {
      "entropy": 5.947932195663452,
      "epoch": 5.946119432017118,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002112312027574248,
      "loss": 4.9197,
      "mean_token_accuracy": 0.22801153510808944,
      "num_tokens": 132566090.0,
      "step": 76425
    },
    {
      "entropy": 5.918825674057007,
      "epoch": 5.946508461388834,
      "grad_norm": 1.5,
      "learning_rate": 0.00021120462408126634,
      "loss": 4.9291,
      "mean_token_accuracy": 0.2184022381901741,
      "num_tokens": 132574416.0,
      "step": 76430
    },
    {
      "entropy": 5.9341014385223385,
      "epoch": 5.946897490760552,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021117804637310123,
      "loss": 4.835,
      "mean_token_accuracy": 0.22244606167078018,
      "num_tokens": 132582094.0,
      "step": 76435
    },
    {
      "entropy": 5.917664909362793,
      "epoch": 5.94728652013227,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002111514696333326,
      "loss": 4.951,
      "mean_token_accuracy": 0.2150795802474022,
      "num_tokens": 132590967.0,
      "step": 76440
    },
    {
      "entropy": 5.980406951904297,
      "epoch": 5.9476755495039875,
      "grad_norm": 1.5,
      "learning_rate": 0.00021112489386236383,
      "loss": 4.9604,
      "mean_token_accuracy": 0.22194251269102097,
      "num_tokens": 132599150.0,
      "step": 76445
    },
    {
      "entropy": 6.034422445297241,
      "epoch": 5.948064578875705,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021109831906059807,
      "loss": 5.0354,
      "mean_token_accuracy": 0.2103809341788292,
      "num_tokens": 132608798.0,
      "step": 76450
    },
    {
      "entropy": 5.96384973526001,
      "epoch": 5.948453608247423,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002110717452284387,
      "loss": 4.9271,
      "mean_token_accuracy": 0.22060575634241103,
      "num_tokens": 132617462.0,
      "step": 76455
    },
    {
      "entropy": 5.975049734115601,
      "epoch": 5.948842637619141,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021104517236628868,
      "loss": 4.8952,
      "mean_token_accuracy": 0.2230422243475914,
      "num_tokens": 132626093.0,
      "step": 76460
    },
    {
      "entropy": 5.980978155136109,
      "epoch": 5.949231666990858,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002110186004745515,
      "loss": 4.9222,
      "mean_token_accuracy": 0.2187782883644104,
      "num_tokens": 132634844.0,
      "step": 76465
    },
    {
      "entropy": 5.94197039604187,
      "epoch": 5.949620696362575,
      "grad_norm": 1.6875,
      "learning_rate": 0.00021099202955363023,
      "loss": 4.8411,
      "mean_token_accuracy": 0.22342236936092377,
      "num_tokens": 132642138.0,
      "step": 76470
    },
    {
      "entropy": 5.988800477981568,
      "epoch": 5.950009725734293,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021096545960392805,
      "loss": 5.0387,
      "mean_token_accuracy": 0.21124572157859803,
      "num_tokens": 132650761.0,
      "step": 76475
    },
    {
      "entropy": 5.983235311508179,
      "epoch": 5.9503987551060105,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00021093889062584803,
      "loss": 4.9367,
      "mean_token_accuracy": 0.2210688352584839,
      "num_tokens": 132660166.0,
      "step": 76480
    },
    {
      "entropy": 6.0448699474334715,
      "epoch": 5.950787784477728,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021091232261979343,
      "loss": 4.9138,
      "mean_token_accuracy": 0.22947245389223098,
      "num_tokens": 132669766.0,
      "step": 76485
    },
    {
      "entropy": 6.060372829437256,
      "epoch": 5.951176813849446,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002108857555861673,
      "loss": 5.0461,
      "mean_token_accuracy": 0.21241808086633682,
      "num_tokens": 132678524.0,
      "step": 76490
    },
    {
      "entropy": 5.976383399963379,
      "epoch": 5.9515658432211636,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021085918952537275,
      "loss": 4.9436,
      "mean_token_accuracy": 0.22428935319185256,
      "num_tokens": 132687152.0,
      "step": 76495
    },
    {
      "entropy": 6.067989826202393,
      "epoch": 5.95195487259288,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00021083262443781297,
      "loss": 5.0917,
      "mean_token_accuracy": 0.2051993802189827,
      "num_tokens": 132695438.0,
      "step": 76500
    },
    {
      "entropy": 5.993937635421753,
      "epoch": 5.952343901964598,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00021080606032389094,
      "loss": 4.9082,
      "mean_token_accuracy": 0.22401574105024338,
      "num_tokens": 132703746.0,
      "step": 76505
    },
    {
      "entropy": 5.991696977615357,
      "epoch": 5.952732931336316,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002107794971840098,
      "loss": 5.0178,
      "mean_token_accuracy": 0.21880230605602263,
      "num_tokens": 132713140.0,
      "step": 76510
    },
    {
      "entropy": 5.986170434951783,
      "epoch": 5.9531219607080335,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021075293501857264,
      "loss": 5.0017,
      "mean_token_accuracy": 0.21491063386201859,
      "num_tokens": 132722291.0,
      "step": 76515
    },
    {
      "entropy": 5.985276126861573,
      "epoch": 5.953510990079751,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021072637382798232,
      "loss": 4.8949,
      "mean_token_accuracy": 0.22182900011539458,
      "num_tokens": 132729794.0,
      "step": 76520
    },
    {
      "entropy": 5.956818771362305,
      "epoch": 5.953900019451469,
      "grad_norm": 1.640625,
      "learning_rate": 0.00021069981361264195,
      "loss": 4.9016,
      "mean_token_accuracy": 0.22393874675035477,
      "num_tokens": 132738052.0,
      "step": 76525
    },
    {
      "entropy": 6.017718839645386,
      "epoch": 5.9542890488231865,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002106732543729546,
      "loss": 4.9528,
      "mean_token_accuracy": 0.23062801510095596,
      "num_tokens": 132746704.0,
      "step": 76530
    },
    {
      "entropy": 5.954107856750488,
      "epoch": 5.954678078194903,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021064669610932325,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2258006066083908,
      "num_tokens": 132755752.0,
      "step": 76535
    },
    {
      "entropy": 5.958869361877442,
      "epoch": 5.955067107566621,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021062013882215086,
      "loss": 4.8758,
      "mean_token_accuracy": 0.23096795827150346,
      "num_tokens": 132764116.0,
      "step": 76540
    },
    {
      "entropy": 5.993494081497192,
      "epoch": 5.955456136938339,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021059358251184046,
      "loss": 5.0023,
      "mean_token_accuracy": 0.21509197354316711,
      "num_tokens": 132772825.0,
      "step": 76545
    },
    {
      "entropy": 5.921010446548462,
      "epoch": 5.9558451663100564,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021056702717879482,
      "loss": 4.9003,
      "mean_token_accuracy": 0.22108598053455353,
      "num_tokens": 132781832.0,
      "step": 76550
    },
    {
      "entropy": 6.059085988998413,
      "epoch": 5.956234195681774,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021054047282341703,
      "loss": 5.0562,
      "mean_token_accuracy": 0.21404702216386795,
      "num_tokens": 132789783.0,
      "step": 76555
    },
    {
      "entropy": 5.956517457962036,
      "epoch": 5.956623225053492,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021051391944610997,
      "loss": 4.8954,
      "mean_token_accuracy": 0.22301306575536728,
      "num_tokens": 132798034.0,
      "step": 76560
    },
    {
      "entropy": 5.983991575241089,
      "epoch": 5.957012254425209,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021048736704727656,
      "loss": 4.9358,
      "mean_token_accuracy": 0.21720358729362488,
      "num_tokens": 132806384.0,
      "step": 76565
    },
    {
      "entropy": 5.990347480773925,
      "epoch": 5.957401283796926,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021046081562731966,
      "loss": 4.9938,
      "mean_token_accuracy": 0.2164604663848877,
      "num_tokens": 132815035.0,
      "step": 76570
    },
    {
      "entropy": 6.012320184707642,
      "epoch": 5.957790313168644,
      "grad_norm": 1.65625,
      "learning_rate": 0.00021043426518664216,
      "loss": 5.0348,
      "mean_token_accuracy": 0.2125747784972191,
      "num_tokens": 132823210.0,
      "step": 76575
    },
    {
      "entropy": 6.007803678512573,
      "epoch": 5.958179342540362,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021040771572564693,
      "loss": 4.8758,
      "mean_token_accuracy": 0.23070555329322814,
      "num_tokens": 132832078.0,
      "step": 76580
    },
    {
      "entropy": 5.963626861572266,
      "epoch": 5.958568371912079,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021038116724473677,
      "loss": 4.9204,
      "mean_token_accuracy": 0.21983421593904495,
      "num_tokens": 132840642.0,
      "step": 76585
    },
    {
      "entropy": 5.93292202949524,
      "epoch": 5.958957401283797,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021035461974431446,
      "loss": 4.924,
      "mean_token_accuracy": 0.22176503390073776,
      "num_tokens": 132850046.0,
      "step": 76590
    },
    {
      "entropy": 5.993535757064819,
      "epoch": 5.959346430655515,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00021032807322478303,
      "loss": 4.9499,
      "mean_token_accuracy": 0.2103102132678032,
      "num_tokens": 132858342.0,
      "step": 76595
    },
    {
      "entropy": 6.027574300765991,
      "epoch": 5.9597354600272325,
      "grad_norm": 1.703125,
      "learning_rate": 0.00021030152768654515,
      "loss": 4.977,
      "mean_token_accuracy": 0.2187955379486084,
      "num_tokens": 132866010.0,
      "step": 76600
    },
    {
      "entropy": 5.932058811187744,
      "epoch": 5.960124489398949,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021027498313000358,
      "loss": 4.891,
      "mean_token_accuracy": 0.22569548189640046,
      "num_tokens": 132874770.0,
      "step": 76605
    },
    {
      "entropy": 6.013638639450074,
      "epoch": 5.960513518770667,
      "grad_norm": 1.5625,
      "learning_rate": 0.00021024843955556112,
      "loss": 5.054,
      "mean_token_accuracy": 0.20791116505861282,
      "num_tokens": 132882582.0,
      "step": 76610
    },
    {
      "entropy": 5.927585220336914,
      "epoch": 5.960902548142385,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021022189696362053,
      "loss": 4.9868,
      "mean_token_accuracy": 0.2119032233953476,
      "num_tokens": 132891848.0,
      "step": 76615
    },
    {
      "entropy": 5.996703195571899,
      "epoch": 5.961291577514102,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002101953553545845,
      "loss": 4.9582,
      "mean_token_accuracy": 0.21794053316116332,
      "num_tokens": 132901508.0,
      "step": 76620
    },
    {
      "entropy": 6.093919849395752,
      "epoch": 5.96168060688582,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002101688147288559,
      "loss": 5.0563,
      "mean_token_accuracy": 0.2100182741880417,
      "num_tokens": 132910220.0,
      "step": 76625
    },
    {
      "entropy": 5.998880195617676,
      "epoch": 5.962069636257538,
      "grad_norm": 1.5,
      "learning_rate": 0.0002101422750868373,
      "loss": 4.9068,
      "mean_token_accuracy": 0.21769822537899017,
      "num_tokens": 132919446.0,
      "step": 76630
    },
    {
      "entropy": 5.993906927108765,
      "epoch": 5.962458665629255,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002101157364289315,
      "loss": 4.919,
      "mean_token_accuracy": 0.22167381197214125,
      "num_tokens": 132927739.0,
      "step": 76635
    },
    {
      "entropy": 5.913827657699585,
      "epoch": 5.962847695000972,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021008919875554105,
      "loss": 4.8946,
      "mean_token_accuracy": 0.2217745900154114,
      "num_tokens": 132935935.0,
      "step": 76640
    },
    {
      "entropy": 5.986498641967773,
      "epoch": 5.96323672437269,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002100626620670687,
      "loss": 5.0035,
      "mean_token_accuracy": 0.21553474217653273,
      "num_tokens": 132943545.0,
      "step": 76645
    },
    {
      "entropy": 6.003711462020874,
      "epoch": 5.963625753744408,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002100361263639171,
      "loss": 5.0122,
      "mean_token_accuracy": 0.21173954159021377,
      "num_tokens": 132952284.0,
      "step": 76650
    },
    {
      "entropy": 5.954073286056518,
      "epoch": 5.964014783116125,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021000959164648892,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22590485513210296,
      "num_tokens": 132961251.0,
      "step": 76655
    },
    {
      "entropy": 6.039183902740478,
      "epoch": 5.964403812487843,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00020998305791518668,
      "loss": 4.9278,
      "mean_token_accuracy": 0.2174799233675003,
      "num_tokens": 132969733.0,
      "step": 76660
    },
    {
      "entropy": 5.995110940933228,
      "epoch": 5.964792841859561,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002099565251704131,
      "loss": 4.9148,
      "mean_token_accuracy": 0.2190564289689064,
      "num_tokens": 132977880.0,
      "step": 76665
    },
    {
      "entropy": 6.000592136383057,
      "epoch": 5.965181871231278,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020992999341257069,
      "loss": 4.9167,
      "mean_token_accuracy": 0.22005137354135512,
      "num_tokens": 132986702.0,
      "step": 76670
    },
    {
      "entropy": 5.993964147567749,
      "epoch": 5.965570900602995,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020990346264206206,
      "loss": 4.998,
      "mean_token_accuracy": 0.216702701151371,
      "num_tokens": 132996506.0,
      "step": 76675
    },
    {
      "entropy": 5.951447296142578,
      "epoch": 5.965959929974713,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020987693285928977,
      "loss": 4.889,
      "mean_token_accuracy": 0.2267771452665329,
      "num_tokens": 133005259.0,
      "step": 76680
    },
    {
      "entropy": 6.004345226287842,
      "epoch": 5.966348959346431,
      "grad_norm": 1.5,
      "learning_rate": 0.00020985040406465644,
      "loss": 4.9723,
      "mean_token_accuracy": 0.21950361132621765,
      "num_tokens": 133013977.0,
      "step": 76685
    },
    {
      "entropy": 5.942533206939697,
      "epoch": 5.966737988718148,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020982387625856447,
      "loss": 4.9008,
      "mean_token_accuracy": 0.2264646530151367,
      "num_tokens": 133022887.0,
      "step": 76690
    },
    {
      "entropy": 5.999416255950928,
      "epoch": 5.967127018089866,
      "grad_norm": 1.625,
      "learning_rate": 0.0002097973494414165,
      "loss": 5.005,
      "mean_token_accuracy": 0.22284844517707825,
      "num_tokens": 133031927.0,
      "step": 76695
    },
    {
      "entropy": 5.968070220947266,
      "epoch": 5.967516047461583,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020977082361361488,
      "loss": 4.9615,
      "mean_token_accuracy": 0.2240196242928505,
      "num_tokens": 133040814.0,
      "step": 76700
    },
    {
      "entropy": 5.986406135559082,
      "epoch": 5.967905076833301,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020974429877556228,
      "loss": 4.9085,
      "mean_token_accuracy": 0.21860576570034027,
      "num_tokens": 133048890.0,
      "step": 76705
    },
    {
      "entropy": 5.956060218811035,
      "epoch": 5.968294106205018,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020971777492766102,
      "loss": 4.9277,
      "mean_token_accuracy": 0.22136975824832916,
      "num_tokens": 133057618.0,
      "step": 76710
    },
    {
      "entropy": 5.967570352554321,
      "epoch": 5.968683135576736,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020969125207031365,
      "loss": 4.9321,
      "mean_token_accuracy": 0.22118736505508424,
      "num_tokens": 133065766.0,
      "step": 76715
    },
    {
      "entropy": 5.907491779327392,
      "epoch": 5.969072164948454,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020966473020392256,
      "loss": 4.8654,
      "mean_token_accuracy": 0.21503092497587203,
      "num_tokens": 133074266.0,
      "step": 76720
    },
    {
      "entropy": 6.038563299179077,
      "epoch": 5.969461194320171,
      "grad_norm": 1.5,
      "learning_rate": 0.0002096382093288901,
      "loss": 4.9939,
      "mean_token_accuracy": 0.21218398958444595,
      "num_tokens": 133082662.0,
      "step": 76725
    },
    {
      "entropy": 5.970776605606079,
      "epoch": 5.969850223691889,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020961168944561888,
      "loss": 4.9487,
      "mean_token_accuracy": 0.21506679207086563,
      "num_tokens": 133091305.0,
      "step": 76730
    },
    {
      "entropy": 5.969977235794067,
      "epoch": 5.970239253063606,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020958517055451125,
      "loss": 4.9759,
      "mean_token_accuracy": 0.21502998918294908,
      "num_tokens": 133100154.0,
      "step": 76735
    },
    {
      "entropy": 6.028014039993286,
      "epoch": 5.970628282435324,
      "grad_norm": 1.703125,
      "learning_rate": 0.00020955865265596946,
      "loss": 5.03,
      "mean_token_accuracy": 0.21870625466108323,
      "num_tokens": 133108637.0,
      "step": 76740
    },
    {
      "entropy": 5.9332670211792,
      "epoch": 5.971017311807041,
      "grad_norm": 1.46875,
      "learning_rate": 0.000209532135750396,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21140479296445847,
      "num_tokens": 133117463.0,
      "step": 76745
    },
    {
      "entropy": 5.966789960861206,
      "epoch": 5.971406341178759,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020950561983819316,
      "loss": 4.9177,
      "mean_token_accuracy": 0.21949040293693542,
      "num_tokens": 133126030.0,
      "step": 76750
    },
    {
      "entropy": 5.962873935699463,
      "epoch": 5.971795370550477,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002094791049197633,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22716207504272462,
      "num_tokens": 133134455.0,
      "step": 76755
    },
    {
      "entropy": 6.019323253631592,
      "epoch": 5.972184399922194,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020945259099550866,
      "loss": 4.9714,
      "mean_token_accuracy": 0.2121277004480362,
      "num_tokens": 133142846.0,
      "step": 76760
    },
    {
      "entropy": 5.95589246749878,
      "epoch": 5.972573429293911,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020942607806583174,
      "loss": 4.8933,
      "mean_token_accuracy": 0.22936320453882217,
      "num_tokens": 133151551.0,
      "step": 76765
    },
    {
      "entropy": 5.934711933135986,
      "epoch": 5.972962458665629,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020939956613113465,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21863277554512023,
      "num_tokens": 133160457.0,
      "step": 76770
    },
    {
      "entropy": 5.876204156875611,
      "epoch": 5.973351488037347,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020937305519181976,
      "loss": 4.803,
      "mean_token_accuracy": 0.23065300285816193,
      "num_tokens": 133168159.0,
      "step": 76775
    },
    {
      "entropy": 5.9455420017242435,
      "epoch": 5.973740517409064,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020934654524828927,
      "loss": 4.964,
      "mean_token_accuracy": 0.22638704478740693,
      "num_tokens": 133176646.0,
      "step": 76780
    },
    {
      "entropy": 6.018430471420288,
      "epoch": 5.974129546780782,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002093200363009455,
      "loss": 4.9874,
      "mean_token_accuracy": 0.21489593237638474,
      "num_tokens": 133184716.0,
      "step": 76785
    },
    {
      "entropy": 6.003715515136719,
      "epoch": 5.9745185761525,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00020929352835019055,
      "loss": 5.048,
      "mean_token_accuracy": 0.21338895112276077,
      "num_tokens": 133192991.0,
      "step": 76790
    },
    {
      "entropy": 5.985891056060791,
      "epoch": 5.974907605524217,
      "grad_norm": 1.5,
      "learning_rate": 0.00020926702139642677,
      "loss": 4.9149,
      "mean_token_accuracy": 0.2205267384648323,
      "num_tokens": 133201785.0,
      "step": 76795
    },
    {
      "entropy": 6.044186973571778,
      "epoch": 5.975296634895935,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020924051544005636,
      "loss": 5.001,
      "mean_token_accuracy": 0.2164405271410942,
      "num_tokens": 133210618.0,
      "step": 76800
    },
    {
      "entropy": 5.98232135772705,
      "epoch": 5.975685664267652,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020921401048148137,
      "loss": 4.9082,
      "mean_token_accuracy": 0.21804079413414001,
      "num_tokens": 133218730.0,
      "step": 76805
    },
    {
      "entropy": 5.987695741653442,
      "epoch": 5.97607469363937,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020918750652110414,
      "loss": 4.9842,
      "mean_token_accuracy": 0.21163113713264464,
      "num_tokens": 133227866.0,
      "step": 76810
    },
    {
      "entropy": 5.998150491714478,
      "epoch": 5.976463723011087,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020916100355932677,
      "loss": 4.9811,
      "mean_token_accuracy": 0.22020472288131715,
      "num_tokens": 133236376.0,
      "step": 76815
    },
    {
      "entropy": 5.958772659301758,
      "epoch": 5.976852752382805,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020913450159655133,
      "loss": 4.8701,
      "mean_token_accuracy": 0.23272328227758407,
      "num_tokens": 133244949.0,
      "step": 76820
    },
    {
      "entropy": 5.998414707183838,
      "epoch": 5.977241781754523,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00020910800063318004,
      "loss": 4.9791,
      "mean_token_accuracy": 0.21842104941606522,
      "num_tokens": 133253082.0,
      "step": 76825
    },
    {
      "entropy": 6.021354913711548,
      "epoch": 5.97763081112624,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002090815006696149,
      "loss": 4.9889,
      "mean_token_accuracy": 0.21992327868938447,
      "num_tokens": 133261716.0,
      "step": 76830
    },
    {
      "entropy": 5.9231987476348875,
      "epoch": 5.978019840497957,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020905500170625808,
      "loss": 4.9026,
      "mean_token_accuracy": 0.2206484094262123,
      "num_tokens": 133271070.0,
      "step": 76835
    },
    {
      "entropy": 6.019783449172974,
      "epoch": 5.978408869869675,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020902850374351168,
      "loss": 4.9563,
      "mean_token_accuracy": 0.21560767889022828,
      "num_tokens": 133280275.0,
      "step": 76840
    },
    {
      "entropy": 5.99537763595581,
      "epoch": 5.978797899241393,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020900200678177766,
      "loss": 4.8692,
      "mean_token_accuracy": 0.22781355381011964,
      "num_tokens": 133288890.0,
      "step": 76845
    },
    {
      "entropy": 6.007684898376465,
      "epoch": 5.97918692861311,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002089755108214582,
      "loss": 5.0261,
      "mean_token_accuracy": 0.22309363335371019,
      "num_tokens": 133298318.0,
      "step": 76850
    },
    {
      "entropy": 5.905185890197754,
      "epoch": 5.979575957984828,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020894901586295529,
      "loss": 4.9037,
      "mean_token_accuracy": 0.23458901792764664,
      "num_tokens": 133306601.0,
      "step": 76855
    },
    {
      "entropy": 6.0069451332092285,
      "epoch": 5.979964987356546,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00020892252190667088,
      "loss": 5.0366,
      "mean_token_accuracy": 0.21249158531427384,
      "num_tokens": 133315367.0,
      "step": 76860
    },
    {
      "entropy": 5.926737070083618,
      "epoch": 5.980354016728263,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000208896028953007,
      "loss": 4.8755,
      "mean_token_accuracy": 0.22236594557762146,
      "num_tokens": 133323672.0,
      "step": 76865
    },
    {
      "entropy": 5.985556554794312,
      "epoch": 5.98074304609998,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020886953700236576,
      "loss": 4.9367,
      "mean_token_accuracy": 0.22075784355401992,
      "num_tokens": 133332111.0,
      "step": 76870
    },
    {
      "entropy": 6.05053596496582,
      "epoch": 5.981132075471698,
      "grad_norm": 1.5,
      "learning_rate": 0.00020884304605514899,
      "loss": 5.0005,
      "mean_token_accuracy": 0.2217547655105591,
      "num_tokens": 133340020.0,
      "step": 76875
    },
    {
      "entropy": 6.051310873031616,
      "epoch": 5.981521104843416,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020881655611175865,
      "loss": 5.036,
      "mean_token_accuracy": 0.21145085841417313,
      "num_tokens": 133349145.0,
      "step": 76880
    },
    {
      "entropy": 5.993061590194702,
      "epoch": 5.981910134215133,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002087900671725968,
      "loss": 5.0472,
      "mean_token_accuracy": 0.2114439055323601,
      "num_tokens": 133358196.0,
      "step": 76885
    },
    {
      "entropy": 5.96725172996521,
      "epoch": 5.982299163586851,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020876357923806527,
      "loss": 4.9781,
      "mean_token_accuracy": 0.21329282373189926,
      "num_tokens": 133367405.0,
      "step": 76890
    },
    {
      "entropy": 6.012993717193604,
      "epoch": 5.982688192958569,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020873709230856608,
      "loss": 4.9876,
      "mean_token_accuracy": 0.21588334441184998,
      "num_tokens": 133376184.0,
      "step": 76895
    },
    {
      "entropy": 6.053491163253784,
      "epoch": 5.9830772223302855,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020871060638450101,
      "loss": 4.9948,
      "mean_token_accuracy": 0.2177640527486801,
      "num_tokens": 133385316.0,
      "step": 76900
    },
    {
      "entropy": 6.007946872711182,
      "epoch": 5.983466251702003,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002086841214662719,
      "loss": 4.895,
      "mean_token_accuracy": 0.22642500102519988,
      "num_tokens": 133393847.0,
      "step": 76905
    },
    {
      "entropy": 5.993705415725708,
      "epoch": 5.983855281073721,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020865763755428075,
      "loss": 4.9658,
      "mean_token_accuracy": 0.22106704860925674,
      "num_tokens": 133403252.0,
      "step": 76910
    },
    {
      "entropy": 5.971817064285278,
      "epoch": 5.984244310445439,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020863115464892935,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21253135800361633,
      "num_tokens": 133413210.0,
      "step": 76915
    },
    {
      "entropy": 5.907345581054687,
      "epoch": 5.984633339817156,
      "grad_norm": 1.625,
      "learning_rate": 0.00020860467275061952,
      "loss": 4.8542,
      "mean_token_accuracy": 0.22638775855302812,
      "num_tokens": 133421105.0,
      "step": 76920
    },
    {
      "entropy": 5.904464483261108,
      "epoch": 5.985022369188874,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020857819185975308,
      "loss": 4.9145,
      "mean_token_accuracy": 0.22159548848867416,
      "num_tokens": 133430268.0,
      "step": 76925
    },
    {
      "entropy": 6.026965570449829,
      "epoch": 5.985411398560592,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020855171197673185,
      "loss": 5.0528,
      "mean_token_accuracy": 0.20412066727876663,
      "num_tokens": 133438815.0,
      "step": 76930
    },
    {
      "entropy": 6.01101131439209,
      "epoch": 5.985800427932309,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002085252331019577,
      "loss": 4.9836,
      "mean_token_accuracy": 0.22079085558652878,
      "num_tokens": 133447178.0,
      "step": 76935
    },
    {
      "entropy": 6.00927848815918,
      "epoch": 5.986189457304026,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020849875523583234,
      "loss": 4.9482,
      "mean_token_accuracy": 0.225357486307621,
      "num_tokens": 133456765.0,
      "step": 76940
    },
    {
      "entropy": 5.938028287887573,
      "epoch": 5.986578486675744,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020847227837875743,
      "loss": 4.897,
      "mean_token_accuracy": 0.22405486255884172,
      "num_tokens": 133464916.0,
      "step": 76945
    },
    {
      "entropy": 6.029953527450561,
      "epoch": 5.986967516047462,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020844580253113486,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21183068305253983,
      "num_tokens": 133472954.0,
      "step": 76950
    },
    {
      "entropy": 5.981648063659668,
      "epoch": 5.987356545419179,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002084193276933663,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21969530284404754,
      "num_tokens": 133481370.0,
      "step": 76955
    },
    {
      "entropy": 6.051704931259155,
      "epoch": 5.987745574790897,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002083928538658535,
      "loss": 5.035,
      "mean_token_accuracy": 0.2084259107708931,
      "num_tokens": 133489839.0,
      "step": 76960
    },
    {
      "entropy": 5.952039623260498,
      "epoch": 5.988134604162614,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020836638104899808,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21670638769865036,
      "num_tokens": 133498669.0,
      "step": 76965
    },
    {
      "entropy": 6.0337859153747555,
      "epoch": 5.9885236335343315,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002083399092432018,
      "loss": 4.986,
      "mean_token_accuracy": 0.2173389732837677,
      "num_tokens": 133508424.0,
      "step": 76970
    },
    {
      "entropy": 6.017998600006104,
      "epoch": 5.988912662906049,
      "grad_norm": 1.6875,
      "learning_rate": 0.00020831343844886625,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21847324818372726,
      "num_tokens": 133517166.0,
      "step": 76975
    },
    {
      "entropy": 6.032848834991455,
      "epoch": 5.989301692277767,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020828696866639324,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21819241642951964,
      "num_tokens": 133526578.0,
      "step": 76980
    },
    {
      "entropy": 5.9617030143737795,
      "epoch": 5.989690721649485,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020826049989618422,
      "loss": 4.9655,
      "mean_token_accuracy": 0.22247709780931474,
      "num_tokens": 133534835.0,
      "step": 76985
    },
    {
      "entropy": 5.9843724250793455,
      "epoch": 5.990079751021202,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020823403213864077,
      "loss": 4.982,
      "mean_token_accuracy": 0.21093514561653137,
      "num_tokens": 133543906.0,
      "step": 76990
    },
    {
      "entropy": 6.011821222305298,
      "epoch": 5.99046878039292,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020820756539416475,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21746853291988372,
      "num_tokens": 133553113.0,
      "step": 76995
    },
    {
      "entropy": 6.017393302917481,
      "epoch": 5.990857809764638,
      "grad_norm": 1.671875,
      "learning_rate": 0.00020818109966315757,
      "loss": 4.9614,
      "mean_token_accuracy": 0.2190835729241371,
      "num_tokens": 133561369.0,
      "step": 77000
    },
    {
      "entropy": 6.002895355224609,
      "epoch": 5.9912468391363545,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020815463494602095,
      "loss": 4.949,
      "mean_token_accuracy": 0.21522486060857773,
      "num_tokens": 133570359.0,
      "step": 77005
    },
    {
      "entropy": 5.86725344657898,
      "epoch": 5.991635868508072,
      "grad_norm": 1.5,
      "learning_rate": 0.00020812817124315626,
      "loss": 4.8139,
      "mean_token_accuracy": 0.2287147432565689,
      "num_tokens": 133579594.0,
      "step": 77010
    },
    {
      "entropy": 5.982420873641968,
      "epoch": 5.99202489787979,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020810170855496514,
      "loss": 4.9808,
      "mean_token_accuracy": 0.2171440303325653,
      "num_tokens": 133587801.0,
      "step": 77015
    },
    {
      "entropy": 5.96649203300476,
      "epoch": 5.992413927251508,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00020807524688184914,
      "loss": 4.9417,
      "mean_token_accuracy": 0.22505539655685425,
      "num_tokens": 133597333.0,
      "step": 77020
    },
    {
      "entropy": 5.9384620666503904,
      "epoch": 5.992802956623225,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002080487862242097,
      "loss": 4.8432,
      "mean_token_accuracy": 0.22624974995851516,
      "num_tokens": 133605425.0,
      "step": 77025
    },
    {
      "entropy": 5.925303554534912,
      "epoch": 5.993191985994943,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002080223265824484,
      "loss": 4.857,
      "mean_token_accuracy": 0.23043868988752364,
      "num_tokens": 133614332.0,
      "step": 77030
    },
    {
      "entropy": 5.9542076587677,
      "epoch": 5.99358101536666,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020799586795696668,
      "loss": 4.8583,
      "mean_token_accuracy": 0.2232855036854744,
      "num_tokens": 133622608.0,
      "step": 77035
    },
    {
      "entropy": 5.913707304000854,
      "epoch": 5.9939700447383775,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020796941034816602,
      "loss": 4.8976,
      "mean_token_accuracy": 0.21951364427804948,
      "num_tokens": 133631117.0,
      "step": 77040
    },
    {
      "entropy": 6.017784690856933,
      "epoch": 5.994359074110095,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002079429537564479,
      "loss": 5.0096,
      "mean_token_accuracy": 0.21839469522237778,
      "num_tokens": 133640483.0,
      "step": 77045
    },
    {
      "entropy": 6.041435861587525,
      "epoch": 5.994748103481813,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020791649818221365,
      "loss": 5.043,
      "mean_token_accuracy": 0.21378972232341767,
      "num_tokens": 133649251.0,
      "step": 77050
    },
    {
      "entropy": 6.000570344924927,
      "epoch": 5.995137132853531,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020789004362586477,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20864510983228685,
      "num_tokens": 133658438.0,
      "step": 77055
    },
    {
      "entropy": 5.964153385162353,
      "epoch": 5.995526162225248,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002078635900878027,
      "loss": 4.9462,
      "mean_token_accuracy": 0.21100929677486419,
      "num_tokens": 133668184.0,
      "step": 77060
    },
    {
      "entropy": 5.916752672195434,
      "epoch": 5.995915191596966,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020783713756842876,
      "loss": 4.8955,
      "mean_token_accuracy": 0.22271228581666946,
      "num_tokens": 133676807.0,
      "step": 77065
    },
    {
      "entropy": 5.902385139465332,
      "epoch": 5.996304220968683,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002078106860681444,
      "loss": 4.8485,
      "mean_token_accuracy": 0.22798672467470169,
      "num_tokens": 133685492.0,
      "step": 77070
    },
    {
      "entropy": 5.940501689910889,
      "epoch": 5.9966932503404005,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020778423558735087,
      "loss": 4.8951,
      "mean_token_accuracy": 0.2226317211985588,
      "num_tokens": 133693683.0,
      "step": 77075
    },
    {
      "entropy": 5.978494930267334,
      "epoch": 5.997082279712118,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002077577861264497,
      "loss": 4.9016,
      "mean_token_accuracy": 0.22277920246124266,
      "num_tokens": 133702323.0,
      "step": 77080
    },
    {
      "entropy": 5.9431970596313475,
      "epoch": 5.997471309083836,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020773133768584208,
      "loss": 5.0221,
      "mean_token_accuracy": 0.20935623496770858,
      "num_tokens": 133711481.0,
      "step": 77085
    },
    {
      "entropy": 5.97619366645813,
      "epoch": 5.9978603384555536,
      "grad_norm": 1.71875,
      "learning_rate": 0.00020770489026592927,
      "loss": 4.9237,
      "mean_token_accuracy": 0.21824227571487426,
      "num_tokens": 133719823.0,
      "step": 77090
    },
    {
      "entropy": 5.99796838760376,
      "epoch": 5.998249367827271,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020767844386711265,
      "loss": 4.9571,
      "mean_token_accuracy": 0.21242578327655792,
      "num_tokens": 133728016.0,
      "step": 77095
    },
    {
      "entropy": 6.016863441467285,
      "epoch": 5.998638397198988,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002076519984897935,
      "loss": 5.0258,
      "mean_token_accuracy": 0.21622181832790374,
      "num_tokens": 133735881.0,
      "step": 77100
    },
    {
      "entropy": 5.923881196975708,
      "epoch": 5.999027426570706,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020762555413437306,
      "loss": 4.9464,
      "mean_token_accuracy": 0.2273606464266777,
      "num_tokens": 133744779.0,
      "step": 77105
    },
    {
      "entropy": 6.005107355117798,
      "epoch": 5.9994164559424235,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002075991108012526,
      "loss": 5.0063,
      "mean_token_accuracy": 0.22038340121507644,
      "num_tokens": 133753199.0,
      "step": 77110
    },
    {
      "entropy": 5.972241163253784,
      "epoch": 5.999805485314141,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002075726684908334,
      "loss": 4.8557,
      "mean_token_accuracy": 0.22678070068359374,
      "num_tokens": 133762417.0,
      "step": 77115
    },
    {
      "entropy": 6.006879647572835,
      "epoch": 6.000155611748687,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020754622720351663,
      "loss": 5.001,
      "mean_token_accuracy": 0.21360349655151367,
      "num_tokens": 133769537.0,
      "step": 77120
    },
    {
      "entropy": 5.963804054260254,
      "epoch": 6.000544641120404,
      "grad_norm": 1.703125,
      "learning_rate": 0.00020751978693970335,
      "loss": 4.8998,
      "mean_token_accuracy": 0.22788792997598648,
      "num_tokens": 133778328.0,
      "step": 77125
    },
    {
      "entropy": 5.984372568130493,
      "epoch": 6.000933670492122,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020749334769979506,
      "loss": 4.9204,
      "mean_token_accuracy": 0.21725817918777465,
      "num_tokens": 133787571.0,
      "step": 77130
    },
    {
      "entropy": 5.921166467666626,
      "epoch": 6.00132269986384,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020746690948419272,
      "loss": 4.8331,
      "mean_token_accuracy": 0.22612046599388122,
      "num_tokens": 133795189.0,
      "step": 77135
    },
    {
      "entropy": 5.918843984603882,
      "epoch": 6.0017117292355575,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020744047229329755,
      "loss": 4.8997,
      "mean_token_accuracy": 0.21915606707334517,
      "num_tokens": 133802840.0,
      "step": 77140
    },
    {
      "entropy": 5.975061655044556,
      "epoch": 6.002100758607275,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002074140361275107,
      "loss": 5.0432,
      "mean_token_accuracy": 0.21217696964740754,
      "num_tokens": 133812648.0,
      "step": 77145
    },
    {
      "entropy": 5.993687582015991,
      "epoch": 6.002489787978992,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020738760098723324,
      "loss": 4.9421,
      "mean_token_accuracy": 0.22535849809646608,
      "num_tokens": 133820769.0,
      "step": 77150
    },
    {
      "entropy": 6.069813632965088,
      "epoch": 6.00287881735071,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00020736116687286634,
      "loss": 4.9749,
      "mean_token_accuracy": 0.2149847850203514,
      "num_tokens": 133828939.0,
      "step": 77155
    },
    {
      "entropy": 5.990317630767822,
      "epoch": 6.003267846722427,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020733473378481112,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2232806235551834,
      "num_tokens": 133837770.0,
      "step": 77160
    },
    {
      "entropy": 5.964057731628418,
      "epoch": 6.003656876094145,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002073083017234685,
      "loss": 4.9178,
      "mean_token_accuracy": 0.22058667242527008,
      "num_tokens": 133846244.0,
      "step": 77165
    },
    {
      "entropy": 5.937447929382325,
      "epoch": 6.004045905465863,
      "grad_norm": 1.703125,
      "learning_rate": 0.00020728187068923975,
      "loss": 4.8657,
      "mean_token_accuracy": 0.22705131769180298,
      "num_tokens": 133854874.0,
      "step": 77170
    },
    {
      "entropy": 5.966950559616089,
      "epoch": 6.0044349348375805,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020725544068252584,
      "loss": 4.9487,
      "mean_token_accuracy": 0.21752160787582397,
      "num_tokens": 133863449.0,
      "step": 77175
    },
    {
      "entropy": 5.87031512260437,
      "epoch": 6.004823964209298,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002072290117037277,
      "loss": 4.7379,
      "mean_token_accuracy": 0.23350879102945327,
      "num_tokens": 133871787.0,
      "step": 77180
    },
    {
      "entropy": 5.9386896133422855,
      "epoch": 6.005212993581015,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020720258375324653,
      "loss": 4.9174,
      "mean_token_accuracy": 0.22336423099040986,
      "num_tokens": 133880368.0,
      "step": 77185
    },
    {
      "entropy": 5.984735107421875,
      "epoch": 6.005602022952733,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002071761568314831,
      "loss": 4.9232,
      "mean_token_accuracy": 0.22758807092905045,
      "num_tokens": 133889687.0,
      "step": 77190
    },
    {
      "entropy": 5.949311447143555,
      "epoch": 6.00599105232445,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002071497309388387,
      "loss": 4.8997,
      "mean_token_accuracy": 0.22296371757984162,
      "num_tokens": 133899450.0,
      "step": 77195
    },
    {
      "entropy": 5.994116115570068,
      "epoch": 6.006380081696168,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020712330607571407,
      "loss": 4.8939,
      "mean_token_accuracy": 0.2270330622792244,
      "num_tokens": 133907777.0,
      "step": 77200
    },
    {
      "entropy": 5.91080117225647,
      "epoch": 6.006769111067886,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020709688224251029,
      "loss": 4.8368,
      "mean_token_accuracy": 0.22681725323200225,
      "num_tokens": 133915852.0,
      "step": 77205
    },
    {
      "entropy": 5.987572193145752,
      "epoch": 6.0071581404396035,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020707045943962817,
      "loss": 4.9849,
      "mean_token_accuracy": 0.2170867443084717,
      "num_tokens": 133925174.0,
      "step": 77210
    },
    {
      "entropy": 5.969158792495728,
      "epoch": 6.007547169811321,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002070440376674687,
      "loss": 4.91,
      "mean_token_accuracy": 0.21896461695432662,
      "num_tokens": 133934902.0,
      "step": 77215
    },
    {
      "entropy": 5.955985593795776,
      "epoch": 6.007936199183038,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020701761692643283,
      "loss": 4.8984,
      "mean_token_accuracy": 0.2231582894921303,
      "num_tokens": 133943608.0,
      "step": 77220
    },
    {
      "entropy": 6.005164623260498,
      "epoch": 6.008325228554756,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020699119721692138,
      "loss": 4.9977,
      "mean_token_accuracy": 0.21900182217359543,
      "num_tokens": 133952556.0,
      "step": 77225
    },
    {
      "entropy": 5.886459922790527,
      "epoch": 6.008714257926473,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020696477853933531,
      "loss": 4.8067,
      "mean_token_accuracy": 0.2316868335008621,
      "num_tokens": 133961156.0,
      "step": 77230
    },
    {
      "entropy": 6.035867309570312,
      "epoch": 6.009103287298191,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002069383608940754,
      "loss": 4.9454,
      "mean_token_accuracy": 0.2193396881222725,
      "num_tokens": 133969857.0,
      "step": 77235
    },
    {
      "entropy": 5.972268724441529,
      "epoch": 6.009492316669909,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020691194428154258,
      "loss": 4.898,
      "mean_token_accuracy": 0.21767839789390564,
      "num_tokens": 133978090.0,
      "step": 77240
    },
    {
      "entropy": 5.937094116210938,
      "epoch": 6.0098813460416265,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00020688552870213762,
      "loss": 4.8327,
      "mean_token_accuracy": 0.22977281957864762,
      "num_tokens": 133986027.0,
      "step": 77245
    },
    {
      "entropy": 5.936698150634766,
      "epoch": 6.010270375413343,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002068591141562614,
      "loss": 4.8228,
      "mean_token_accuracy": 0.23438769727945327,
      "num_tokens": 133994607.0,
      "step": 77250
    },
    {
      "entropy": 5.9959992408752445,
      "epoch": 6.010659404785061,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020683270064431454,
      "loss": 4.9731,
      "mean_token_accuracy": 0.2207173690199852,
      "num_tokens": 134003612.0,
      "step": 77255
    },
    {
      "entropy": 5.990382719039917,
      "epoch": 6.011048434156779,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000206806288166698,
      "loss": 4.9473,
      "mean_token_accuracy": 0.22227606326341628,
      "num_tokens": 134012022.0,
      "step": 77260
    },
    {
      "entropy": 5.984481716156006,
      "epoch": 6.011437463528496,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020677987672381248,
      "loss": 4.8701,
      "mean_token_accuracy": 0.227724127471447,
      "num_tokens": 134020304.0,
      "step": 77265
    },
    {
      "entropy": 6.009970045089721,
      "epoch": 6.011826492900214,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002067534663160588,
      "loss": 4.985,
      "mean_token_accuracy": 0.2256305769085884,
      "num_tokens": 134028949.0,
      "step": 77270
    },
    {
      "entropy": 5.973845386505127,
      "epoch": 6.012215522271932,
      "grad_norm": 1.671875,
      "learning_rate": 0.00020672705694383759,
      "loss": 4.9376,
      "mean_token_accuracy": 0.22260891348123552,
      "num_tokens": 134036996.0,
      "step": 77275
    },
    {
      "entropy": 5.935384893417359,
      "epoch": 6.0126045516436495,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020670064860754973,
      "loss": 4.8811,
      "mean_token_accuracy": 0.22404216080904008,
      "num_tokens": 134045682.0,
      "step": 77280
    },
    {
      "entropy": 6.010344171524048,
      "epoch": 6.012993581015366,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002066742413075957,
      "loss": 4.9386,
      "mean_token_accuracy": 0.22231684029102325,
      "num_tokens": 134054355.0,
      "step": 77285
    },
    {
      "entropy": 6.012735795974732,
      "epoch": 6.013382610387084,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020664783504437635,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21430301517248154,
      "num_tokens": 134062978.0,
      "step": 77290
    },
    {
      "entropy": 6.029959011077881,
      "epoch": 6.013771639758802,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002066214298182923,
      "loss": 4.9736,
      "mean_token_accuracy": 0.21935460418462754,
      "num_tokens": 134072481.0,
      "step": 77295
    },
    {
      "entropy": 6.016332340240479,
      "epoch": 6.014160669130519,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020659502562974424,
      "loss": 4.8694,
      "mean_token_accuracy": 0.2219953417778015,
      "num_tokens": 134080603.0,
      "step": 77300
    },
    {
      "entropy": 6.000202178955078,
      "epoch": 6.014549698502237,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020656862247913277,
      "loss": 4.9309,
      "mean_token_accuracy": 0.2187982589006424,
      "num_tokens": 134089847.0,
      "step": 77305
    },
    {
      "entropy": 5.940379858016968,
      "epoch": 6.014938727873955,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020654222036685854,
      "loss": 4.8917,
      "mean_token_accuracy": 0.22553602010011672,
      "num_tokens": 134098648.0,
      "step": 77310
    },
    {
      "entropy": 5.9866907596588135,
      "epoch": 6.0153277572456725,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020651581929332215,
      "loss": 4.9201,
      "mean_token_accuracy": 0.224651537835598,
      "num_tokens": 134106926.0,
      "step": 77315
    },
    {
      "entropy": 5.974052810668946,
      "epoch": 6.015716786617389,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020648941925892418,
      "loss": 4.8679,
      "mean_token_accuracy": 0.22212464958429337,
      "num_tokens": 134116151.0,
      "step": 77320
    },
    {
      "entropy": 5.978938293457031,
      "epoch": 6.016105815989107,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020646302026406516,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21990459263324738,
      "num_tokens": 134125546.0,
      "step": 77325
    },
    {
      "entropy": 6.007490015029907,
      "epoch": 6.016494845360825,
      "grad_norm": 1.5,
      "learning_rate": 0.00020643662230914584,
      "loss": 4.9041,
      "mean_token_accuracy": 0.22710235118865968,
      "num_tokens": 134133993.0,
      "step": 77330
    },
    {
      "entropy": 5.898662757873535,
      "epoch": 6.016883874732542,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020641022539456661,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2238541454076767,
      "num_tokens": 134142749.0,
      "step": 77335
    },
    {
      "entropy": 5.995209074020385,
      "epoch": 6.01727290410426,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020638382952072805,
      "loss": 4.9829,
      "mean_token_accuracy": 0.21985611021518708,
      "num_tokens": 134151731.0,
      "step": 77340
    },
    {
      "entropy": 6.005035543441773,
      "epoch": 6.017661933475978,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020635743468803064,
      "loss": 4.9655,
      "mean_token_accuracy": 0.21878334283828735,
      "num_tokens": 134161137.0,
      "step": 77345
    },
    {
      "entropy": 5.978911304473877,
      "epoch": 6.018050962847695,
      "grad_norm": 1.734375,
      "learning_rate": 0.0002063310408968749,
      "loss": 4.8601,
      "mean_token_accuracy": 0.231124809384346,
      "num_tokens": 134169883.0,
      "step": 77350
    },
    {
      "entropy": 6.02721266746521,
      "epoch": 6.018439992219412,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002063046481476613,
      "loss": 5.033,
      "mean_token_accuracy": 0.21880795955657958,
      "num_tokens": 134178399.0,
      "step": 77355
    },
    {
      "entropy": 6.0200762271881105,
      "epoch": 6.01882902159113,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0002062782564407904,
      "loss": 4.9516,
      "mean_token_accuracy": 0.2192690849304199,
      "num_tokens": 134186270.0,
      "step": 77360
    },
    {
      "entropy": 5.920985221862793,
      "epoch": 6.019218050962848,
      "grad_norm": 1.65625,
      "learning_rate": 0.00020625186577666256,
      "loss": 4.8096,
      "mean_token_accuracy": 0.2376904994249344,
      "num_tokens": 134194809.0,
      "step": 77365
    },
    {
      "entropy": 6.036172676086426,
      "epoch": 6.019607080334565,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020622547615567816,
      "loss": 5.0527,
      "mean_token_accuracy": 0.20707651525735854,
      "num_tokens": 134203212.0,
      "step": 77370
    },
    {
      "entropy": 6.003637838363647,
      "epoch": 6.019996109706283,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002061990875782378,
      "loss": 4.9639,
      "mean_token_accuracy": 0.22037897557020186,
      "num_tokens": 134211971.0,
      "step": 77375
    },
    {
      "entropy": 5.983209609985352,
      "epoch": 6.020385139078001,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020617270004474165,
      "loss": 4.9117,
      "mean_token_accuracy": 0.22632469832897187,
      "num_tokens": 134220970.0,
      "step": 77380
    },
    {
      "entropy": 5.953566598892212,
      "epoch": 6.020774168449718,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020614631355559033,
      "loss": 4.9063,
      "mean_token_accuracy": 0.21737792044878007,
      "num_tokens": 134230084.0,
      "step": 77385
    },
    {
      "entropy": 5.940289163589478,
      "epoch": 6.021163197821435,
      "grad_norm": 1.53125,
      "learning_rate": 0.000206119928111184,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22485526502132416,
      "num_tokens": 134239051.0,
      "step": 77390
    },
    {
      "entropy": 5.921415758132935,
      "epoch": 6.021552227193153,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00020609354371192324,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22107823342084884,
      "num_tokens": 134247398.0,
      "step": 77395
    },
    {
      "entropy": 5.992382574081421,
      "epoch": 6.021941256564871,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020606716035820823,
      "loss": 4.9737,
      "mean_token_accuracy": 0.2180938631296158,
      "num_tokens": 134256109.0,
      "step": 77400
    },
    {
      "entropy": 5.976199054718018,
      "epoch": 6.022330285936588,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020604077805043932,
      "loss": 4.9254,
      "mean_token_accuracy": 0.22226646393537522,
      "num_tokens": 134264879.0,
      "step": 77405
    },
    {
      "entropy": 6.028080081939697,
      "epoch": 6.022719315308306,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0002060143967890169,
      "loss": 4.9083,
      "mean_token_accuracy": 0.21855670362710952,
      "num_tokens": 134272949.0,
      "step": 77410
    },
    {
      "entropy": 5.942157936096192,
      "epoch": 6.023108344680024,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020598801657434118,
      "loss": 4.913,
      "mean_token_accuracy": 0.2198380634188652,
      "num_tokens": 134281831.0,
      "step": 77415
    },
    {
      "entropy": 5.924440002441406,
      "epoch": 6.023497374051741,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020596163740681245,
      "loss": 4.9343,
      "mean_token_accuracy": 0.22290041744709016,
      "num_tokens": 134291410.0,
      "step": 77420
    },
    {
      "entropy": 5.979711484909058,
      "epoch": 6.023886403423458,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000205935259286831,
      "loss": 4.9175,
      "mean_token_accuracy": 0.22350115478038787,
      "num_tokens": 134299707.0,
      "step": 77425
    },
    {
      "entropy": 6.0375651836395265,
      "epoch": 6.024275432795176,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002059088822147971,
      "loss": 4.9543,
      "mean_token_accuracy": 0.2227134734392166,
      "num_tokens": 134308052.0,
      "step": 77430
    },
    {
      "entropy": 5.970802354812622,
      "epoch": 6.024664462166894,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020588250619111087,
      "loss": 4.925,
      "mean_token_accuracy": 0.21959677934646607,
      "num_tokens": 134316669.0,
      "step": 77435
    },
    {
      "entropy": 5.9212476253509525,
      "epoch": 6.025053491538611,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002058561312161727,
      "loss": 4.7572,
      "mean_token_accuracy": 0.23666637837886811,
      "num_tokens": 134324924.0,
      "step": 77440
    },
    {
      "entropy": 5.981415557861328,
      "epoch": 6.025442520910329,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020582975729038262,
      "loss": 5.0004,
      "mean_token_accuracy": 0.21752534061670303,
      "num_tokens": 134333563.0,
      "step": 77445
    },
    {
      "entropy": 6.002754545211792,
      "epoch": 6.025831550282046,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020580338441414088,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21931889802217483,
      "num_tokens": 134342260.0,
      "step": 77450
    },
    {
      "entropy": 5.953294467926026,
      "epoch": 6.026220579653764,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020577701258784754,
      "loss": 4.9302,
      "mean_token_accuracy": 0.21714317798614502,
      "num_tokens": 134350770.0,
      "step": 77455
    },
    {
      "entropy": 5.984437465667725,
      "epoch": 6.026609609025481,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020575064181190295,
      "loss": 4.8308,
      "mean_token_accuracy": 0.22686077654361725,
      "num_tokens": 134358896.0,
      "step": 77460
    },
    {
      "entropy": 6.010274600982666,
      "epoch": 6.026998638397199,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020572427208670713,
      "loss": 5.0011,
      "mean_token_accuracy": 0.21117804199457169,
      "num_tokens": 134367056.0,
      "step": 77465
    },
    {
      "entropy": 5.977928543090821,
      "epoch": 6.027387667768917,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020569790341266027,
      "loss": 4.9884,
      "mean_token_accuracy": 0.21931353360414504,
      "num_tokens": 134376360.0,
      "step": 77470
    },
    {
      "entropy": 5.967798042297363,
      "epoch": 6.027776697140634,
      "grad_norm": 1.6875,
      "learning_rate": 0.0002056715357901624,
      "loss": 4.8978,
      "mean_token_accuracy": 0.21902000159025192,
      "num_tokens": 134385087.0,
      "step": 77475
    },
    {
      "entropy": 6.044174766540527,
      "epoch": 6.028165726512352,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020564516921961362,
      "loss": 4.9531,
      "mean_token_accuracy": 0.21599837839603425,
      "num_tokens": 134393561.0,
      "step": 77480
    },
    {
      "entropy": 6.01698260307312,
      "epoch": 6.028554755884069,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000205618803701414,
      "loss": 4.9356,
      "mean_token_accuracy": 0.2264395162463188,
      "num_tokens": 134402385.0,
      "step": 77485
    },
    {
      "entropy": 5.97895073890686,
      "epoch": 6.028943785255787,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002055924392359636,
      "loss": 4.8949,
      "mean_token_accuracy": 0.21980111598968505,
      "num_tokens": 134410739.0,
      "step": 77490
    },
    {
      "entropy": 5.956437015533448,
      "epoch": 6.029332814627504,
      "grad_norm": 1.5,
      "learning_rate": 0.0002055660758236625,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2247818276286125,
      "num_tokens": 134419431.0,
      "step": 77495
    },
    {
      "entropy": 5.9876786231994625,
      "epoch": 6.029721843999222,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020553971346491063,
      "loss": 4.9936,
      "mean_token_accuracy": 0.21837008148431777,
      "num_tokens": 134427952.0,
      "step": 77500
    },
    {
      "entropy": 5.90460319519043,
      "epoch": 6.03011087337094,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00020551335216010804,
      "loss": 4.8427,
      "mean_token_accuracy": 0.2301195353269577,
      "num_tokens": 134436713.0,
      "step": 77505
    },
    {
      "entropy": 5.950352668762207,
      "epoch": 6.030499902742657,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002054869919096548,
      "loss": 4.8732,
      "mean_token_accuracy": 0.22478476613759996,
      "num_tokens": 134444920.0,
      "step": 77510
    },
    {
      "entropy": 5.938903522491455,
      "epoch": 6.030888932114375,
      "grad_norm": 1.625,
      "learning_rate": 0.00020546063271395076,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22249113023281097,
      "num_tokens": 134453442.0,
      "step": 77515
    },
    {
      "entropy": 6.082631349563599,
      "epoch": 6.031277961486092,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020543427457339592,
      "loss": 5.0532,
      "mean_token_accuracy": 0.2126928374171257,
      "num_tokens": 134462328.0,
      "step": 77520
    },
    {
      "entropy": 5.911359643936157,
      "epoch": 6.0316669908578096,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020540791748839028,
      "loss": 4.8858,
      "mean_token_accuracy": 0.22006149291992189,
      "num_tokens": 134470777.0,
      "step": 77525
    },
    {
      "entropy": 5.9269492626190186,
      "epoch": 6.032056020229527,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020538156145933373,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22235708236694335,
      "num_tokens": 134479400.0,
      "step": 77530
    },
    {
      "entropy": 6.001809978485108,
      "epoch": 6.032445049601245,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020535520648662614,
      "loss": 4.9648,
      "mean_token_accuracy": 0.22024015635251998,
      "num_tokens": 134488026.0,
      "step": 77535
    },
    {
      "entropy": 6.007876586914063,
      "epoch": 6.032834078972963,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020532885257066748,
      "loss": 4.9715,
      "mean_token_accuracy": 0.22467001974582673,
      "num_tokens": 134496681.0,
      "step": 77540
    },
    {
      "entropy": 5.928050899505616,
      "epoch": 6.03322310834468,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020530249971185754,
      "loss": 4.8801,
      "mean_token_accuracy": 0.2203107476234436,
      "num_tokens": 134505025.0,
      "step": 77545
    },
    {
      "entropy": 5.926378488540649,
      "epoch": 6.033612137716397,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00020527614791059635,
      "loss": 4.835,
      "mean_token_accuracy": 0.22455671429634094,
      "num_tokens": 134513594.0,
      "step": 77550
    },
    {
      "entropy": 6.013893413543701,
      "epoch": 6.034001167088115,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020524979716728353,
      "loss": 4.9701,
      "mean_token_accuracy": 0.22002838999032975,
      "num_tokens": 134522471.0,
      "step": 77555
    },
    {
      "entropy": 5.965164852142334,
      "epoch": 6.0343901964598325,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020522344748231903,
      "loss": 4.9743,
      "mean_token_accuracy": 0.221189084649086,
      "num_tokens": 134530675.0,
      "step": 77560
    },
    {
      "entropy": 5.943581676483154,
      "epoch": 6.03477922583155,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020519709885610267,
      "loss": 4.8491,
      "mean_token_accuracy": 0.2228779211640358,
      "num_tokens": 134539786.0,
      "step": 77565
    },
    {
      "entropy": 5.953074932098389,
      "epoch": 6.035168255203268,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020517075128903423,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2229670539498329,
      "num_tokens": 134547989.0,
      "step": 77570
    },
    {
      "entropy": 6.004550981521606,
      "epoch": 6.035557284574986,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020514440478151352,
      "loss": 4.9666,
      "mean_token_accuracy": 0.2208867222070694,
      "num_tokens": 134556613.0,
      "step": 77575
    },
    {
      "entropy": 6.0341614246368405,
      "epoch": 6.035946313946703,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020511805933394022,
      "loss": 5.0514,
      "mean_token_accuracy": 0.21251089572906495,
      "num_tokens": 134564932.0,
      "step": 77580
    },
    {
      "entropy": 5.991259145736694,
      "epoch": 6.03633534331842,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020509171494671418,
      "loss": 4.9337,
      "mean_token_accuracy": 0.22043437212705613,
      "num_tokens": 134573551.0,
      "step": 77585
    },
    {
      "entropy": 5.973898601531983,
      "epoch": 6.036724372690138,
      "grad_norm": 1.5,
      "learning_rate": 0.00020506537162023503,
      "loss": 4.9304,
      "mean_token_accuracy": 0.21939200460910796,
      "num_tokens": 134581763.0,
      "step": 77590
    },
    {
      "entropy": 5.958333969116211,
      "epoch": 6.0371134020618555,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020503902935490266,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21743723154067993,
      "num_tokens": 134591096.0,
      "step": 77595
    },
    {
      "entropy": 5.992141056060791,
      "epoch": 6.037502431433573,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020501268815111657,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21838610470294953,
      "num_tokens": 134599626.0,
      "step": 77600
    },
    {
      "entropy": 6.053526067733765,
      "epoch": 6.037891460805291,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020498634800927663,
      "loss": 5.009,
      "mean_token_accuracy": 0.21248390823602675,
      "num_tokens": 134607726.0,
      "step": 77605
    },
    {
      "entropy": 5.980777883529663,
      "epoch": 6.038280490177009,
      "grad_norm": 1.671875,
      "learning_rate": 0.00020496000892978233,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22154248803853988,
      "num_tokens": 134615977.0,
      "step": 77610
    },
    {
      "entropy": 5.944348239898682,
      "epoch": 6.038669519548726,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002049336709130335,
      "loss": 4.8949,
      "mean_token_accuracy": 0.22854854464530944,
      "num_tokens": 134624636.0,
      "step": 77615
    },
    {
      "entropy": 5.898325634002686,
      "epoch": 6.039058548920443,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020490733395942969,
      "loss": 4.8386,
      "mean_token_accuracy": 0.23115002661943435,
      "num_tokens": 134633856.0,
      "step": 77620
    },
    {
      "entropy": 5.980371522903442,
      "epoch": 6.039447578292161,
      "grad_norm": 1.765625,
      "learning_rate": 0.00020488099806937043,
      "loss": 4.939,
      "mean_token_accuracy": 0.22500165104866027,
      "num_tokens": 134642195.0,
      "step": 77625
    },
    {
      "entropy": 5.971798610687256,
      "epoch": 6.0398366076638785,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002048546632432555,
      "loss": 4.9132,
      "mean_token_accuracy": 0.22144813984632492,
      "num_tokens": 134650336.0,
      "step": 77630
    },
    {
      "entropy": 5.9069019794464115,
      "epoch": 6.040225637035596,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020482832948148444,
      "loss": 4.8568,
      "mean_token_accuracy": 0.2290749430656433,
      "num_tokens": 134659576.0,
      "step": 77635
    },
    {
      "entropy": 6.002629947662354,
      "epoch": 6.040614666407314,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002048019967844567,
      "loss": 4.996,
      "mean_token_accuracy": 0.22060558944940567,
      "num_tokens": 134667939.0,
      "step": 77640
    },
    {
      "entropy": 5.933436298370362,
      "epoch": 6.041003695779032,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020477566515257197,
      "loss": 4.832,
      "mean_token_accuracy": 0.22532147765159607,
      "num_tokens": 134676024.0,
      "step": 77645
    },
    {
      "entropy": 5.935768079757691,
      "epoch": 6.041392725150749,
      "grad_norm": 1.640625,
      "learning_rate": 0.00020474933458622974,
      "loss": 4.9014,
      "mean_token_accuracy": 0.21894921660423278,
      "num_tokens": 134685210.0,
      "step": 77650
    },
    {
      "entropy": 5.951097679138184,
      "epoch": 6.041781754522466,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020472300508582946,
      "loss": 4.9193,
      "mean_token_accuracy": 0.21972052305936812,
      "num_tokens": 134693951.0,
      "step": 77655
    },
    {
      "entropy": 5.988010740280151,
      "epoch": 6.042170783894184,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020469667665177084,
      "loss": 4.8769,
      "mean_token_accuracy": 0.22732580453157425,
      "num_tokens": 134701999.0,
      "step": 77660
    },
    {
      "entropy": 5.930520820617676,
      "epoch": 6.0425598132659015,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020467034928445323,
      "loss": 4.9292,
      "mean_token_accuracy": 0.21766661256551742,
      "num_tokens": 134710933.0,
      "step": 77665
    },
    {
      "entropy": 5.975750732421875,
      "epoch": 6.042948842637619,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020464402298427608,
      "loss": 4.9808,
      "mean_token_accuracy": 0.21860561668872833,
      "num_tokens": 134719577.0,
      "step": 77670
    },
    {
      "entropy": 5.932960939407349,
      "epoch": 6.043337872009337,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020461769775163897,
      "loss": 4.837,
      "mean_token_accuracy": 0.22973683327436448,
      "num_tokens": 134727472.0,
      "step": 77675
    },
    {
      "entropy": 6.02776255607605,
      "epoch": 6.043726901381055,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020459137358694126,
      "loss": 5.0186,
      "mean_token_accuracy": 0.21540241688489914,
      "num_tokens": 134736351.0,
      "step": 77680
    },
    {
      "entropy": 5.987282371520996,
      "epoch": 6.044115930752771,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020456505049058234,
      "loss": 5.0067,
      "mean_token_accuracy": 0.21220969557762145,
      "num_tokens": 134744953.0,
      "step": 77685
    },
    {
      "entropy": 6.047860765457154,
      "epoch": 6.044504960124489,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020453872846296168,
      "loss": 5.002,
      "mean_token_accuracy": 0.20939269363880159,
      "num_tokens": 134753309.0,
      "step": 77690
    },
    {
      "entropy": 6.013701963424682,
      "epoch": 6.044893989496207,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020451240750447868,
      "loss": 4.9524,
      "mean_token_accuracy": 0.22526639550924302,
      "num_tokens": 134761977.0,
      "step": 77695
    },
    {
      "entropy": 5.902167654037475,
      "epoch": 6.0452830188679245,
      "grad_norm": 1.5,
      "learning_rate": 0.00020448608761553273,
      "loss": 4.8695,
      "mean_token_accuracy": 0.2245625600218773,
      "num_tokens": 134770781.0,
      "step": 77700
    },
    {
      "entropy": 5.9834754943847654,
      "epoch": 6.045672048239642,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020445976879652312,
      "loss": 4.968,
      "mean_token_accuracy": 0.21440071910619735,
      "num_tokens": 134779640.0,
      "step": 77705
    },
    {
      "entropy": 5.979857587814331,
      "epoch": 6.04606107761136,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00020443345104784934,
      "loss": 4.9093,
      "mean_token_accuracy": 0.2272307440638542,
      "num_tokens": 134789130.0,
      "step": 77710
    },
    {
      "entropy": 5.918622732162476,
      "epoch": 6.046450106983078,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020440713436991055,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2257406696677208,
      "num_tokens": 134797214.0,
      "step": 77715
    },
    {
      "entropy": 5.976826763153076,
      "epoch": 6.046839136354794,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000204380818763106,
      "loss": 4.975,
      "mean_token_accuracy": 0.21334893852472306,
      "num_tokens": 134805403.0,
      "step": 77720
    },
    {
      "entropy": 5.947823619842529,
      "epoch": 6.047228165726512,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020435450422783525,
      "loss": 4.9025,
      "mean_token_accuracy": 0.2205536648631096,
      "num_tokens": 134813409.0,
      "step": 77725
    },
    {
      "entropy": 5.916941833496094,
      "epoch": 6.04761719509823,
      "grad_norm": 1.796875,
      "learning_rate": 0.00020432819076449745,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22390880733728408,
      "num_tokens": 134822821.0,
      "step": 77730
    },
    {
      "entropy": 6.01198263168335,
      "epoch": 6.0480062244699475,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020430187837349186,
      "loss": 4.9639,
      "mean_token_accuracy": 0.22288390547037124,
      "num_tokens": 134831286.0,
      "step": 77735
    },
    {
      "entropy": 5.938953399658203,
      "epoch": 6.048395253841665,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020427556705521777,
      "loss": 4.8049,
      "mean_token_accuracy": 0.23457493185997008,
      "num_tokens": 134840273.0,
      "step": 77740
    },
    {
      "entropy": 5.966055488586425,
      "epoch": 6.048784283213383,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002042492568100744,
      "loss": 4.8923,
      "mean_token_accuracy": 0.21983334124088288,
      "num_tokens": 134848254.0,
      "step": 77745
    },
    {
      "entropy": 5.898483848571777,
      "epoch": 6.049173312585101,
      "grad_norm": 1.734375,
      "learning_rate": 0.0002042229476384609,
      "loss": 4.8194,
      "mean_token_accuracy": 0.23587443083524703,
      "num_tokens": 134856045.0,
      "step": 77750
    },
    {
      "entropy": 5.915934324264526,
      "epoch": 6.049562341956817,
      "grad_norm": 1.5,
      "learning_rate": 0.00020419663954077656,
      "loss": 4.8835,
      "mean_token_accuracy": 0.22112328559160233,
      "num_tokens": 134865157.0,
      "step": 77755
    },
    {
      "entropy": 5.9438755989074705,
      "epoch": 6.049951371328535,
      "grad_norm": 1.65625,
      "learning_rate": 0.00020417033251742046,
      "loss": 4.9548,
      "mean_token_accuracy": 0.21761519014835357,
      "num_tokens": 134873859.0,
      "step": 77760
    },
    {
      "entropy": 5.984092473983765,
      "epoch": 6.050340400700253,
      "grad_norm": 1.734375,
      "learning_rate": 0.0002041440265687919,
      "loss": 4.8079,
      "mean_token_accuracy": 0.23234899640083312,
      "num_tokens": 134882386.0,
      "step": 77765
    },
    {
      "entropy": 6.037471437454224,
      "epoch": 6.0507294300719705,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020411772169528987,
      "loss": 4.9796,
      "mean_token_accuracy": 0.21096444129943848,
      "num_tokens": 134890701.0,
      "step": 77770
    },
    {
      "entropy": 5.9838213443756105,
      "epoch": 6.051118459443688,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020409141789731362,
      "loss": 4.9753,
      "mean_token_accuracy": 0.2134937271475792,
      "num_tokens": 134898409.0,
      "step": 77775
    },
    {
      "entropy": 5.95708737373352,
      "epoch": 6.051507488815406,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020406511517526228,
      "loss": 4.9843,
      "mean_token_accuracy": 0.2122500404715538,
      "num_tokens": 134907159.0,
      "step": 77780
    },
    {
      "entropy": 5.959597873687744,
      "epoch": 6.051896518187123,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020403881352953486,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22884802967309953,
      "num_tokens": 134915206.0,
      "step": 77785
    },
    {
      "entropy": 5.939011859893799,
      "epoch": 6.05228554755884,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020401251296053047,
      "loss": 4.9231,
      "mean_token_accuracy": 0.2273995578289032,
      "num_tokens": 134924275.0,
      "step": 77790
    },
    {
      "entropy": 6.003844118118286,
      "epoch": 6.052674576930558,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020398621346864827,
      "loss": 4.9993,
      "mean_token_accuracy": 0.21560055315494536,
      "num_tokens": 134933280.0,
      "step": 77795
    },
    {
      "entropy": 6.023497104644775,
      "epoch": 6.053063606302276,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00020395991505428723,
      "loss": 4.9491,
      "mean_token_accuracy": 0.2228104665875435,
      "num_tokens": 134941810.0,
      "step": 77800
    },
    {
      "entropy": 5.991574764251709,
      "epoch": 6.0534526356739935,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020393361771784647,
      "loss": 4.9746,
      "mean_token_accuracy": 0.21099766343832016,
      "num_tokens": 134950419.0,
      "step": 77805
    },
    {
      "entropy": 6.0198320865631105,
      "epoch": 6.053841665045711,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0002039073214597249,
      "loss": 4.9262,
      "mean_token_accuracy": 0.22320237755775452,
      "num_tokens": 134959150.0,
      "step": 77810
    },
    {
      "entropy": 5.941063594818115,
      "epoch": 6.054230694417429,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020388102628032156,
      "loss": 4.891,
      "mean_token_accuracy": 0.22152338176965714,
      "num_tokens": 134967778.0,
      "step": 77815
    },
    {
      "entropy": 6.0032140731811525,
      "epoch": 6.054619723789146,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020385473218003546,
      "loss": 4.9272,
      "mean_token_accuracy": 0.2187305822968483,
      "num_tokens": 134975969.0,
      "step": 77820
    },
    {
      "entropy": 5.965050077438354,
      "epoch": 6.055008753160863,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020382843915926557,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23009021878242492,
      "num_tokens": 134985370.0,
      "step": 77825
    },
    {
      "entropy": 5.902154064178466,
      "epoch": 6.055397782532581,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020380214721841083,
      "loss": 4.8204,
      "mean_token_accuracy": 0.22571178078651427,
      "num_tokens": 134994374.0,
      "step": 77830
    },
    {
      "entropy": 5.954391860961914,
      "epoch": 6.055786811904299,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00020377585635787015,
      "loss": 4.9311,
      "mean_token_accuracy": 0.21615663468837737,
      "num_tokens": 135001914.0,
      "step": 77835
    },
    {
      "entropy": 5.98260235786438,
      "epoch": 6.0561758412760165,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00020374956657804254,
      "loss": 5.0912,
      "mean_token_accuracy": 0.21001579612493515,
      "num_tokens": 135010561.0,
      "step": 77840
    },
    {
      "entropy": 5.963683271408081,
      "epoch": 6.056564870647734,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002037232778793268,
      "loss": 4.8851,
      "mean_token_accuracy": 0.22478196024894714,
      "num_tokens": 135019308.0,
      "step": 77845
    },
    {
      "entropy": 5.985214424133301,
      "epoch": 6.056953900019452,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002036969902621219,
      "loss": 4.9118,
      "mean_token_accuracy": 0.21541074067354202,
      "num_tokens": 135027818.0,
      "step": 77850
    },
    {
      "entropy": 5.938775968551636,
      "epoch": 6.057342929391169,
      "grad_norm": 1.5,
      "learning_rate": 0.0002036707037268266,
      "loss": 4.9439,
      "mean_token_accuracy": 0.22086464166641234,
      "num_tokens": 135036978.0,
      "step": 77855
    },
    {
      "entropy": 5.941401386260987,
      "epoch": 6.057731958762886,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020364441827383994,
      "loss": 4.9399,
      "mean_token_accuracy": 0.2192774996161461,
      "num_tokens": 135046113.0,
      "step": 77860
    },
    {
      "entropy": 5.9581053256988525,
      "epoch": 6.058120988134604,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020361813390356066,
      "loss": 4.8329,
      "mean_token_accuracy": 0.22635435909032822,
      "num_tokens": 135054072.0,
      "step": 77865
    },
    {
      "entropy": 5.892794418334961,
      "epoch": 6.058510017506322,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020359185061638753,
      "loss": 4.8054,
      "mean_token_accuracy": 0.2313562661409378,
      "num_tokens": 135062606.0,
      "step": 77870
    },
    {
      "entropy": 6.012898588180542,
      "epoch": 6.0588990468780395,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002035655684127194,
      "loss": 4.9966,
      "mean_token_accuracy": 0.2153088390827179,
      "num_tokens": 135071893.0,
      "step": 77875
    },
    {
      "entropy": 5.952249908447266,
      "epoch": 6.059288076249757,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0002035392872929551,
      "loss": 4.8936,
      "mean_token_accuracy": 0.22607151865959169,
      "num_tokens": 135081339.0,
      "step": 77880
    },
    {
      "entropy": 6.018096542358398,
      "epoch": 6.059677105621474,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00020351300725749338,
      "loss": 4.9979,
      "mean_token_accuracy": 0.22137173861265183,
      "num_tokens": 135089862.0,
      "step": 77885
    },
    {
      "entropy": 5.946950674057007,
      "epoch": 6.060066134993192,
      "grad_norm": 1.578125,
      "learning_rate": 0.000203486728306733,
      "loss": 4.9467,
      "mean_token_accuracy": 0.21569077372550965,
      "num_tokens": 135097426.0,
      "step": 77890
    },
    {
      "entropy": 5.929494190216064,
      "epoch": 6.060455164364909,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020346045044107269,
      "loss": 4.8748,
      "mean_token_accuracy": 0.22807603925466538,
      "num_tokens": 135105894.0,
      "step": 77895
    },
    {
      "entropy": 5.996301221847534,
      "epoch": 6.060844193736627,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002034341736609112,
      "loss": 4.8947,
      "mean_token_accuracy": 0.2321320429444313,
      "num_tokens": 135113880.0,
      "step": 77900
    },
    {
      "entropy": 6.082368993759156,
      "epoch": 6.061233223108345,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00020340789796664715,
      "loss": 4.9953,
      "mean_token_accuracy": 0.21067555397748947,
      "num_tokens": 135122622.0,
      "step": 77905
    },
    {
      "entropy": 5.977445077896118,
      "epoch": 6.0616222524800625,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020338162335867938,
      "loss": 4.9126,
      "mean_token_accuracy": 0.22017618119716645,
      "num_tokens": 135130998.0,
      "step": 77910
    },
    {
      "entropy": 6.016769123077393,
      "epoch": 6.06201128185178,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020335534983740645,
      "loss": 4.9437,
      "mean_token_accuracy": 0.22748716473579406,
      "num_tokens": 135140043.0,
      "step": 77915
    },
    {
      "entropy": 5.920091533660889,
      "epoch": 6.062400311223497,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020332907740322693,
      "loss": 4.8987,
      "mean_token_accuracy": 0.22399997860193252,
      "num_tokens": 135147914.0,
      "step": 77920
    },
    {
      "entropy": 5.949782752990723,
      "epoch": 6.062789340595215,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020330280605653967,
      "loss": 4.9341,
      "mean_token_accuracy": 0.22286033630371094,
      "num_tokens": 135156359.0,
      "step": 77925
    },
    {
      "entropy": 5.983214521408081,
      "epoch": 6.063178369966932,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020327653579774331,
      "loss": 4.9527,
      "mean_token_accuracy": 0.22346986085176468,
      "num_tokens": 135165369.0,
      "step": 77930
    },
    {
      "entropy": 5.981209087371826,
      "epoch": 6.06356739933865,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020325026662723622,
      "loss": 4.9988,
      "mean_token_accuracy": 0.21941498517990113,
      "num_tokens": 135174423.0,
      "step": 77935
    },
    {
      "entropy": 5.944496440887451,
      "epoch": 6.063956428710368,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020322399854541724,
      "loss": 4.8982,
      "mean_token_accuracy": 0.2249224454164505,
      "num_tokens": 135183260.0,
      "step": 77940
    },
    {
      "entropy": 5.963327407836914,
      "epoch": 6.0643454580820855,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002031977315526848,
      "loss": 4.9712,
      "mean_token_accuracy": 0.2163778617978096,
      "num_tokens": 135192236.0,
      "step": 77945
    },
    {
      "entropy": 6.014355754852295,
      "epoch": 6.064734487453803,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020317146564943745,
      "loss": 4.9284,
      "mean_token_accuracy": 0.22324275970458984,
      "num_tokens": 135201087.0,
      "step": 77950
    },
    {
      "entropy": 5.984144830703736,
      "epoch": 6.06512351682552,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002031452008360738,
      "loss": 4.9509,
      "mean_token_accuracy": 0.222816164791584,
      "num_tokens": 135210214.0,
      "step": 77955
    },
    {
      "entropy": 5.911145544052124,
      "epoch": 6.065512546197238,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002031189371129924,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23140286654233932,
      "num_tokens": 135219085.0,
      "step": 77960
    },
    {
      "entropy": 5.994936561584472,
      "epoch": 6.065901575568955,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002030926744805917,
      "loss": 4.9707,
      "mean_token_accuracy": 0.2178179457783699,
      "num_tokens": 135227437.0,
      "step": 77965
    },
    {
      "entropy": 5.9807776927948,
      "epoch": 6.066290604940673,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002030664129392702,
      "loss": 5.0124,
      "mean_token_accuracy": 0.21513006687164307,
      "num_tokens": 135236345.0,
      "step": 77970
    },
    {
      "entropy": 5.9765973567962645,
      "epoch": 6.066679634312391,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020304015248942642,
      "loss": 4.9436,
      "mean_token_accuracy": 0.22375525981187822,
      "num_tokens": 135245269.0,
      "step": 77975
    },
    {
      "entropy": 6.03334527015686,
      "epoch": 6.0670686636841085,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020301389313145886,
      "loss": 4.9608,
      "mean_token_accuracy": 0.22217994183301926,
      "num_tokens": 135253809.0,
      "step": 77980
    },
    {
      "entropy": 5.997375202178955,
      "epoch": 6.067457693055825,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020298763486576566,
      "loss": 4.8808,
      "mean_token_accuracy": 0.2261785924434662,
      "num_tokens": 135263583.0,
      "step": 77985
    },
    {
      "entropy": 5.9226892471313475,
      "epoch": 6.067846722427543,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020296137769274563,
      "loss": 4.8795,
      "mean_token_accuracy": 0.2286846086382866,
      "num_tokens": 135272027.0,
      "step": 77990
    },
    {
      "entropy": 5.973358917236328,
      "epoch": 6.068235751799261,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00020293512161279704,
      "loss": 5.0167,
      "mean_token_accuracy": 0.2188491106033325,
      "num_tokens": 135280320.0,
      "step": 77995
    },
    {
      "entropy": 5.903932332992554,
      "epoch": 6.068624781170978,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00020290886662631825,
      "loss": 4.8596,
      "mean_token_accuracy": 0.2267376646399498,
      "num_tokens": 135288757.0,
      "step": 78000
    },
    {
      "epoch": 6.068624781170978,
      "eval_entropy": 5.674746320887869,
      "eval_loss": 5.044197082519531,
      "eval_mean_token_accuracy": 0.22419949895464766,
      "eval_num_tokens": 135288757.0,
      "eval_runtime": 21.6275,
      "eval_samples_per_second": 1921.533,
      "eval_steps_per_second": 240.203,
      "step": 78000
    },
    {
      "entropy": 5.985635566711426,
      "epoch": 6.069013810542696,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020288261273370767,
      "loss": 4.9276,
      "mean_token_accuracy": 0.224013289809227,
      "num_tokens": 135297685.0,
      "step": 78005
    },
    {
      "entropy": 5.890841388702393,
      "epoch": 6.069402839914414,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020285635993536372,
      "loss": 4.7195,
      "mean_token_accuracy": 0.23789566159248351,
      "num_tokens": 135305975.0,
      "step": 78010
    },
    {
      "entropy": 5.9819926738739015,
      "epoch": 6.0697918692861315,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020283010823168464,
      "loss": 4.9691,
      "mean_token_accuracy": 0.22070698887109758,
      "num_tokens": 135315210.0,
      "step": 78015
    },
    {
      "entropy": 5.904526853561402,
      "epoch": 6.070180898657848,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020280385762306874,
      "loss": 4.8561,
      "mean_token_accuracy": 0.22912377417087554,
      "num_tokens": 135323360.0,
      "step": 78020
    },
    {
      "entropy": 5.986085891723633,
      "epoch": 6.070569928029566,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020277760810991446,
      "loss": 4.911,
      "mean_token_accuracy": 0.22665352523326873,
      "num_tokens": 135332326.0,
      "step": 78025
    },
    {
      "entropy": 6.007417964935303,
      "epoch": 6.070958957401284,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020275135969262,
      "loss": 5.0495,
      "mean_token_accuracy": 0.2141617015004158,
      "num_tokens": 135341759.0,
      "step": 78030
    },
    {
      "entropy": 6.044129514694214,
      "epoch": 6.071347986773001,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020272511237158364,
      "loss": 4.9632,
      "mean_token_accuracy": 0.2190430134534836,
      "num_tokens": 135349150.0,
      "step": 78035
    },
    {
      "entropy": 6.009207534790039,
      "epoch": 6.071737016144719,
      "grad_norm": 1.625,
      "learning_rate": 0.00020269886614720367,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22112569212913513,
      "num_tokens": 135357483.0,
      "step": 78040
    },
    {
      "entropy": 5.9999665260314945,
      "epoch": 6.072126045516437,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00020267262101987833,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21539801955223084,
      "num_tokens": 135365937.0,
      "step": 78045
    },
    {
      "entropy": 6.034775066375732,
      "epoch": 6.0725150748881545,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020264637699000587,
      "loss": 5.0689,
      "mean_token_accuracy": 0.21587740182876586,
      "num_tokens": 135374372.0,
      "step": 78050
    },
    {
      "entropy": 5.972180318832398,
      "epoch": 6.072904104259871,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002026201340579844,
      "loss": 4.9243,
      "mean_token_accuracy": 0.21376640498638153,
      "num_tokens": 135382625.0,
      "step": 78055
    },
    {
      "entropy": 6.0044841289520265,
      "epoch": 6.073293133631589,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020259389222421225,
      "loss": 4.9157,
      "mean_token_accuracy": 0.22653304785490036,
      "num_tokens": 135391037.0,
      "step": 78060
    },
    {
      "entropy": 5.9874166488647464,
      "epoch": 6.073682163003307,
      "grad_norm": 1.65625,
      "learning_rate": 0.00020256765148908752,
      "loss": 4.9402,
      "mean_token_accuracy": 0.22228442132472992,
      "num_tokens": 135400278.0,
      "step": 78065
    },
    {
      "entropy": 5.974554395675659,
      "epoch": 6.074071192375024,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002025414118530084,
      "loss": 4.9126,
      "mean_token_accuracy": 0.223488487303257,
      "num_tokens": 135409189.0,
      "step": 78070
    },
    {
      "entropy": 5.922073078155518,
      "epoch": 6.074460221746742,
      "grad_norm": 1.671875,
      "learning_rate": 0.00020251517331637303,
      "loss": 4.8647,
      "mean_token_accuracy": 0.2338866338133812,
      "num_tokens": 135417926.0,
      "step": 78075
    },
    {
      "entropy": 5.969542741775513,
      "epoch": 6.07484925111846,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020248893587957955,
      "loss": 4.9349,
      "mean_token_accuracy": 0.21988017559051515,
      "num_tokens": 135426236.0,
      "step": 78080
    },
    {
      "entropy": 5.952415084838867,
      "epoch": 6.0752382804901774,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020246269954302603,
      "loss": 4.9768,
      "mean_token_accuracy": 0.2148200422525406,
      "num_tokens": 135434816.0,
      "step": 78085
    },
    {
      "entropy": 5.940057468414307,
      "epoch": 6.075627309861894,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020243646430711065,
      "loss": 4.8097,
      "mean_token_accuracy": 0.2401748701930046,
      "num_tokens": 135443194.0,
      "step": 78090
    },
    {
      "entropy": 5.895291090011597,
      "epoch": 6.076016339233612,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020241023017223138,
      "loss": 4.8163,
      "mean_token_accuracy": 0.22246948778629302,
      "num_tokens": 135451568.0,
      "step": 78095
    },
    {
      "entropy": 5.874649286270142,
      "epoch": 6.07640536860533,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020238399713878635,
      "loss": 4.7788,
      "mean_token_accuracy": 0.23226355761289597,
      "num_tokens": 135459633.0,
      "step": 78100
    },
    {
      "entropy": 6.016199684143066,
      "epoch": 6.076794397977047,
      "grad_norm": 1.625,
      "learning_rate": 0.00020235776520717352,
      "loss": 5.0104,
      "mean_token_accuracy": 0.2190709486603737,
      "num_tokens": 135467939.0,
      "step": 78105
    },
    {
      "entropy": 6.004969787597656,
      "epoch": 6.077183427348765,
      "grad_norm": 1.734375,
      "learning_rate": 0.000202331534377791,
      "loss": 4.9775,
      "mean_token_accuracy": 0.21960571110248567,
      "num_tokens": 135476200.0,
      "step": 78110
    },
    {
      "entropy": 5.9907396793365475,
      "epoch": 6.077572456720483,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020230530465103686,
      "loss": 4.8747,
      "mean_token_accuracy": 0.2299434170126915,
      "num_tokens": 135484564.0,
      "step": 78115
    },
    {
      "entropy": 5.974975538253784,
      "epoch": 6.0779614860921996,
      "grad_norm": 1.6875,
      "learning_rate": 0.00020227907602730888,
      "loss": 4.9431,
      "mean_token_accuracy": 0.22075805068016052,
      "num_tokens": 135493359.0,
      "step": 78120
    },
    {
      "entropy": 5.924606084823608,
      "epoch": 6.078350515463917,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020225284850700526,
      "loss": 4.9237,
      "mean_token_accuracy": 0.2304464966058731,
      "num_tokens": 135502440.0,
      "step": 78125
    },
    {
      "entropy": 6.001686382293701,
      "epoch": 6.078739544835635,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020222662209052388,
      "loss": 4.9352,
      "mean_token_accuracy": 0.22169570922851561,
      "num_tokens": 135511185.0,
      "step": 78130
    },
    {
      "entropy": 5.9566662311553955,
      "epoch": 6.079128574207353,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020220039677826263,
      "loss": 4.8743,
      "mean_token_accuracy": 0.21854217648506163,
      "num_tokens": 135519683.0,
      "step": 78135
    },
    {
      "entropy": 6.007566022872925,
      "epoch": 6.07951760357907,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020217417257061953,
      "loss": 4.9451,
      "mean_token_accuracy": 0.22036110758781433,
      "num_tokens": 135527963.0,
      "step": 78140
    },
    {
      "entropy": 5.933487367630005,
      "epoch": 6.079906632950788,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00020214794946799247,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2287461429834366,
      "num_tokens": 135536110.0,
      "step": 78145
    },
    {
      "entropy": 5.980324697494507,
      "epoch": 6.080295662322506,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002021217274707793,
      "loss": 4.904,
      "mean_token_accuracy": 0.21983834356069565,
      "num_tokens": 135544391.0,
      "step": 78150
    },
    {
      "entropy": 5.901234292984009,
      "epoch": 6.0806846916942225,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020209550657937787,
      "loss": 4.8735,
      "mean_token_accuracy": 0.22051397562026978,
      "num_tokens": 135553449.0,
      "step": 78155
    },
    {
      "entropy": 5.968926620483399,
      "epoch": 6.08107372106594,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020206928679418612,
      "loss": 4.9678,
      "mean_token_accuracy": 0.21841342449188234,
      "num_tokens": 135562457.0,
      "step": 78160
    },
    {
      "entropy": 5.924316120147705,
      "epoch": 6.081462750437658,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020204306811560187,
      "loss": 4.8599,
      "mean_token_accuracy": 0.23003023117780685,
      "num_tokens": 135570820.0,
      "step": 78165
    },
    {
      "entropy": 5.961174964904785,
      "epoch": 6.081851779809376,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020201685054402303,
      "loss": 4.8848,
      "mean_token_accuracy": 0.2240026295185089,
      "num_tokens": 135579086.0,
      "step": 78170
    },
    {
      "entropy": 5.973650550842285,
      "epoch": 6.082240809181093,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002019906340798472,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22466128766536714,
      "num_tokens": 135587723.0,
      "step": 78175
    },
    {
      "entropy": 5.9488404273986815,
      "epoch": 6.082629838552811,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020196441872347232,
      "loss": 4.9351,
      "mean_token_accuracy": 0.22238851636648177,
      "num_tokens": 135596396.0,
      "step": 78180
    },
    {
      "entropy": 5.9618528366088865,
      "epoch": 6.083018867924529,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020193820447529605,
      "loss": 4.876,
      "mean_token_accuracy": 0.22049222141504288,
      "num_tokens": 135604193.0,
      "step": 78185
    },
    {
      "entropy": 5.927823114395141,
      "epoch": 6.0834078972962455,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002019119913357163,
      "loss": 4.9647,
      "mean_token_accuracy": 0.21910331696271895,
      "num_tokens": 135612719.0,
      "step": 78190
    },
    {
      "entropy": 5.905794429779053,
      "epoch": 6.083796926667963,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020188577930513074,
      "loss": 4.9063,
      "mean_token_accuracy": 0.2226342722773552,
      "num_tokens": 135621350.0,
      "step": 78195
    },
    {
      "entropy": 6.03681492805481,
      "epoch": 6.084185956039681,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002018595683839372,
      "loss": 5.008,
      "mean_token_accuracy": 0.2172956347465515,
      "num_tokens": 135629799.0,
      "step": 78200
    },
    {
      "entropy": 5.981302213668823,
      "epoch": 6.084574985411399,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020183335857253327,
      "loss": 4.9389,
      "mean_token_accuracy": 0.22085150927305222,
      "num_tokens": 135638679.0,
      "step": 78205
    },
    {
      "entropy": 5.939679384231567,
      "epoch": 6.084964014783116,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020180714987131658,
      "loss": 4.8383,
      "mean_token_accuracy": 0.22711574733257295,
      "num_tokens": 135646638.0,
      "step": 78210
    },
    {
      "entropy": 5.958037376403809,
      "epoch": 6.085353044154834,
      "grad_norm": 1.640625,
      "learning_rate": 0.000201780942280685,
      "loss": 4.94,
      "mean_token_accuracy": 0.2228827506303787,
      "num_tokens": 135654861.0,
      "step": 78215
    },
    {
      "entropy": 5.9381880283355715,
      "epoch": 6.085742073526551,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000201754735801036,
      "loss": 4.8244,
      "mean_token_accuracy": 0.2219199001789093,
      "num_tokens": 135662506.0,
      "step": 78220
    },
    {
      "entropy": 5.963792037963867,
      "epoch": 6.0861311028982685,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020172853043276735,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2203740045428276,
      "num_tokens": 135670970.0,
      "step": 78225
    },
    {
      "entropy": 5.986278676986695,
      "epoch": 6.086520132269986,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002017023261762766,
      "loss": 4.9627,
      "mean_token_accuracy": 0.2270832121372223,
      "num_tokens": 135680058.0,
      "step": 78230
    },
    {
      "entropy": 6.010351514816284,
      "epoch": 6.086909161641704,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002016761230319613,
      "loss": 4.9853,
      "mean_token_accuracy": 0.21775639355182647,
      "num_tokens": 135688835.0,
      "step": 78235
    },
    {
      "entropy": 5.985901117324829,
      "epoch": 6.087298191013422,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020164992100021924,
      "loss": 4.9348,
      "mean_token_accuracy": 0.2217437282204628,
      "num_tokens": 135697019.0,
      "step": 78240
    },
    {
      "entropy": 6.035257148742676,
      "epoch": 6.087687220385139,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020162372008144792,
      "loss": 4.977,
      "mean_token_accuracy": 0.2180163249373436,
      "num_tokens": 135706148.0,
      "step": 78245
    },
    {
      "entropy": 5.969739246368408,
      "epoch": 6.088076249756857,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020159752027604477,
      "loss": 4.9249,
      "mean_token_accuracy": 0.21598249673843384,
      "num_tokens": 135714859.0,
      "step": 78250
    },
    {
      "entropy": 5.97525315284729,
      "epoch": 6.088465279128574,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020157132158440732,
      "loss": 4.9099,
      "mean_token_accuracy": 0.21825291961431503,
      "num_tokens": 135722671.0,
      "step": 78255
    },
    {
      "entropy": 5.8818765640258786,
      "epoch": 6.0888543085002915,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020154512400693338,
      "loss": 4.8211,
      "mean_token_accuracy": 0.22820490300655366,
      "num_tokens": 135731192.0,
      "step": 78260
    },
    {
      "entropy": 5.9135274410247805,
      "epoch": 6.089243337872009,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002015189275440202,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22689738124608994,
      "num_tokens": 135739394.0,
      "step": 78265
    },
    {
      "entropy": 6.028908586502075,
      "epoch": 6.089632367243727,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020149273219606528,
      "loss": 4.9841,
      "mean_token_accuracy": 0.2104635387659073,
      "num_tokens": 135748708.0,
      "step": 78270
    },
    {
      "entropy": 6.017696762084961,
      "epoch": 6.090021396615445,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002014665379634662,
      "loss": 4.9102,
      "mean_token_accuracy": 0.22166254818439485,
      "num_tokens": 135756686.0,
      "step": 78275
    },
    {
      "entropy": 5.969715976715088,
      "epoch": 6.090410425987162,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020144034484662044,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22737436145544052,
      "num_tokens": 135765062.0,
      "step": 78280
    },
    {
      "entropy": 5.960886383056641,
      "epoch": 6.09079945535888,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00020141415284592528,
      "loss": 4.9609,
      "mean_token_accuracy": 0.21668445318937302,
      "num_tokens": 135773651.0,
      "step": 78285
    },
    {
      "entropy": 5.986602544784546,
      "epoch": 6.091188484730597,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020138796196177827,
      "loss": 4.9672,
      "mean_token_accuracy": 0.22843015491962432,
      "num_tokens": 135781910.0,
      "step": 78290
    },
    {
      "entropy": 5.940346384048462,
      "epoch": 6.0915775141023145,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020136177219457675,
      "loss": 4.9133,
      "mean_token_accuracy": 0.21467115730047226,
      "num_tokens": 135790118.0,
      "step": 78295
    },
    {
      "entropy": 5.889401435852051,
      "epoch": 6.091966543474032,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0002013355835447181,
      "loss": 4.8478,
      "mean_token_accuracy": 0.22108627557754518,
      "num_tokens": 135799109.0,
      "step": 78300
    },
    {
      "entropy": 6.015622186660766,
      "epoch": 6.09235557284575,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020130939601259974,
      "loss": 4.9757,
      "mean_token_accuracy": 0.22181079983711244,
      "num_tokens": 135808542.0,
      "step": 78305
    },
    {
      "entropy": 5.994078779220581,
      "epoch": 6.092744602217468,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020128320959861902,
      "loss": 4.9054,
      "mean_token_accuracy": 0.2278132528066635,
      "num_tokens": 135817617.0,
      "step": 78310
    },
    {
      "entropy": 6.033066654205323,
      "epoch": 6.093133631589185,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020125702430317323,
      "loss": 5.0221,
      "mean_token_accuracy": 0.2120734080672264,
      "num_tokens": 135827192.0,
      "step": 78315
    },
    {
      "entropy": 6.0025825023651125,
      "epoch": 6.093522660960902,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002012308401266597,
      "loss": 4.9805,
      "mean_token_accuracy": 0.2124684363603592,
      "num_tokens": 135835581.0,
      "step": 78320
    },
    {
      "entropy": 5.947249126434326,
      "epoch": 6.09391169033262,
      "grad_norm": 1.5,
      "learning_rate": 0.00020120465706947584,
      "loss": 4.8556,
      "mean_token_accuracy": 0.22199095338582991,
      "num_tokens": 135844329.0,
      "step": 78325
    },
    {
      "entropy": 5.981993532180786,
      "epoch": 6.0943007197043375,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020117847513201876,
      "loss": 4.9023,
      "mean_token_accuracy": 0.22581162452697753,
      "num_tokens": 135853141.0,
      "step": 78330
    },
    {
      "entropy": 6.012953615188598,
      "epoch": 6.094689749076055,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002011522943146859,
      "loss": 4.9484,
      "mean_token_accuracy": 0.21931291371583939,
      "num_tokens": 135860911.0,
      "step": 78335
    },
    {
      "entropy": 5.961538791656494,
      "epoch": 6.095078778447773,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002011261146178744,
      "loss": 4.9188,
      "mean_token_accuracy": 0.222324900329113,
      "num_tokens": 135869672.0,
      "step": 78340
    },
    {
      "entropy": 5.911401653289795,
      "epoch": 6.095467807819491,
      "grad_norm": 1.5,
      "learning_rate": 0.00020109993604198152,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22937599271535875,
      "num_tokens": 135878076.0,
      "step": 78345
    },
    {
      "entropy": 5.91434760093689,
      "epoch": 6.095856837191208,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002010737585874045,
      "loss": 4.9257,
      "mean_token_accuracy": 0.21909331530332565,
      "num_tokens": 135886783.0,
      "step": 78350
    },
    {
      "entropy": 5.983807849884033,
      "epoch": 6.096245866562925,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002010475822545405,
      "loss": 4.9721,
      "mean_token_accuracy": 0.21677321046590806,
      "num_tokens": 135895121.0,
      "step": 78355
    },
    {
      "entropy": 5.939267015457153,
      "epoch": 6.096634895934643,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002010214070437868,
      "loss": 4.9304,
      "mean_token_accuracy": 0.22152261435985565,
      "num_tokens": 135903981.0,
      "step": 78360
    },
    {
      "entropy": 5.964714050292969,
      "epoch": 6.0970239253063605,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020099523295554045,
      "loss": 4.8854,
      "mean_token_accuracy": 0.22363710403442383,
      "num_tokens": 135911847.0,
      "step": 78365
    },
    {
      "entropy": 5.938686943054199,
      "epoch": 6.097412954678078,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020096905999019871,
      "loss": 4.8925,
      "mean_token_accuracy": 0.22739910334348679,
      "num_tokens": 135920926.0,
      "step": 78370
    },
    {
      "entropy": 5.927397012710571,
      "epoch": 6.097801984049796,
      "grad_norm": 1.625,
      "learning_rate": 0.00020094288814815858,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2275019630789757,
      "num_tokens": 135928902.0,
      "step": 78375
    },
    {
      "entropy": 5.96088056564331,
      "epoch": 6.098191013421514,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020091671742981725,
      "loss": 4.9537,
      "mean_token_accuracy": 0.22085745483636857,
      "num_tokens": 135938201.0,
      "step": 78380
    },
    {
      "entropy": 6.004913187026977,
      "epoch": 6.098580042793231,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00020089054783557176,
      "loss": 4.8685,
      "mean_token_accuracy": 0.22405725866556167,
      "num_tokens": 135946584.0,
      "step": 78385
    },
    {
      "entropy": 5.958474254608154,
      "epoch": 6.098969072164948,
      "grad_norm": 1.640625,
      "learning_rate": 0.00020086437936581936,
      "loss": 4.8839,
      "mean_token_accuracy": 0.21675440073013305,
      "num_tokens": 135954758.0,
      "step": 78390
    },
    {
      "entropy": 5.905165719985962,
      "epoch": 6.099358101536666,
      "grad_norm": 1.640625,
      "learning_rate": 0.00020083821202095697,
      "loss": 4.8748,
      "mean_token_accuracy": 0.22586527168750764,
      "num_tokens": 135964088.0,
      "step": 78395
    },
    {
      "entropy": 5.974940586090088,
      "epoch": 6.0997471309083835,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020081204580138162,
      "loss": 4.9715,
      "mean_token_accuracy": 0.21640000492334366,
      "num_tokens": 135972669.0,
      "step": 78400
    },
    {
      "entropy": 5.946333932876587,
      "epoch": 6.100136160280101,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020078588070749048,
      "loss": 4.851,
      "mean_token_accuracy": 0.22575690299272538,
      "num_tokens": 135981287.0,
      "step": 78405
    },
    {
      "entropy": 5.9553179264068605,
      "epoch": 6.100525189651819,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020075971673968042,
      "loss": 4.8407,
      "mean_token_accuracy": 0.220459420979023,
      "num_tokens": 135989412.0,
      "step": 78410
    },
    {
      "entropy": 5.992336893081665,
      "epoch": 6.100914219023537,
      "grad_norm": 1.671875,
      "learning_rate": 0.00020073355389834852,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2298018679022789,
      "num_tokens": 135998070.0,
      "step": 78415
    },
    {
      "entropy": 5.970849704742432,
      "epoch": 6.101303248395254,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020070739218389173,
      "loss": 4.9141,
      "mean_token_accuracy": 0.22805558145046234,
      "num_tokens": 136007219.0,
      "step": 78420
    },
    {
      "entropy": 5.991208457946778,
      "epoch": 6.101692277766971,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00020068123159670698,
      "loss": 4.9129,
      "mean_token_accuracy": 0.23118423372507096,
      "num_tokens": 136015658.0,
      "step": 78425
    },
    {
      "entropy": 5.9620085716247555,
      "epoch": 6.102081307138689,
      "grad_norm": 1.5,
      "learning_rate": 0.00020065507213719132,
      "loss": 4.9987,
      "mean_token_accuracy": 0.21187917441129683,
      "num_tokens": 136024669.0,
      "step": 78430
    },
    {
      "entropy": 5.9589409828186035,
      "epoch": 6.1024703365104065,
      "grad_norm": 1.5,
      "learning_rate": 0.00020062891380574156,
      "loss": 4.936,
      "mean_token_accuracy": 0.2157642126083374,
      "num_tokens": 136033183.0,
      "step": 78435
    },
    {
      "entropy": 5.93713903427124,
      "epoch": 6.102859365882124,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020060275660275468,
      "loss": 4.8003,
      "mean_token_accuracy": 0.23457692712545394,
      "num_tokens": 136042125.0,
      "step": 78440
    },
    {
      "entropy": 5.933380556106568,
      "epoch": 6.103248395253842,
      "grad_norm": 1.625,
      "learning_rate": 0.00020057660052862753,
      "loss": 4.8727,
      "mean_token_accuracy": 0.22898467779159545,
      "num_tokens": 136050415.0,
      "step": 78445
    },
    {
      "entropy": 5.920478677749633,
      "epoch": 6.10363742462556,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020055044558375695,
      "loss": 4.9291,
      "mean_token_accuracy": 0.21685829013586044,
      "num_tokens": 136059208.0,
      "step": 78450
    },
    {
      "entropy": 5.970703363418579,
      "epoch": 6.104026453997276,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020052429176853993,
      "loss": 4.9713,
      "mean_token_accuracy": 0.218191060423851,
      "num_tokens": 136067814.0,
      "step": 78455
    },
    {
      "entropy": 5.956562328338623,
      "epoch": 6.104415483368994,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00020049813908337323,
      "loss": 4.9678,
      "mean_token_accuracy": 0.22928624451160431,
      "num_tokens": 136076450.0,
      "step": 78460
    },
    {
      "entropy": 5.894328689575195,
      "epoch": 6.104804512740712,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002004719875286537,
      "loss": 4.8219,
      "mean_token_accuracy": 0.22623507380485536,
      "num_tokens": 136085393.0,
      "step": 78465
    },
    {
      "entropy": 5.924663782119751,
      "epoch": 6.1051935421124295,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002004458371047782,
      "loss": 4.8827,
      "mean_token_accuracy": 0.2290019229054451,
      "num_tokens": 136093942.0,
      "step": 78470
    },
    {
      "entropy": 5.972107076644898,
      "epoch": 6.105582571484147,
      "grad_norm": 1.65625,
      "learning_rate": 0.00020041968781214336,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22824961096048355,
      "num_tokens": 136102357.0,
      "step": 78475
    },
    {
      "entropy": 5.956307077407837,
      "epoch": 6.105971600855865,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002003935396511461,
      "loss": 4.9295,
      "mean_token_accuracy": 0.2217320129275322,
      "num_tokens": 136111966.0,
      "step": 78480
    },
    {
      "entropy": 5.954202556610108,
      "epoch": 6.106360630227583,
      "grad_norm": 1.640625,
      "learning_rate": 0.00020036739262218313,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21754074841737747,
      "num_tokens": 136120461.0,
      "step": 78485
    },
    {
      "entropy": 5.9700439453125,
      "epoch": 6.106749659599299,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020034124672565112,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21563442498445512,
      "num_tokens": 136128414.0,
      "step": 78490
    },
    {
      "entropy": 5.998425388336182,
      "epoch": 6.107138688971017,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002003151019619469,
      "loss": 4.9309,
      "mean_token_accuracy": 0.2237512856721878,
      "num_tokens": 136137363.0,
      "step": 78495
    },
    {
      "entropy": 6.058087778091431,
      "epoch": 6.107527718342735,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020028895833146715,
      "loss": 5.0696,
      "mean_token_accuracy": 0.20810706913471222,
      "num_tokens": 136146494.0,
      "step": 78500
    },
    {
      "entropy": 6.05325927734375,
      "epoch": 6.1079167477144525,
      "grad_norm": 1.5,
      "learning_rate": 0.00020026281583460854,
      "loss": 5.0224,
      "mean_token_accuracy": 0.2140989735722542,
      "num_tokens": 136155011.0,
      "step": 78505
    },
    {
      "entropy": 6.022303199768066,
      "epoch": 6.10830577708617,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00020023667447176776,
      "loss": 4.9573,
      "mean_token_accuracy": 0.21531793028116225,
      "num_tokens": 136163206.0,
      "step": 78510
    },
    {
      "entropy": 5.968208932876587,
      "epoch": 6.108694806457888,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00020021053424334142,
      "loss": 4.9764,
      "mean_token_accuracy": 0.22114571034908295,
      "num_tokens": 136171863.0,
      "step": 78515
    },
    {
      "entropy": 5.96893720626831,
      "epoch": 6.109083835829606,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020018439514972615,
      "loss": 4.8881,
      "mean_token_accuracy": 0.22343537360429763,
      "num_tokens": 136180685.0,
      "step": 78520
    },
    {
      "entropy": 5.969165086746216,
      "epoch": 6.109472865201322,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002001582571913186,
      "loss": 4.8855,
      "mean_token_accuracy": 0.2221762016415596,
      "num_tokens": 136189410.0,
      "step": 78525
    },
    {
      "entropy": 5.932357597351074,
      "epoch": 6.10986189457304,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020013212036851542,
      "loss": 4.8925,
      "mean_token_accuracy": 0.21919392943382263,
      "num_tokens": 136198645.0,
      "step": 78530
    },
    {
      "entropy": 5.836781454086304,
      "epoch": 6.110250923944758,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020010598468171316,
      "loss": 4.6828,
      "mean_token_accuracy": 0.24857892990112304,
      "num_tokens": 136207691.0,
      "step": 78535
    },
    {
      "entropy": 5.985394096374511,
      "epoch": 6.1106399533164755,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002000798501313083,
      "loss": 4.9704,
      "mean_token_accuracy": 0.21573147773742676,
      "num_tokens": 136216838.0,
      "step": 78540
    },
    {
      "entropy": 6.031375074386597,
      "epoch": 6.111028982688193,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020005371671769756,
      "loss": 4.9895,
      "mean_token_accuracy": 0.21868302077054977,
      "num_tokens": 136225421.0,
      "step": 78545
    },
    {
      "entropy": 5.961360836029053,
      "epoch": 6.111418012059911,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020002758444127738,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22049076557159425,
      "num_tokens": 136233919.0,
      "step": 78550
    },
    {
      "entropy": 5.922882318496704,
      "epoch": 6.111807041431628,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020000145330244424,
      "loss": 4.7975,
      "mean_token_accuracy": 0.22663377970457077,
      "num_tokens": 136241662.0,
      "step": 78555
    },
    {
      "entropy": 5.9643824100494385,
      "epoch": 6.112196070803345,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001999753233015947,
      "loss": 5.0022,
      "mean_token_accuracy": 0.21466088742017747,
      "num_tokens": 136251153.0,
      "step": 78560
    },
    {
      "entropy": 5.958481740951538,
      "epoch": 6.112585100175063,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00019994919443912514,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23033823370933532,
      "num_tokens": 136259392.0,
      "step": 78565
    },
    {
      "entropy": 5.9595917701721195,
      "epoch": 6.112974129546781,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00019992306671543213,
      "loss": 4.9442,
      "mean_token_accuracy": 0.22140335887670518,
      "num_tokens": 136268438.0,
      "step": 78570
    },
    {
      "entropy": 5.961529731750488,
      "epoch": 6.1133631589184985,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019989694013091214,
      "loss": 4.968,
      "mean_token_accuracy": 0.2194148138165474,
      "num_tokens": 136277501.0,
      "step": 78575
    },
    {
      "entropy": 5.889406394958496,
      "epoch": 6.113752188290216,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019987081468596146,
      "loss": 4.8353,
      "mean_token_accuracy": 0.23202701508998871,
      "num_tokens": 136285699.0,
      "step": 78580
    },
    {
      "entropy": 6.0094305038452145,
      "epoch": 6.114141217661934,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00019984469038097658,
      "loss": 4.9749,
      "mean_token_accuracy": 0.222022707760334,
      "num_tokens": 136294560.0,
      "step": 78585
    },
    {
      "entropy": 6.0004417419433596,
      "epoch": 6.114530247033651,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000199818567216354,
      "loss": 4.9225,
      "mean_token_accuracy": 0.22777402102947236,
      "num_tokens": 136303523.0,
      "step": 78590
    },
    {
      "entropy": 6.034603834152222,
      "epoch": 6.114919276405368,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019979244519248996,
      "loss": 5.0494,
      "mean_token_accuracy": 0.21380944401025773,
      "num_tokens": 136312897.0,
      "step": 78595
    },
    {
      "entropy": 5.898952007293701,
      "epoch": 6.115308305777086,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001997663243097808,
      "loss": 4.7808,
      "mean_token_accuracy": 0.2306784823536873,
      "num_tokens": 136321550.0,
      "step": 78600
    },
    {
      "entropy": 5.992227554321289,
      "epoch": 6.115697335148804,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019974020456862295,
      "loss": 4.9815,
      "mean_token_accuracy": 0.2275203660130501,
      "num_tokens": 136330477.0,
      "step": 78605
    },
    {
      "entropy": 5.97981333732605,
      "epoch": 6.1160863645205215,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019971408596941283,
      "loss": 4.9931,
      "mean_token_accuracy": 0.21446480304002763,
      "num_tokens": 136338956.0,
      "step": 78610
    },
    {
      "entropy": 5.984450435638427,
      "epoch": 6.116475393892239,
      "grad_norm": 1.5,
      "learning_rate": 0.0001996879685125465,
      "loss": 4.8987,
      "mean_token_accuracy": 0.22448585778474808,
      "num_tokens": 136347124.0,
      "step": 78615
    },
    {
      "entropy": 5.93726224899292,
      "epoch": 6.116864423263957,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019966185219842038,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2140411004424095,
      "num_tokens": 136355456.0,
      "step": 78620
    },
    {
      "entropy": 5.9282817363739015,
      "epoch": 6.117253452635674,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019963573702743077,
      "loss": 4.9551,
      "mean_token_accuracy": 0.2192707657814026,
      "num_tokens": 136364264.0,
      "step": 78625
    },
    {
      "entropy": 5.967901134490967,
      "epoch": 6.117642482007391,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019960962299997394,
      "loss": 4.8721,
      "mean_token_accuracy": 0.2309507042169571,
      "num_tokens": 136373105.0,
      "step": 78630
    },
    {
      "entropy": 5.91287055015564,
      "epoch": 6.118031511379109,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019958351011644605,
      "loss": 4.8756,
      "mean_token_accuracy": 0.21847671866416932,
      "num_tokens": 136381968.0,
      "step": 78635
    },
    {
      "entropy": 5.934564828872681,
      "epoch": 6.118420540750827,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001995573983772433,
      "loss": 4.828,
      "mean_token_accuracy": 0.2331065520644188,
      "num_tokens": 136389803.0,
      "step": 78640
    },
    {
      "entropy": 5.997598552703858,
      "epoch": 6.1188095701225445,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00019953128778276197,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22141017019748688,
      "num_tokens": 136397704.0,
      "step": 78645
    },
    {
      "entropy": 6.0067493438720705,
      "epoch": 6.119198599494262,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001995051783333982,
      "loss": 4.9909,
      "mean_token_accuracy": 0.21464648246765136,
      "num_tokens": 136406042.0,
      "step": 78650
    },
    {
      "entropy": 6.001147890090943,
      "epoch": 6.119587628865979,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019947907002954818,
      "loss": 4.9574,
      "mean_token_accuracy": 0.2186083197593689,
      "num_tokens": 136414128.0,
      "step": 78655
    },
    {
      "entropy": 5.971075582504272,
      "epoch": 6.119976658237697,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019945296287160818,
      "loss": 4.9194,
      "mean_token_accuracy": 0.22189908474683762,
      "num_tokens": 136422710.0,
      "step": 78660
    },
    {
      "entropy": 5.9549116611480715,
      "epoch": 6.120365687609414,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001994268568599742,
      "loss": 5.0005,
      "mean_token_accuracy": 0.20999650955200194,
      "num_tokens": 136430743.0,
      "step": 78665
    },
    {
      "entropy": 5.952028751373291,
      "epoch": 6.120754716981132,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019940075199504232,
      "loss": 4.8831,
      "mean_token_accuracy": 0.2252148911356926,
      "num_tokens": 136440033.0,
      "step": 78670
    },
    {
      "entropy": 5.974386882781983,
      "epoch": 6.12114374635285,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019937464827720874,
      "loss": 4.8616,
      "mean_token_accuracy": 0.23385551422834397,
      "num_tokens": 136448077.0,
      "step": 78675
    },
    {
      "entropy": 5.999250507354736,
      "epoch": 6.121532775724567,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019934854570686944,
      "loss": 4.984,
      "mean_token_accuracy": 0.2167230427265167,
      "num_tokens": 136456930.0,
      "step": 78680
    },
    {
      "entropy": 5.897667360305786,
      "epoch": 6.121921805096285,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019932244428442065,
      "loss": 4.8525,
      "mean_token_accuracy": 0.227958944439888,
      "num_tokens": 136464990.0,
      "step": 78685
    },
    {
      "entropy": 5.878890180587769,
      "epoch": 6.122310834468002,
      "grad_norm": 1.71875,
      "learning_rate": 0.00019929634401025825,
      "loss": 4.8516,
      "mean_token_accuracy": 0.2264557361602783,
      "num_tokens": 136473459.0,
      "step": 78690
    },
    {
      "entropy": 5.987198209762573,
      "epoch": 6.12269986383972,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001992702448847783,
      "loss": 5.0076,
      "mean_token_accuracy": 0.21289132237434388,
      "num_tokens": 136481643.0,
      "step": 78695
    },
    {
      "entropy": 5.907851791381836,
      "epoch": 6.123088893211437,
      "grad_norm": 1.5,
      "learning_rate": 0.00019924414690837688,
      "loss": 4.8332,
      "mean_token_accuracy": 0.22933602929115296,
      "num_tokens": 136490394.0,
      "step": 78700
    },
    {
      "entropy": 5.954915475845337,
      "epoch": 6.123477922583155,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019921805008145,
      "loss": 4.9589,
      "mean_token_accuracy": 0.21602309942245485,
      "num_tokens": 136499383.0,
      "step": 78705
    },
    {
      "entropy": 5.96660590171814,
      "epoch": 6.123866951954873,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019919195440439358,
      "loss": 4.9154,
      "mean_token_accuracy": 0.22569091022014617,
      "num_tokens": 136508294.0,
      "step": 78710
    },
    {
      "entropy": 5.9572892665863035,
      "epoch": 6.12425598132659,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019916585987760356,
      "loss": 4.8792,
      "mean_token_accuracy": 0.2245973527431488,
      "num_tokens": 136516819.0,
      "step": 78715
    },
    {
      "entropy": 5.965902376174927,
      "epoch": 6.124645010698308,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00019913976650147585,
      "loss": 4.885,
      "mean_token_accuracy": 0.2214389756321907,
      "num_tokens": 136525512.0,
      "step": 78720
    },
    {
      "entropy": 5.967025995254517,
      "epoch": 6.125034040070025,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019911367427640647,
      "loss": 4.9129,
      "mean_token_accuracy": 0.2209485426545143,
      "num_tokens": 136534673.0,
      "step": 78725
    },
    {
      "entropy": 5.958683633804322,
      "epoch": 6.125423069441743,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019908758320279134,
      "loss": 4.9064,
      "mean_token_accuracy": 0.21643196195363998,
      "num_tokens": 136544079.0,
      "step": 78730
    },
    {
      "entropy": 5.974506807327271,
      "epoch": 6.12581209881346,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019906149328102635,
      "loss": 4.9069,
      "mean_token_accuracy": 0.2247644543647766,
      "num_tokens": 136552712.0,
      "step": 78735
    },
    {
      "entropy": 6.013232612609864,
      "epoch": 6.126201128185178,
      "grad_norm": 1.625,
      "learning_rate": 0.00019903540451150732,
      "loss": 4.9928,
      "mean_token_accuracy": 0.2182633548974991,
      "num_tokens": 136561444.0,
      "step": 78740
    },
    {
      "entropy": 5.910155534744263,
      "epoch": 6.126590157556896,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019900931689463008,
      "loss": 4.8679,
      "mean_token_accuracy": 0.2251318320631981,
      "num_tokens": 136570619.0,
      "step": 78745
    },
    {
      "entropy": 5.927820873260498,
      "epoch": 6.126979186928613,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019898323043079054,
      "loss": 4.8526,
      "mean_token_accuracy": 0.22456406205892562,
      "num_tokens": 136579230.0,
      "step": 78750
    },
    {
      "entropy": 5.99037504196167,
      "epoch": 6.127368216300331,
      "grad_norm": 1.671875,
      "learning_rate": 0.00019895714512038436,
      "loss": 4.9928,
      "mean_token_accuracy": 0.21802987456321715,
      "num_tokens": 136587614.0,
      "step": 78755
    },
    {
      "entropy": 5.993751335144043,
      "epoch": 6.127757245672048,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019893106096380763,
      "loss": 4.8875,
      "mean_token_accuracy": 0.22469125390052797,
      "num_tokens": 136595509.0,
      "step": 78760
    },
    {
      "entropy": 5.979387092590332,
      "epoch": 6.128146275043766,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019890497796145584,
      "loss": 4.9283,
      "mean_token_accuracy": 0.2224093794822693,
      "num_tokens": 136604160.0,
      "step": 78765
    },
    {
      "entropy": 5.965836000442505,
      "epoch": 6.128535304415483,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019887889611372502,
      "loss": 4.9494,
      "mean_token_accuracy": 0.22006654888391494,
      "num_tokens": 136613260.0,
      "step": 78770
    },
    {
      "entropy": 5.949742031097412,
      "epoch": 6.128924333787201,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001988528154210107,
      "loss": 4.8694,
      "mean_token_accuracy": 0.23093582391738893,
      "num_tokens": 136621189.0,
      "step": 78775
    },
    {
      "entropy": 5.983337545394898,
      "epoch": 6.129313363158919,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001988267358837087,
      "loss": 4.9898,
      "mean_token_accuracy": 0.21921508014202118,
      "num_tokens": 136630732.0,
      "step": 78780
    },
    {
      "entropy": 6.01609845161438,
      "epoch": 6.129702392530636,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001988006575022147,
      "loss": 4.938,
      "mean_token_accuracy": 0.22378284186124803,
      "num_tokens": 136639194.0,
      "step": 78785
    },
    {
      "entropy": 5.922202491760254,
      "epoch": 6.130091421902353,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019877458027692458,
      "loss": 4.8713,
      "mean_token_accuracy": 0.21950026005506515,
      "num_tokens": 136647478.0,
      "step": 78790
    },
    {
      "entropy": 5.996760463714599,
      "epoch": 6.130480451274071,
      "grad_norm": 1.5,
      "learning_rate": 0.00019874850420823376,
      "loss": 5.0454,
      "mean_token_accuracy": 0.21112051606178284,
      "num_tokens": 136655967.0,
      "step": 78795
    },
    {
      "entropy": 5.984359312057495,
      "epoch": 6.130869480645789,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019872242929653812,
      "loss": 4.9134,
      "mean_token_accuracy": 0.21882197111845017,
      "num_tokens": 136666207.0,
      "step": 78800
    },
    {
      "entropy": 5.971501159667969,
      "epoch": 6.131258510017506,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019869635554223313,
      "loss": 4.8592,
      "mean_token_accuracy": 0.2239970788359642,
      "num_tokens": 136674661.0,
      "step": 78805
    },
    {
      "entropy": 5.944386577606201,
      "epoch": 6.131647539389224,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019867028294571454,
      "loss": 4.8922,
      "mean_token_accuracy": 0.22543565779924393,
      "num_tokens": 136683213.0,
      "step": 78810
    },
    {
      "entropy": 5.996570444107055,
      "epoch": 6.132036568760942,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019864421150737787,
      "loss": 5.014,
      "mean_token_accuracy": 0.20677502304315568,
      "num_tokens": 136692565.0,
      "step": 78815
    },
    {
      "entropy": 5.9893182754516605,
      "epoch": 6.132425598132659,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019861814122761883,
      "loss": 4.9032,
      "mean_token_accuracy": 0.2206261858344078,
      "num_tokens": 136700712.0,
      "step": 78820
    },
    {
      "entropy": 5.91471266746521,
      "epoch": 6.132814627504376,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001985920721068329,
      "loss": 4.8796,
      "mean_token_accuracy": 0.23263075947761536,
      "num_tokens": 136709414.0,
      "step": 78825
    },
    {
      "entropy": 5.893651866912842,
      "epoch": 6.133203656876094,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019856600414541566,
      "loss": 4.803,
      "mean_token_accuracy": 0.2273870438337326,
      "num_tokens": 136718324.0,
      "step": 78830
    },
    {
      "entropy": 5.9667566299438475,
      "epoch": 6.133592686247812,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019853993734376263,
      "loss": 4.9674,
      "mean_token_accuracy": 0.22334721088409423,
      "num_tokens": 136727026.0,
      "step": 78835
    },
    {
      "entropy": 5.943632650375366,
      "epoch": 6.133981715619529,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019851387170226936,
      "loss": 4.8731,
      "mean_token_accuracy": 0.21944154649972916,
      "num_tokens": 136736140.0,
      "step": 78840
    },
    {
      "entropy": 5.925212621688843,
      "epoch": 6.134370744991247,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001984878072213313,
      "loss": 4.8879,
      "mean_token_accuracy": 0.22796605825424193,
      "num_tokens": 136745600.0,
      "step": 78845
    },
    {
      "entropy": 6.007452154159546,
      "epoch": 6.134759774362965,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019846174390134396,
      "loss": 5.0104,
      "mean_token_accuracy": 0.21322810351848603,
      "num_tokens": 136755499.0,
      "step": 78850
    },
    {
      "entropy": 5.949771213531494,
      "epoch": 6.1351488037346815,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019843568174270293,
      "loss": 4.828,
      "mean_token_accuracy": 0.22947811782360078,
      "num_tokens": 136763429.0,
      "step": 78855
    },
    {
      "entropy": 5.978890609741211,
      "epoch": 6.135537833106399,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019840962074580348,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22466979175806046,
      "num_tokens": 136771450.0,
      "step": 78860
    },
    {
      "entropy": 5.9711432456970215,
      "epoch": 6.135926862478117,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001983835609110412,
      "loss": 4.925,
      "mean_token_accuracy": 0.2249930590391159,
      "num_tokens": 136779694.0,
      "step": 78865
    },
    {
      "entropy": 5.962436819076538,
      "epoch": 6.136315891849835,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019835750223881144,
      "loss": 4.9465,
      "mean_token_accuracy": 0.22128317803144454,
      "num_tokens": 136788373.0,
      "step": 78870
    },
    {
      "entropy": 5.947157764434815,
      "epoch": 6.136704921221552,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019833144472950953,
      "loss": 4.8901,
      "mean_token_accuracy": 0.22303520888090134,
      "num_tokens": 136796546.0,
      "step": 78875
    },
    {
      "entropy": 5.96454610824585,
      "epoch": 6.13709395059327,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000198305388383531,
      "loss": 4.9106,
      "mean_token_accuracy": 0.2254924327135086,
      "num_tokens": 136805349.0,
      "step": 78880
    },
    {
      "entropy": 6.010669565200805,
      "epoch": 6.137482979964988,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00019827933320127106,
      "loss": 4.9488,
      "mean_token_accuracy": 0.22360581308603286,
      "num_tokens": 136814936.0,
      "step": 78885
    },
    {
      "entropy": 5.990770626068115,
      "epoch": 6.1378720093367045,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019825327918312515,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22744717001914977,
      "num_tokens": 136822974.0,
      "step": 78890
    },
    {
      "entropy": 5.972367906570435,
      "epoch": 6.138261038708422,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019822722632948858,
      "loss": 4.8329,
      "mean_token_accuracy": 0.22646649330854415,
      "num_tokens": 136830898.0,
      "step": 78895
    },
    {
      "entropy": 6.053360414505005,
      "epoch": 6.13865006808014,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001982011746407567,
      "loss": 5.0345,
      "mean_token_accuracy": 0.21833558082580568,
      "num_tokens": 136839349.0,
      "step": 78900
    },
    {
      "entropy": 5.955715131759644,
      "epoch": 6.139039097451858,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001981751241173247,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2290905609726906,
      "num_tokens": 136847599.0,
      "step": 78905
    },
    {
      "entropy": 5.97434515953064,
      "epoch": 6.139428126823575,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019814907475958793,
      "loss": 4.8743,
      "mean_token_accuracy": 0.22201920300722122,
      "num_tokens": 136856276.0,
      "step": 78910
    },
    {
      "entropy": 5.980785942077636,
      "epoch": 6.139817156195293,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001981230265679415,
      "loss": 4.9245,
      "mean_token_accuracy": 0.22012439966201783,
      "num_tokens": 136865844.0,
      "step": 78915
    },
    {
      "entropy": 5.879122924804688,
      "epoch": 6.140206185567011,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019809697954278088,
      "loss": 4.8413,
      "mean_token_accuracy": 0.23089629858732225,
      "num_tokens": 136874719.0,
      "step": 78920
    },
    {
      "entropy": 5.9317444324493405,
      "epoch": 6.1405952149387275,
      "grad_norm": 1.671875,
      "learning_rate": 0.0001980709336845012,
      "loss": 4.876,
      "mean_token_accuracy": 0.22598715722560883,
      "num_tokens": 136883323.0,
      "step": 78925
    },
    {
      "entropy": 6.0239190578460695,
      "epoch": 6.140984244310445,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019804488899349772,
      "loss": 5.0145,
      "mean_token_accuracy": 0.2152661070227623,
      "num_tokens": 136892019.0,
      "step": 78930
    },
    {
      "entropy": 5.940943622589112,
      "epoch": 6.141373273682163,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019801884547016553,
      "loss": 4.9075,
      "mean_token_accuracy": 0.22422507405281067,
      "num_tokens": 136901188.0,
      "step": 78935
    },
    {
      "entropy": 5.981811475753784,
      "epoch": 6.141762303053881,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001979928031148998,
      "loss": 4.9283,
      "mean_token_accuracy": 0.22906084656715392,
      "num_tokens": 136910531.0,
      "step": 78940
    },
    {
      "entropy": 5.953662538528443,
      "epoch": 6.142151332425598,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001979667619280957,
      "loss": 4.9662,
      "mean_token_accuracy": 0.21604142636060714,
      "num_tokens": 136920045.0,
      "step": 78945
    },
    {
      "entropy": 5.986766147613525,
      "epoch": 6.142540361797316,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019794072191014838,
      "loss": 4.9043,
      "mean_token_accuracy": 0.22126989513635636,
      "num_tokens": 136929696.0,
      "step": 78950
    },
    {
      "entropy": 5.971985769271851,
      "epoch": 6.142929391169034,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019791468306145298,
      "loss": 4.8738,
      "mean_token_accuracy": 0.224392768740654,
      "num_tokens": 136937546.0,
      "step": 78955
    },
    {
      "entropy": 5.951761054992676,
      "epoch": 6.1433184205407505,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001978886453824046,
      "loss": 4.9304,
      "mean_token_accuracy": 0.21947243064641953,
      "num_tokens": 136947295.0,
      "step": 78960
    },
    {
      "entropy": 5.983538198471069,
      "epoch": 6.143707449912468,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001978626088733982,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22671755850315095,
      "num_tokens": 136955361.0,
      "step": 78965
    },
    {
      "entropy": 5.971900558471679,
      "epoch": 6.144096479284186,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019783657353482904,
      "loss": 4.9096,
      "mean_token_accuracy": 0.23156654238700866,
      "num_tokens": 136964212.0,
      "step": 78970
    },
    {
      "entropy": 5.97320408821106,
      "epoch": 6.144485508655904,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019781053936709198,
      "loss": 4.9129,
      "mean_token_accuracy": 0.2203710913658142,
      "num_tokens": 136972432.0,
      "step": 78975
    },
    {
      "entropy": 5.927911949157715,
      "epoch": 6.144874538027621,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019778450637058216,
      "loss": 4.8422,
      "mean_token_accuracy": 0.22761031687259675,
      "num_tokens": 136980496.0,
      "step": 78980
    },
    {
      "entropy": 5.959466981887817,
      "epoch": 6.145263567399339,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019775847454569446,
      "loss": 4.894,
      "mean_token_accuracy": 0.22535288631916045,
      "num_tokens": 136988739.0,
      "step": 78985
    },
    {
      "entropy": 5.942465591430664,
      "epoch": 6.145652596771056,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019773244389282408,
      "loss": 4.8608,
      "mean_token_accuracy": 0.21792869120836258,
      "num_tokens": 136996493.0,
      "step": 78990
    },
    {
      "entropy": 5.926329708099365,
      "epoch": 6.1460416261427735,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001977064144123658,
      "loss": 4.8947,
      "mean_token_accuracy": 0.2257568806409836,
      "num_tokens": 137005542.0,
      "step": 78995
    },
    {
      "entropy": 5.91122612953186,
      "epoch": 6.146430655514491,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019768038610471473,
      "loss": 4.858,
      "mean_token_accuracy": 0.2233612507581711,
      "num_tokens": 137014396.0,
      "step": 79000
    },
    {
      "entropy": 5.96497688293457,
      "epoch": 6.146819684886209,
      "grad_norm": 1.671875,
      "learning_rate": 0.00019765435897026575,
      "loss": 5.0373,
      "mean_token_accuracy": 0.2129688560962677,
      "num_tokens": 137023410.0,
      "step": 79005
    },
    {
      "entropy": 6.008572053909302,
      "epoch": 6.147208714257927,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019762833300941373,
      "loss": 4.9062,
      "mean_token_accuracy": 0.2198584035038948,
      "num_tokens": 137031869.0,
      "step": 79010
    },
    {
      "entropy": 5.98424015045166,
      "epoch": 6.147597743629644,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019760230822255358,
      "loss": 4.8468,
      "mean_token_accuracy": 0.23060715198516846,
      "num_tokens": 137040570.0,
      "step": 79015
    },
    {
      "entropy": 5.996257543563843,
      "epoch": 6.147986773001362,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019757628461008025,
      "loss": 4.9843,
      "mean_token_accuracy": 0.21626638621091843,
      "num_tokens": 137048962.0,
      "step": 79020
    },
    {
      "entropy": 6.025349712371826,
      "epoch": 6.148375802373079,
      "grad_norm": 1.671875,
      "learning_rate": 0.0001975502621723886,
      "loss": 4.9559,
      "mean_token_accuracy": 0.21729299277067185,
      "num_tokens": 137056989.0,
      "step": 79025
    },
    {
      "entropy": 5.938254308700562,
      "epoch": 6.1487648317447965,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019752424090987338,
      "loss": 4.8855,
      "mean_token_accuracy": 0.22599090337753297,
      "num_tokens": 137065798.0,
      "step": 79030
    },
    {
      "entropy": 5.970391273498535,
      "epoch": 6.149153861116514,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019749822082292947,
      "loss": 4.9822,
      "mean_token_accuracy": 0.21601008027791976,
      "num_tokens": 137074650.0,
      "step": 79035
    },
    {
      "entropy": 5.909721612930298,
      "epoch": 6.149542890488232,
      "grad_norm": 1.6875,
      "learning_rate": 0.00019747220191195175,
      "loss": 4.8705,
      "mean_token_accuracy": 0.2242736592888832,
      "num_tokens": 137082881.0,
      "step": 79040
    },
    {
      "entropy": 5.976190662384033,
      "epoch": 6.14993191985995,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019744618417733494,
      "loss": 4.9843,
      "mean_token_accuracy": 0.21500724852085112,
      "num_tokens": 137091456.0,
      "step": 79045
    },
    {
      "entropy": 5.963031864166259,
      "epoch": 6.150320949231667,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001974201676194738,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22342006862163544,
      "num_tokens": 137100882.0,
      "step": 79050
    },
    {
      "entropy": 5.978305244445801,
      "epoch": 6.150709978603385,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019739415223876322,
      "loss": 4.9527,
      "mean_token_accuracy": 0.22125126272439957,
      "num_tokens": 137109596.0,
      "step": 79055
    },
    {
      "entropy": 6.027043724060059,
      "epoch": 6.151099007975102,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019736813803559783,
      "loss": 5.0167,
      "mean_token_accuracy": 0.21568835377693177,
      "num_tokens": 137117941.0,
      "step": 79060
    },
    {
      "entropy": 5.9933525085449215,
      "epoch": 6.1514880373468195,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001973421250103724,
      "loss": 4.9597,
      "mean_token_accuracy": 0.2129237860441208,
      "num_tokens": 137127069.0,
      "step": 79065
    },
    {
      "entropy": 5.996827125549316,
      "epoch": 6.151877066718537,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019731611316348158,
      "loss": 4.9822,
      "mean_token_accuracy": 0.21508847028017045,
      "num_tokens": 137135626.0,
      "step": 79070
    },
    {
      "entropy": 5.92444486618042,
      "epoch": 6.152266096090255,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019729010249532015,
      "loss": 4.8761,
      "mean_token_accuracy": 0.22490203827619554,
      "num_tokens": 137144552.0,
      "step": 79075
    },
    {
      "entropy": 5.9769879341125485,
      "epoch": 6.152655125461973,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019726409300628266,
      "loss": 4.9541,
      "mean_token_accuracy": 0.22209542542695998,
      "num_tokens": 137153028.0,
      "step": 79080
    },
    {
      "entropy": 5.9671900272369385,
      "epoch": 6.15304415483369,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019723808469676384,
      "loss": 4.9703,
      "mean_token_accuracy": 0.21768055409193038,
      "num_tokens": 137162034.0,
      "step": 79085
    },
    {
      "entropy": 6.019922590255737,
      "epoch": 6.153433184205407,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019721207756715842,
      "loss": 4.9565,
      "mean_token_accuracy": 0.22168430387973787,
      "num_tokens": 137170796.0,
      "step": 79090
    },
    {
      "entropy": 5.9583827495574955,
      "epoch": 6.153822213577125,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019718607161786083,
      "loss": 4.8891,
      "mean_token_accuracy": 0.2255186378955841,
      "num_tokens": 137179090.0,
      "step": 79095
    },
    {
      "entropy": 5.916441535949707,
      "epoch": 6.1542112429488425,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001971600668492658,
      "loss": 4.8354,
      "mean_token_accuracy": 0.22876286655664443,
      "num_tokens": 137187952.0,
      "step": 79100
    },
    {
      "entropy": 6.02269229888916,
      "epoch": 6.15460027232056,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00019713406326176775,
      "loss": 4.9249,
      "mean_token_accuracy": 0.2217179149389267,
      "num_tokens": 137196063.0,
      "step": 79105
    },
    {
      "entropy": 5.986225652694702,
      "epoch": 6.154989301692278,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00019710806085576144,
      "loss": 5.0036,
      "mean_token_accuracy": 0.21810629218816757,
      "num_tokens": 137204238.0,
      "step": 79110
    },
    {
      "entropy": 5.912337970733643,
      "epoch": 6.155378331063996,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001970820596316412,
      "loss": 4.8122,
      "mean_token_accuracy": 0.23265117406845093,
      "num_tokens": 137212609.0,
      "step": 79115
    },
    {
      "entropy": 6.0232611179351805,
      "epoch": 6.155767360435713,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00019705605958980182,
      "loss": 5.0393,
      "mean_token_accuracy": 0.21273597478866577,
      "num_tokens": 137220722.0,
      "step": 79120
    },
    {
      "entropy": 5.9417329788208,
      "epoch": 6.15615638980743,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019703006073063767,
      "loss": 4.8677,
      "mean_token_accuracy": 0.22172589004039764,
      "num_tokens": 137229414.0,
      "step": 79125
    },
    {
      "entropy": 6.011443042755127,
      "epoch": 6.156545419179148,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00019700406305454322,
      "loss": 4.9987,
      "mean_token_accuracy": 0.21313070356845856,
      "num_tokens": 137238373.0,
      "step": 79130
    },
    {
      "entropy": 5.930182600021363,
      "epoch": 6.1569344485508655,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00019697806656191296,
      "loss": 4.8703,
      "mean_token_accuracy": 0.22149323374032975,
      "num_tokens": 137246410.0,
      "step": 79135
    },
    {
      "entropy": 5.9598595142364506,
      "epoch": 6.157323477922583,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001969520712531414,
      "loss": 4.9707,
      "mean_token_accuracy": 0.2149055615067482,
      "num_tokens": 137255714.0,
      "step": 79140
    },
    {
      "entropy": 5.943698120117188,
      "epoch": 6.157712507294301,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019692607712862287,
      "loss": 4.8926,
      "mean_token_accuracy": 0.21975095123052596,
      "num_tokens": 137264618.0,
      "step": 79145
    },
    {
      "entropy": 5.953506422042847,
      "epoch": 6.158101536666019,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00019690008418875188,
      "loss": 4.8495,
      "mean_token_accuracy": 0.2311854749917984,
      "num_tokens": 137272377.0,
      "step": 79150
    },
    {
      "entropy": 5.952644872665405,
      "epoch": 6.158490566037736,
      "grad_norm": 1.765625,
      "learning_rate": 0.00019687409243392278,
      "loss": 4.8997,
      "mean_token_accuracy": 0.2220569968223572,
      "num_tokens": 137281267.0,
      "step": 79155
    },
    {
      "entropy": 5.8947947978973385,
      "epoch": 6.158879595409453,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019684810186453,
      "loss": 4.8505,
      "mean_token_accuracy": 0.2208537772297859,
      "num_tokens": 137290474.0,
      "step": 79160
    },
    {
      "entropy": 5.90294451713562,
      "epoch": 6.159268624781171,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019682211248096793,
      "loss": 4.8343,
      "mean_token_accuracy": 0.23102210611104965,
      "num_tokens": 137299060.0,
      "step": 79165
    },
    {
      "entropy": 5.9517073154449465,
      "epoch": 6.1596576541528885,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001967961242836308,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21876242607831956,
      "num_tokens": 137307440.0,
      "step": 79170
    },
    {
      "entropy": 5.93889684677124,
      "epoch": 6.160046683524606,
      "grad_norm": 1.578125,
      "learning_rate": 0.0001967701372729131,
      "loss": 4.872,
      "mean_token_accuracy": 0.22552878111600877,
      "num_tokens": 137315842.0,
      "step": 79175
    },
    {
      "entropy": 5.973778009414673,
      "epoch": 6.160435712896324,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019674415144920894,
      "loss": 4.9474,
      "mean_token_accuracy": 0.2187244176864624,
      "num_tokens": 137324297.0,
      "step": 79180
    },
    {
      "entropy": 5.978236818313599,
      "epoch": 6.1608247422680416,
      "grad_norm": 1.640625,
      "learning_rate": 0.0001967181668129127,
      "loss": 4.9039,
      "mean_token_accuracy": 0.22910358607769013,
      "num_tokens": 137332529.0,
      "step": 79185
    },
    {
      "entropy": 5.873778676986694,
      "epoch": 6.161213771639758,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019669218336441875,
      "loss": 4.8317,
      "mean_token_accuracy": 0.2273958742618561,
      "num_tokens": 137341070.0,
      "step": 79190
    },
    {
      "entropy": 5.965333271026611,
      "epoch": 6.161602801011476,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019666620110412125,
      "loss": 4.9328,
      "mean_token_accuracy": 0.21960118561983108,
      "num_tokens": 137350179.0,
      "step": 79195
    },
    {
      "entropy": 5.949791336059571,
      "epoch": 6.161991830383194,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019664022003241456,
      "loss": 4.9126,
      "mean_token_accuracy": 0.21660742461681365,
      "num_tokens": 137358216.0,
      "step": 79200
    },
    {
      "entropy": 5.930989837646484,
      "epoch": 6.1623808597549115,
      "grad_norm": 1.640625,
      "learning_rate": 0.00019661424014969276,
      "loss": 4.9684,
      "mean_token_accuracy": 0.2154742181301117,
      "num_tokens": 137366184.0,
      "step": 79205
    },
    {
      "entropy": 6.028099918365479,
      "epoch": 6.162769889126629,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019658826145635007,
      "loss": 5.0211,
      "mean_token_accuracy": 0.21172949224710463,
      "num_tokens": 137375515.0,
      "step": 79210
    },
    {
      "entropy": 5.977033090591431,
      "epoch": 6.163158918498347,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019656228395278076,
      "loss": 5.0029,
      "mean_token_accuracy": 0.21573712080717086,
      "num_tokens": 137384955.0,
      "step": 79215
    },
    {
      "entropy": 5.990919399261474,
      "epoch": 6.1635479478700645,
      "grad_norm": 1.5,
      "learning_rate": 0.0001965363076393789,
      "loss": 4.9516,
      "mean_token_accuracy": 0.22270189225673676,
      "num_tokens": 137394338.0,
      "step": 79220
    },
    {
      "entropy": 5.980376482009888,
      "epoch": 6.163936977241781,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019651033251653866,
      "loss": 4.9429,
      "mean_token_accuracy": 0.21708168536424638,
      "num_tokens": 137403552.0,
      "step": 79225
    },
    {
      "entropy": 5.90495023727417,
      "epoch": 6.164326006613499,
      "grad_norm": 1.625,
      "learning_rate": 0.00019648435858465424,
      "loss": 4.8757,
      "mean_token_accuracy": 0.22392307072877884,
      "num_tokens": 137412659.0,
      "step": 79230
    },
    {
      "entropy": 5.949674606323242,
      "epoch": 6.164715035985217,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019645838584411967,
      "loss": 4.9305,
      "mean_token_accuracy": 0.22761750668287278,
      "num_tokens": 137421907.0,
      "step": 79235
    },
    {
      "entropy": 5.99226942062378,
      "epoch": 6.1651040653569344,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019643241429532916,
      "loss": 4.9704,
      "mean_token_accuracy": 0.22198613733053207,
      "num_tokens": 137430323.0,
      "step": 79240
    },
    {
      "entropy": 5.942294788360596,
      "epoch": 6.165493094728652,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019640644393867663,
      "loss": 4.8998,
      "mean_token_accuracy": 0.22801827192306517,
      "num_tokens": 137438572.0,
      "step": 79245
    },
    {
      "entropy": 5.899667406082154,
      "epoch": 6.16588212410037,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019638047477455616,
      "loss": 4.7994,
      "mean_token_accuracy": 0.23496681153774263,
      "num_tokens": 137447252.0,
      "step": 79250
    },
    {
      "entropy": 5.993371105194091,
      "epoch": 6.1662711534720875,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00019635450680336198,
      "loss": 4.9129,
      "mean_token_accuracy": 0.22194566130638121,
      "num_tokens": 137455349.0,
      "step": 79255
    },
    {
      "entropy": 5.895857191085815,
      "epoch": 6.166660182843804,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019632854002548796,
      "loss": 4.7982,
      "mean_token_accuracy": 0.2311977669596672,
      "num_tokens": 137463526.0,
      "step": 79260
    },
    {
      "entropy": 5.877495765686035,
      "epoch": 6.167049212215522,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019630257444132803,
      "loss": 4.8283,
      "mean_token_accuracy": 0.22517864853143693,
      "num_tokens": 137471985.0,
      "step": 79265
    },
    {
      "entropy": 5.992117309570313,
      "epoch": 6.16743824158724,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019627661005127634,
      "loss": 5.0062,
      "mean_token_accuracy": 0.21477101892232894,
      "num_tokens": 137480972.0,
      "step": 79270
    },
    {
      "entropy": 5.9760339736938475,
      "epoch": 6.167827270958957,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001962506468557268,
      "loss": 4.8599,
      "mean_token_accuracy": 0.23819465637207032,
      "num_tokens": 137489471.0,
      "step": 79275
    },
    {
      "entropy": 6.02827672958374,
      "epoch": 6.168216300330675,
      "grad_norm": 1.671875,
      "learning_rate": 0.0001962246848550734,
      "loss": 5.0189,
      "mean_token_accuracy": 0.21359410583972932,
      "num_tokens": 137498267.0,
      "step": 79280
    },
    {
      "entropy": 5.923919439315796,
      "epoch": 6.168605329702393,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019619872404971002,
      "loss": 4.9228,
      "mean_token_accuracy": 0.21790040880441666,
      "num_tokens": 137506604.0,
      "step": 79285
    },
    {
      "entropy": 5.989980459213257,
      "epoch": 6.1689943590741105,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019617276444003053,
      "loss": 4.9587,
      "mean_token_accuracy": 0.21990588158369065,
      "num_tokens": 137515354.0,
      "step": 79290
    },
    {
      "entropy": 5.927034521102906,
      "epoch": 6.169383388445827,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019614680602642886,
      "loss": 4.8203,
      "mean_token_accuracy": 0.23616329133510588,
      "num_tokens": 137523879.0,
      "step": 79295
    },
    {
      "entropy": 6.009552907943726,
      "epoch": 6.169772417817545,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00019612084880929888,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2243254154920578,
      "num_tokens": 137532195.0,
      "step": 79300
    },
    {
      "entropy": 5.948806190490723,
      "epoch": 6.170161447189263,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019609489278903452,
      "loss": 4.885,
      "mean_token_accuracy": 0.22203729152679444,
      "num_tokens": 137540122.0,
      "step": 79305
    },
    {
      "entropy": 5.892035627365113,
      "epoch": 6.17055047656098,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019606893796602953,
      "loss": 4.8427,
      "mean_token_accuracy": 0.22220426201820373,
      "num_tokens": 137549180.0,
      "step": 79310
    },
    {
      "entropy": 6.005953454971314,
      "epoch": 6.170939505932698,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001960429843406778,
      "loss": 4.974,
      "mean_token_accuracy": 0.2069421410560608,
      "num_tokens": 137558363.0,
      "step": 79315
    },
    {
      "entropy": 5.989331865310669,
      "epoch": 6.171328535304416,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0001960170319133731,
      "loss": 4.927,
      "mean_token_accuracy": 0.22333947122097014,
      "num_tokens": 137566894.0,
      "step": 79320
    },
    {
      "entropy": 5.946901321411133,
      "epoch": 6.171717564676133,
      "grad_norm": 1.625,
      "learning_rate": 0.00019599108068450922,
      "loss": 4.9291,
      "mean_token_accuracy": 0.2165914550423622,
      "num_tokens": 137576106.0,
      "step": 79325
    },
    {
      "entropy": 5.933172082901001,
      "epoch": 6.17210659404785,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019596513065447995,
      "loss": 4.8518,
      "mean_token_accuracy": 0.2256227970123291,
      "num_tokens": 137585392.0,
      "step": 79330
    },
    {
      "entropy": 5.977402782440185,
      "epoch": 6.172495623419568,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019593918182367903,
      "loss": 4.9463,
      "mean_token_accuracy": 0.21733978837728501,
      "num_tokens": 137593687.0,
      "step": 79335
    },
    {
      "entropy": 5.967447090148926,
      "epoch": 6.172884652791286,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019591323419250017,
      "loss": 4.9087,
      "mean_token_accuracy": 0.21444923728704451,
      "num_tokens": 137602358.0,
      "step": 79340
    },
    {
      "entropy": 5.9588024616241455,
      "epoch": 6.173273682163003,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00019588728776133714,
      "loss": 4.8787,
      "mean_token_accuracy": 0.23151189237833023,
      "num_tokens": 137611140.0,
      "step": 79345
    },
    {
      "entropy": 5.965338897705078,
      "epoch": 6.173662711534721,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00019586134253058357,
      "loss": 4.9036,
      "mean_token_accuracy": 0.224187932908535,
      "num_tokens": 137619859.0,
      "step": 79350
    },
    {
      "entropy": 5.934031391143799,
      "epoch": 6.174051740906439,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00019583539850063315,
      "loss": 4.8716,
      "mean_token_accuracy": 0.2273362696170807,
      "num_tokens": 137627885.0,
      "step": 79355
    },
    {
      "entropy": 5.9629127979278564,
      "epoch": 6.174440770278156,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019580945567187956,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21639098078012467,
      "num_tokens": 137636903.0,
      "step": 79360
    },
    {
      "entropy": 5.962752628326416,
      "epoch": 6.174829799649873,
      "grad_norm": 1.625,
      "learning_rate": 0.0001957835140447165,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22388675212860107,
      "num_tokens": 137645418.0,
      "step": 79365
    },
    {
      "entropy": 6.024716854095459,
      "epoch": 6.175218829021591,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0001957575736195375,
      "loss": 4.9861,
      "mean_token_accuracy": 0.222356940805912,
      "num_tokens": 137654348.0,
      "step": 79370
    },
    {
      "entropy": 5.971407651901245,
      "epoch": 6.175607858393309,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001957316343967362,
      "loss": 4.858,
      "mean_token_accuracy": 0.22231405675411225,
      "num_tokens": 137662264.0,
      "step": 79375
    },
    {
      "entropy": 6.009288549423218,
      "epoch": 6.175996887765026,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019570569637670604,
      "loss": 4.9494,
      "mean_token_accuracy": 0.2167347177863121,
      "num_tokens": 137670730.0,
      "step": 79380
    },
    {
      "entropy": 5.95785870552063,
      "epoch": 6.176385917136744,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019567975955984079,
      "loss": 4.9001,
      "mean_token_accuracy": 0.22899568527936937,
      "num_tokens": 137680156.0,
      "step": 79385
    },
    {
      "entropy": 6.048834753036499,
      "epoch": 6.176774946508462,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019565382394653403,
      "loss": 4.9417,
      "mean_token_accuracy": 0.22017960995435715,
      "num_tokens": 137688878.0,
      "step": 79390
    },
    {
      "entropy": 6.065632009506226,
      "epoch": 6.177163975880179,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001956278895371792,
      "loss": 5.0151,
      "mean_token_accuracy": 0.2181781381368637,
      "num_tokens": 137697321.0,
      "step": 79395
    },
    {
      "entropy": 5.957924747467041,
      "epoch": 6.177553005251896,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019560195633216976,
      "loss": 4.9339,
      "mean_token_accuracy": 0.22174030244350434,
      "num_tokens": 137706436.0,
      "step": 79400
    },
    {
      "entropy": 6.017624378204346,
      "epoch": 6.177942034623614,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019557602433189925,
      "loss": 4.9952,
      "mean_token_accuracy": 0.21589977741241456,
      "num_tokens": 137715391.0,
      "step": 79405
    },
    {
      "entropy": 5.956071090698242,
      "epoch": 6.178331063995332,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019555009353676116,
      "loss": 4.8868,
      "mean_token_accuracy": 0.22691207975149155,
      "num_tokens": 137723901.0,
      "step": 79410
    },
    {
      "entropy": 5.941185426712036,
      "epoch": 6.178720093367049,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019552416394714894,
      "loss": 4.9698,
      "mean_token_accuracy": 0.21797839403152466,
      "num_tokens": 137733109.0,
      "step": 79415
    },
    {
      "entropy": 5.9432957649230955,
      "epoch": 6.179109122738767,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019549823556345603,
      "loss": 4.877,
      "mean_token_accuracy": 0.2284097358584404,
      "num_tokens": 137742777.0,
      "step": 79420
    },
    {
      "entropy": 5.972169208526611,
      "epoch": 6.179498152110484,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019547230838607593,
      "loss": 4.9663,
      "mean_token_accuracy": 0.21303676962852477,
      "num_tokens": 137751919.0,
      "step": 79425
    },
    {
      "entropy": 5.955071544647216,
      "epoch": 6.179887181482202,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019544638241540186,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21741227060556412,
      "num_tokens": 137760947.0,
      "step": 79430
    },
    {
      "entropy": 5.942549419403076,
      "epoch": 6.180276210853919,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019542045765182737,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22595133036375045,
      "num_tokens": 137769127.0,
      "step": 79435
    },
    {
      "entropy": 5.991839504241943,
      "epoch": 6.180665240225637,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019539453409574576,
      "loss": 4.8982,
      "mean_token_accuracy": 0.22174542248249055,
      "num_tokens": 137778062.0,
      "step": 79440
    },
    {
      "entropy": 6.029169750213623,
      "epoch": 6.181054269597355,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019536861174755045,
      "loss": 5.1001,
      "mean_token_accuracy": 0.20719262063503266,
      "num_tokens": 137787024.0,
      "step": 79445
    },
    {
      "entropy": 5.953707265853882,
      "epoch": 6.181443298969072,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019534269060763454,
      "loss": 4.9305,
      "mean_token_accuracy": 0.21859549283981322,
      "num_tokens": 137794934.0,
      "step": 79450
    },
    {
      "entropy": 5.936577415466308,
      "epoch": 6.18183232834079,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00019531677067639164,
      "loss": 4.8852,
      "mean_token_accuracy": 0.22338711768388747,
      "num_tokens": 137803650.0,
      "step": 79455
    },
    {
      "entropy": 6.0225691318511965,
      "epoch": 6.182221357712507,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019529085195421492,
      "loss": 5.0306,
      "mean_token_accuracy": 0.21426304131746293,
      "num_tokens": 137813057.0,
      "step": 79460
    },
    {
      "entropy": 5.9748406410217285,
      "epoch": 6.182610387084225,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019526493444149763,
      "loss": 4.9232,
      "mean_token_accuracy": 0.21800258308649062,
      "num_tokens": 137821840.0,
      "step": 79465
    },
    {
      "entropy": 5.931950092315674,
      "epoch": 6.182999416455942,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019523901813863303,
      "loss": 4.8759,
      "mean_token_accuracy": 0.22692939192056655,
      "num_tokens": 137829634.0,
      "step": 79470
    },
    {
      "entropy": 5.998644399642944,
      "epoch": 6.18338844582766,
      "grad_norm": 1.6875,
      "learning_rate": 0.00019521310304601447,
      "loss": 4.894,
      "mean_token_accuracy": 0.22558328211307527,
      "num_tokens": 137838357.0,
      "step": 79475
    },
    {
      "entropy": 5.998757600784302,
      "epoch": 6.183777475199378,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000195187189164035,
      "loss": 4.9403,
      "mean_token_accuracy": 0.2135901302099228,
      "num_tokens": 137845953.0,
      "step": 79480
    },
    {
      "entropy": 5.94584379196167,
      "epoch": 6.184166504571095,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001951612764930879,
      "loss": 4.9892,
      "mean_token_accuracy": 0.21195582002401353,
      "num_tokens": 137854551.0,
      "step": 79485
    },
    {
      "entropy": 5.941933250427246,
      "epoch": 6.184555533942813,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019513536503356637,
      "loss": 4.9247,
      "mean_token_accuracy": 0.22057410329580307,
      "num_tokens": 137863231.0,
      "step": 79490
    },
    {
      "entropy": 6.003873825073242,
      "epoch": 6.18494456331453,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019510945478586351,
      "loss": 4.9836,
      "mean_token_accuracy": 0.21771241575479508,
      "num_tokens": 137871613.0,
      "step": 79495
    },
    {
      "entropy": 5.9731488704681395,
      "epoch": 6.185333592686248,
      "grad_norm": 1.5,
      "learning_rate": 0.00019508354575037258,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2221928969025612,
      "num_tokens": 137880319.0,
      "step": 79500
    },
    {
      "entropy": 5.960566711425781,
      "epoch": 6.185722622057965,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001950576379274866,
      "loss": 4.9199,
      "mean_token_accuracy": 0.22295173704624177,
      "num_tokens": 137889489.0,
      "step": 79505
    },
    {
      "entropy": 6.000883388519287,
      "epoch": 6.186111651429683,
      "grad_norm": 1.6875,
      "learning_rate": 0.0001950317313175987,
      "loss": 4.9715,
      "mean_token_accuracy": 0.21598806232213974,
      "num_tokens": 137897930.0,
      "step": 79510
    },
    {
      "entropy": 6.0559666633605955,
      "epoch": 6.186500680801401,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019500582592110207,
      "loss": 4.9724,
      "mean_token_accuracy": 0.21480585932731627,
      "num_tokens": 137906947.0,
      "step": 79515
    },
    {
      "entropy": 5.97120475769043,
      "epoch": 6.186889710173118,
      "grad_norm": 1.640625,
      "learning_rate": 0.00019497992173838963,
      "loss": 4.891,
      "mean_token_accuracy": 0.22061660438776015,
      "num_tokens": 137916007.0,
      "step": 79520
    },
    {
      "entropy": 5.993329238891602,
      "epoch": 6.187278739544835,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019495401876985457,
      "loss": 4.9589,
      "mean_token_accuracy": 0.2177450478076935,
      "num_tokens": 137925149.0,
      "step": 79525
    },
    {
      "entropy": 5.933354759216309,
      "epoch": 6.187667768916553,
      "grad_norm": 1.734375,
      "learning_rate": 0.00019492811701588982,
      "loss": 4.9226,
      "mean_token_accuracy": 0.22297506630420685,
      "num_tokens": 137933970.0,
      "step": 79530
    },
    {
      "entropy": 6.009746170043945,
      "epoch": 6.188056798288271,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001949022164768885,
      "loss": 4.9331,
      "mean_token_accuracy": 0.21894347220659255,
      "num_tokens": 137942579.0,
      "step": 79535
    },
    {
      "entropy": 5.977053260803222,
      "epoch": 6.188445827659988,
      "grad_norm": 1.703125,
      "learning_rate": 0.0001948763171532435,
      "loss": 4.9087,
      "mean_token_accuracy": 0.23211622834205628,
      "num_tokens": 137951049.0,
      "step": 79540
    },
    {
      "entropy": 5.934276342391968,
      "epoch": 6.188834857031706,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001948504190453479,
      "loss": 4.8654,
      "mean_token_accuracy": 0.23529740571975707,
      "num_tokens": 137958726.0,
      "step": 79545
    },
    {
      "entropy": 5.928645038604737,
      "epoch": 6.189223886403424,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019482452215359463,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22487399131059646,
      "num_tokens": 137968050.0,
      "step": 79550
    },
    {
      "entropy": 5.96699070930481,
      "epoch": 6.189612915775141,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00019479862647837665,
      "loss": 4.9085,
      "mean_token_accuracy": 0.22640119045972823,
      "num_tokens": 137976162.0,
      "step": 79555
    },
    {
      "entropy": 5.939764738082886,
      "epoch": 6.190001945146858,
      "grad_norm": 1.703125,
      "learning_rate": 0.00019477273202008678,
      "loss": 4.9292,
      "mean_token_accuracy": 0.22290742993354798,
      "num_tokens": 137985045.0,
      "step": 79560
    },
    {
      "entropy": 5.958351373672485,
      "epoch": 6.190390974518576,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019474683877911808,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2167590156197548,
      "num_tokens": 137993715.0,
      "step": 79565
    },
    {
      "entropy": 6.026981258392334,
      "epoch": 6.190780003890294,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019472094675586333,
      "loss": 5.0338,
      "mean_token_accuracy": 0.21942880153656005,
      "num_tokens": 138002280.0,
      "step": 79570
    },
    {
      "entropy": 6.036942005157471,
      "epoch": 6.191169033262011,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001946950559507154,
      "loss": 4.9291,
      "mean_token_accuracy": 0.2160528063774109,
      "num_tokens": 138011027.0,
      "step": 79575
    },
    {
      "entropy": 6.005313396453857,
      "epoch": 6.191558062633729,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019466916636406718,
      "loss": 4.988,
      "mean_token_accuracy": 0.2130245327949524,
      "num_tokens": 138019149.0,
      "step": 79580
    },
    {
      "entropy": 6.048730516433716,
      "epoch": 6.191947092005447,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019464327799631153,
      "loss": 5.0152,
      "mean_token_accuracy": 0.21961454749107362,
      "num_tokens": 138027551.0,
      "step": 79585
    },
    {
      "entropy": 5.9462381362915036,
      "epoch": 6.192336121377164,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019461739084784124,
      "loss": 4.9831,
      "mean_token_accuracy": 0.21267153024673463,
      "num_tokens": 138036368.0,
      "step": 79590
    },
    {
      "entropy": 5.952931356430054,
      "epoch": 6.192725150748881,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001945915049190491,
      "loss": 4.9439,
      "mean_token_accuracy": 0.2210552379488945,
      "num_tokens": 138044934.0,
      "step": 79595
    },
    {
      "entropy": 5.980203628540039,
      "epoch": 6.193114180120599,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019456562021032787,
      "loss": 4.896,
      "mean_token_accuracy": 0.22648950219154357,
      "num_tokens": 138053030.0,
      "step": 79600
    },
    {
      "entropy": 5.913761854171753,
      "epoch": 6.193503209492317,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019453973672207037,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22770137339830399,
      "num_tokens": 138062188.0,
      "step": 79605
    },
    {
      "entropy": 5.960789585113526,
      "epoch": 6.193892238864034,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001945138544546693,
      "loss": 4.8961,
      "mean_token_accuracy": 0.22554732114076614,
      "num_tokens": 138071042.0,
      "step": 79610
    },
    {
      "entropy": 5.956880140304565,
      "epoch": 6.194281268235752,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001944879734085173,
      "loss": 4.9198,
      "mean_token_accuracy": 0.2234371393918991,
      "num_tokens": 138079370.0,
      "step": 79615
    },
    {
      "entropy": 6.0080560684204105,
      "epoch": 6.19467029760747,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00019446209358400723,
      "loss": 4.9544,
      "mean_token_accuracy": 0.2169836699962616,
      "num_tokens": 138087758.0,
      "step": 79620
    },
    {
      "entropy": 5.9419698238372805,
      "epoch": 6.195059326979187,
      "grad_norm": 1.625,
      "learning_rate": 0.0001944362149815317,
      "loss": 4.9461,
      "mean_token_accuracy": 0.22101276665925979,
      "num_tokens": 138096001.0,
      "step": 79625
    },
    {
      "entropy": 5.989284372329712,
      "epoch": 6.195448356350904,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019441033760148337,
      "loss": 5.0095,
      "mean_token_accuracy": 0.21683213412761687,
      "num_tokens": 138105003.0,
      "step": 79630
    },
    {
      "entropy": 5.981075239181519,
      "epoch": 6.195837385722622,
      "grad_norm": 1.5,
      "learning_rate": 0.00019438446144425493,
      "loss": 4.9097,
      "mean_token_accuracy": 0.21443400382995606,
      "num_tokens": 138113097.0,
      "step": 79635
    },
    {
      "entropy": 6.066406774520874,
      "epoch": 6.19622641509434,
      "grad_norm": 1.65625,
      "learning_rate": 0.00019435858651023892,
      "loss": 5.0188,
      "mean_token_accuracy": 0.21824993640184404,
      "num_tokens": 138122305.0,
      "step": 79640
    },
    {
      "entropy": 5.899138164520264,
      "epoch": 6.196615444466057,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00019433271279982795,
      "loss": 4.8582,
      "mean_token_accuracy": 0.23093915730714798,
      "num_tokens": 138130633.0,
      "step": 79645
    },
    {
      "entropy": 5.945448589324951,
      "epoch": 6.197004473837775,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001943068403134147,
      "loss": 4.9976,
      "mean_token_accuracy": 0.21781107485294343,
      "num_tokens": 138139166.0,
      "step": 79650
    },
    {
      "entropy": 5.995903253555298,
      "epoch": 6.197393503209493,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019428096905139175,
      "loss": 4.9673,
      "mean_token_accuracy": 0.22360957860946656,
      "num_tokens": 138148072.0,
      "step": 79655
    },
    {
      "entropy": 6.009780216217041,
      "epoch": 6.1977825325812095,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019425509901415162,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2153704881668091,
      "num_tokens": 138155859.0,
      "step": 79660
    },
    {
      "entropy": 6.040027141571045,
      "epoch": 6.198171561952927,
      "grad_norm": 1.6875,
      "learning_rate": 0.00019422923020208688,
      "loss": 4.9844,
      "mean_token_accuracy": 0.20676391273736955,
      "num_tokens": 138164570.0,
      "step": 79665
    },
    {
      "entropy": 5.976370477676392,
      "epoch": 6.198560591324645,
      "grad_norm": 1.625,
      "learning_rate": 0.00019420336261558995,
      "loss": 4.8973,
      "mean_token_accuracy": 0.22038774192333221,
      "num_tokens": 138172794.0,
      "step": 79670
    },
    {
      "entropy": 5.938037395477295,
      "epoch": 6.198949620696363,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019417749625505343,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22565864026546478,
      "num_tokens": 138182016.0,
      "step": 79675
    },
    {
      "entropy": 5.980306005477905,
      "epoch": 6.19933865006808,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019415163112086975,
      "loss": 5.0,
      "mean_token_accuracy": 0.21616332083940507,
      "num_tokens": 138191298.0,
      "step": 79680
    },
    {
      "entropy": 5.939406251907348,
      "epoch": 6.199727679439798,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019412576721343134,
      "loss": 4.893,
      "mean_token_accuracy": 0.2198581576347351,
      "num_tokens": 138200253.0,
      "step": 79685
    },
    {
      "entropy": 5.98449182510376,
      "epoch": 6.200116708811516,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019409990453313076,
      "loss": 4.934,
      "mean_token_accuracy": 0.22124186009168625,
      "num_tokens": 138209406.0,
      "step": 79690
    },
    {
      "entropy": 5.97465181350708,
      "epoch": 6.2005057381832325,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001940740430803603,
      "loss": 4.9082,
      "mean_token_accuracy": 0.22218364477157593,
      "num_tokens": 138218486.0,
      "step": 79695
    },
    {
      "entropy": 5.937479734420776,
      "epoch": 6.20089476755495,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019404818285551252,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2345029518008232,
      "num_tokens": 138226734.0,
      "step": 79700
    },
    {
      "entropy": 5.994907379150391,
      "epoch": 6.201283796926668,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0001940223238589796,
      "loss": 4.949,
      "mean_token_accuracy": 0.22809687554836272,
      "num_tokens": 138235290.0,
      "step": 79705
    },
    {
      "entropy": 5.926184797286988,
      "epoch": 6.201672826298386,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019399646609115408,
      "loss": 4.855,
      "mean_token_accuracy": 0.23414897173643112,
      "num_tokens": 138243664.0,
      "step": 79710
    },
    {
      "entropy": 5.970319128036499,
      "epoch": 6.202061855670103,
      "grad_norm": 1.671875,
      "learning_rate": 0.00019397060955242824,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21271145939826966,
      "num_tokens": 138252450.0,
      "step": 79715
    },
    {
      "entropy": 5.977016592025757,
      "epoch": 6.202450885041821,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019394475424319442,
      "loss": 5.0397,
      "mean_token_accuracy": 0.21145578920841218,
      "num_tokens": 138260719.0,
      "step": 79720
    },
    {
      "entropy": 6.0019426345825195,
      "epoch": 6.202839914413538,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019391890016384495,
      "loss": 4.9199,
      "mean_token_accuracy": 0.22061970233917236,
      "num_tokens": 138268831.0,
      "step": 79725
    },
    {
      "entropy": 5.960845804214477,
      "epoch": 6.2032289437852555,
      "grad_norm": 1.75,
      "learning_rate": 0.00019389304731477213,
      "loss": 4.9235,
      "mean_token_accuracy": 0.2217656269669533,
      "num_tokens": 138276899.0,
      "step": 79730
    },
    {
      "entropy": 5.985551357269287,
      "epoch": 6.203617973156973,
      "grad_norm": 1.703125,
      "learning_rate": 0.00019386719569636823,
      "loss": 4.9721,
      "mean_token_accuracy": 0.22180745899677276,
      "num_tokens": 138284962.0,
      "step": 79735
    },
    {
      "entropy": 6.063968992233276,
      "epoch": 6.204007002528691,
      "grad_norm": 1.6875,
      "learning_rate": 0.00019384134530902553,
      "loss": 5.0379,
      "mean_token_accuracy": 0.21763056367635727,
      "num_tokens": 138293711.0,
      "step": 79740
    },
    {
      "entropy": 5.963944721221924,
      "epoch": 6.204396031900409,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001938154961531362,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22805121839046477,
      "num_tokens": 138302032.0,
      "step": 79745
    },
    {
      "entropy": 5.941394996643067,
      "epoch": 6.204785061272126,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001937896482290925,
      "loss": 4.893,
      "mean_token_accuracy": 0.22244299203157425,
      "num_tokens": 138310301.0,
      "step": 79750
    },
    {
      "entropy": 5.97747163772583,
      "epoch": 6.205174090643844,
      "grad_norm": 1.625,
      "learning_rate": 0.00019376380153728667,
      "loss": 4.9389,
      "mean_token_accuracy": 0.22543756514787675,
      "num_tokens": 138318964.0,
      "step": 79755
    },
    {
      "entropy": 5.994526672363281,
      "epoch": 6.205563120015561,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019373795607811078,
      "loss": 4.9023,
      "mean_token_accuracy": 0.2275601953268051,
      "num_tokens": 138327285.0,
      "step": 79760
    },
    {
      "entropy": 6.001405334472656,
      "epoch": 6.2059521493872785,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019371211185195708,
      "loss": 4.8537,
      "mean_token_accuracy": 0.2228407859802246,
      "num_tokens": 138336088.0,
      "step": 79765
    },
    {
      "entropy": 6.035896921157837,
      "epoch": 6.206341178758996,
      "grad_norm": 1.625,
      "learning_rate": 0.0001936862688592177,
      "loss": 4.9358,
      "mean_token_accuracy": 0.22250894010066985,
      "num_tokens": 138344493.0,
      "step": 79770
    },
    {
      "entropy": 5.869189262390137,
      "epoch": 6.206730208130714,
      "grad_norm": 1.5,
      "learning_rate": 0.0001936604271002848,
      "loss": 4.8491,
      "mean_token_accuracy": 0.2243872806429863,
      "num_tokens": 138352931.0,
      "step": 79775
    },
    {
      "entropy": 5.9611390113830565,
      "epoch": 6.2071192375024316,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019363458657555038,
      "loss": 4.9021,
      "mean_token_accuracy": 0.21720437854528427,
      "num_tokens": 138361972.0,
      "step": 79780
    },
    {
      "entropy": 5.913915681838989,
      "epoch": 6.207508266874149,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019360874728540668,
      "loss": 4.8797,
      "mean_token_accuracy": 0.2219817414879799,
      "num_tokens": 138370941.0,
      "step": 79785
    },
    {
      "entropy": 5.9492237091064455,
      "epoch": 6.207897296245867,
      "grad_norm": 1.5,
      "learning_rate": 0.00019358290923024562,
      "loss": 4.9968,
      "mean_token_accuracy": 0.2170637547969818,
      "num_tokens": 138380185.0,
      "step": 79790
    },
    {
      "entropy": 6.0023480415344235,
      "epoch": 6.208286325617584,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001935570724104594,
      "loss": 4.9814,
      "mean_token_accuracy": 0.2154964402318001,
      "num_tokens": 138389790.0,
      "step": 79795
    },
    {
      "entropy": 6.0200567722320555,
      "epoch": 6.2086753549893015,
      "grad_norm": 1.5,
      "learning_rate": 0.00019353123682643998,
      "loss": 4.9216,
      "mean_token_accuracy": 0.22072939127683638,
      "num_tokens": 138399066.0,
      "step": 79800
    },
    {
      "entropy": 5.995497608184815,
      "epoch": 6.209064384361019,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019350540247857928,
      "loss": 4.9589,
      "mean_token_accuracy": 0.21752391904592513,
      "num_tokens": 138408014.0,
      "step": 79805
    },
    {
      "entropy": 6.018437814712525,
      "epoch": 6.209453413732737,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019347956936726946,
      "loss": 5.0329,
      "mean_token_accuracy": 0.2173070952296257,
      "num_tokens": 138416996.0,
      "step": 79810
    },
    {
      "entropy": 5.9925621509552,
      "epoch": 6.2098424431044545,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001934537374929024,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21811287403106688,
      "num_tokens": 138425496.0,
      "step": 79815
    },
    {
      "entropy": 5.984643411636353,
      "epoch": 6.210231472476172,
      "grad_norm": 1.625,
      "learning_rate": 0.00019342790685587008,
      "loss": 4.8766,
      "mean_token_accuracy": 0.2224881500005722,
      "num_tokens": 138433970.0,
      "step": 79820
    },
    {
      "entropy": 5.972914552688598,
      "epoch": 6.21062050184789,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001934020774565645,
      "loss": 4.8714,
      "mean_token_accuracy": 0.22625983208417894,
      "num_tokens": 138442703.0,
      "step": 79825
    },
    {
      "entropy": 5.9512190341949465,
      "epoch": 6.211009531219607,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001933762492953775,
      "loss": 4.905,
      "mean_token_accuracy": 0.21802688837051393,
      "num_tokens": 138451263.0,
      "step": 79830
    },
    {
      "entropy": 6.008898305892944,
      "epoch": 6.2113985605913244,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019335042237270095,
      "loss": 4.9189,
      "mean_token_accuracy": 0.22262635678052903,
      "num_tokens": 138459154.0,
      "step": 79835
    },
    {
      "entropy": 6.017796039581299,
      "epoch": 6.211787589963042,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019332459668892676,
      "loss": 4.9379,
      "mean_token_accuracy": 0.22495976984500884,
      "num_tokens": 138467923.0,
      "step": 79840
    },
    {
      "entropy": 5.957146835327149,
      "epoch": 6.21217661933476,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019329877224444675,
      "loss": 4.9034,
      "mean_token_accuracy": 0.22493798285722733,
      "num_tokens": 138476453.0,
      "step": 79845
    },
    {
      "entropy": 5.981274223327636,
      "epoch": 6.2125656487064775,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001932729490396529,
      "loss": 4.9056,
      "mean_token_accuracy": 0.2251599356532097,
      "num_tokens": 138485680.0,
      "step": 79850
    },
    {
      "entropy": 5.973556089401245,
      "epoch": 6.212954678078195,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019324712707493707,
      "loss": 4.93,
      "mean_token_accuracy": 0.2147836282849312,
      "num_tokens": 138494448.0,
      "step": 79855
    },
    {
      "entropy": 5.9563719749450685,
      "epoch": 6.213343707449912,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0001932213063506909,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2253995269536972,
      "num_tokens": 138502518.0,
      "step": 79860
    },
    {
      "entropy": 5.91676344871521,
      "epoch": 6.21373273682163,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019319548686730625,
      "loss": 4.9201,
      "mean_token_accuracy": 0.21966246962547303,
      "num_tokens": 138511218.0,
      "step": 79865
    },
    {
      "entropy": 5.964319133758545,
      "epoch": 6.214121766193347,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019316966862517482,
      "loss": 4.9401,
      "mean_token_accuracy": 0.21427278518676757,
      "num_tokens": 138520030.0,
      "step": 79870
    },
    {
      "entropy": 5.997924566268921,
      "epoch": 6.214510795565065,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001931438516246885,
      "loss": 4.9269,
      "mean_token_accuracy": 0.22006016224622726,
      "num_tokens": 138529130.0,
      "step": 79875
    },
    {
      "entropy": 5.99649863243103,
      "epoch": 6.214899824936783,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001931180358662389,
      "loss": 4.9421,
      "mean_token_accuracy": 0.22425933182239532,
      "num_tokens": 138537881.0,
      "step": 79880
    },
    {
      "entropy": 5.971860551834107,
      "epoch": 6.2152888543085005,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00019309222135021776,
      "loss": 4.8815,
      "mean_token_accuracy": 0.23005254119634627,
      "num_tokens": 138546176.0,
      "step": 79885
    },
    {
      "entropy": 5.931798315048217,
      "epoch": 6.215677883680218,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001930664080770168,
      "loss": 4.8342,
      "mean_token_accuracy": 0.23020280450582503,
      "num_tokens": 138554918.0,
      "step": 79890
    },
    {
      "entropy": 5.911038160324097,
      "epoch": 6.216066913051935,
      "grad_norm": 1.5,
      "learning_rate": 0.00019304059604702772,
      "loss": 4.861,
      "mean_token_accuracy": 0.22534206211566926,
      "num_tokens": 138564022.0,
      "step": 79895
    },
    {
      "entropy": 5.974187183380127,
      "epoch": 6.216455942423653,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019301478526064213,
      "loss": 4.9754,
      "mean_token_accuracy": 0.2210260733962059,
      "num_tokens": 138573009.0,
      "step": 79900
    },
    {
      "entropy": 5.9545948028564455,
      "epoch": 6.21684497179537,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019298897571825175,
      "loss": 4.9668,
      "mean_token_accuracy": 0.22072848528623581,
      "num_tokens": 138580910.0,
      "step": 79905
    },
    {
      "entropy": 5.961371278762817,
      "epoch": 6.217234001167088,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019296316742024806,
      "loss": 4.9284,
      "mean_token_accuracy": 0.2203731968998909,
      "num_tokens": 138589434.0,
      "step": 79910
    },
    {
      "entropy": 5.929994678497314,
      "epoch": 6.217623030538806,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019293736036702259,
      "loss": 4.8536,
      "mean_token_accuracy": 0.21923349499702455,
      "num_tokens": 138598173.0,
      "step": 79915
    },
    {
      "entropy": 5.990842628479004,
      "epoch": 6.2180120599105235,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001929115545589672,
      "loss": 4.9078,
      "mean_token_accuracy": 0.2217731148004532,
      "num_tokens": 138607244.0,
      "step": 79920
    },
    {
      "entropy": 5.92891411781311,
      "epoch": 6.218401089282241,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001928857499964733,
      "loss": 4.87,
      "mean_token_accuracy": 0.22926084697246552,
      "num_tokens": 138616166.0,
      "step": 79925
    },
    {
      "entropy": 5.956151485443115,
      "epoch": 6.218790118653958,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019285994667993245,
      "loss": 4.9304,
      "mean_token_accuracy": 0.2282259300351143,
      "num_tokens": 138624792.0,
      "step": 79930
    },
    {
      "entropy": 6.0048271179199215,
      "epoch": 6.219179148025676,
      "grad_norm": 1.625,
      "learning_rate": 0.00019283414460973626,
      "loss": 4.9544,
      "mean_token_accuracy": 0.21998497992753982,
      "num_tokens": 138633646.0,
      "step": 79935
    },
    {
      "entropy": 5.921176433563232,
      "epoch": 6.219568177397393,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019280834378627605,
      "loss": 4.8089,
      "mean_token_accuracy": 0.22570632994174958,
      "num_tokens": 138641924.0,
      "step": 79940
    },
    {
      "entropy": 5.883162784576416,
      "epoch": 6.219957206769111,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001927825442099434,
      "loss": 4.7859,
      "mean_token_accuracy": 0.23931537717580795,
      "num_tokens": 138650482.0,
      "step": 79945
    },
    {
      "entropy": 5.92144284248352,
      "epoch": 6.220346236140829,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001927567458811298,
      "loss": 4.8627,
      "mean_token_accuracy": 0.22535786628723145,
      "num_tokens": 138658805.0,
      "step": 79950
    },
    {
      "entropy": 5.980515575408935,
      "epoch": 6.2207352655125465,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019273094880022662,
      "loss": 5.009,
      "mean_token_accuracy": 0.21804411262273787,
      "num_tokens": 138667888.0,
      "step": 79955
    },
    {
      "entropy": 5.942375612258911,
      "epoch": 6.221124294884264,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019270515296762543,
      "loss": 4.8333,
      "mean_token_accuracy": 0.22584726363420488,
      "num_tokens": 138676068.0,
      "step": 79960
    },
    {
      "entropy": 6.003423976898193,
      "epoch": 6.221513324255981,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019267935838371752,
      "loss": 5.004,
      "mean_token_accuracy": 0.2144085004925728,
      "num_tokens": 138685041.0,
      "step": 79965
    },
    {
      "entropy": 5.937396287918091,
      "epoch": 6.221902353627699,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001926535650488943,
      "loss": 4.8636,
      "mean_token_accuracy": 0.22527590095996858,
      "num_tokens": 138693816.0,
      "step": 79970
    },
    {
      "entropy": 5.927091598510742,
      "epoch": 6.222291382999416,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001926277729635472,
      "loss": 4.8673,
      "mean_token_accuracy": 0.2290528193116188,
      "num_tokens": 138702765.0,
      "step": 79975
    },
    {
      "entropy": 5.927750873565674,
      "epoch": 6.222680412371134,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019260198212806747,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22678946405649186,
      "num_tokens": 138711679.0,
      "step": 79980
    },
    {
      "entropy": 5.9225057601928714,
      "epoch": 6.223069441742852,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001925761925428466,
      "loss": 4.8485,
      "mean_token_accuracy": 0.22968191057443618,
      "num_tokens": 138719781.0,
      "step": 79985
    },
    {
      "entropy": 5.98445234298706,
      "epoch": 6.2234584711145695,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019255040420827576,
      "loss": 4.9757,
      "mean_token_accuracy": 0.21483351290225983,
      "num_tokens": 138728399.0,
      "step": 79990
    },
    {
      "entropy": 5.985298538208008,
      "epoch": 6.223847500486286,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019252461712474629,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2170774444937706,
      "num_tokens": 138737481.0,
      "step": 79995
    },
    {
      "entropy": 6.078380489349366,
      "epoch": 6.224236529858004,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001924988312926495,
      "loss": 5.0777,
      "mean_token_accuracy": 0.213588710129261,
      "num_tokens": 138746453.0,
      "step": 80000
    },
    {
      "entropy": 6.001856470108033,
      "epoch": 6.224625559229722,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019247304671237664,
      "loss": 4.9607,
      "mean_token_accuracy": 0.2262277141213417,
      "num_tokens": 138754676.0,
      "step": 80005
    },
    {
      "entropy": 5.945106124877929,
      "epoch": 6.225014588601439,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00019244726338431896,
      "loss": 4.8626,
      "mean_token_accuracy": 0.2203497916460037,
      "num_tokens": 138763403.0,
      "step": 80010
    },
    {
      "entropy": 6.00696291923523,
      "epoch": 6.225403617973157,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019242148130886773,
      "loss": 4.9261,
      "mean_token_accuracy": 0.22686820477247238,
      "num_tokens": 138772597.0,
      "step": 80015
    },
    {
      "entropy": 5.9677405834198,
      "epoch": 6.225792647344875,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019239570048641398,
      "loss": 4.8874,
      "mean_token_accuracy": 0.22184973061084748,
      "num_tokens": 138781364.0,
      "step": 80020
    },
    {
      "entropy": 6.067014694213867,
      "epoch": 6.2261816767165925,
      "grad_norm": 1.5625,
      "learning_rate": 0.000192369920917349,
      "loss": 5.0475,
      "mean_token_accuracy": 0.21243318319320678,
      "num_tokens": 138789720.0,
      "step": 80025
    },
    {
      "entropy": 5.969296264648437,
      "epoch": 6.226570706088309,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000192344142602064,
      "loss": 4.939,
      "mean_token_accuracy": 0.22091440707445145,
      "num_tokens": 138798582.0,
      "step": 80030
    },
    {
      "entropy": 5.941404676437378,
      "epoch": 6.226959735460027,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019231836554095,
      "loss": 4.8986,
      "mean_token_accuracy": 0.22727575153112411,
      "num_tokens": 138807580.0,
      "step": 80035
    },
    {
      "entropy": 5.884705495834351,
      "epoch": 6.227348764831745,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019229258973439827,
      "loss": 4.8417,
      "mean_token_accuracy": 0.22424816936254502,
      "num_tokens": 138815988.0,
      "step": 80040
    },
    {
      "entropy": 5.989632225036621,
      "epoch": 6.227737794203462,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019226681518279975,
      "loss": 4.905,
      "mean_token_accuracy": 0.2193565160036087,
      "num_tokens": 138824049.0,
      "step": 80045
    },
    {
      "entropy": 5.9522132396698,
      "epoch": 6.22812682357518,
      "grad_norm": 1.65625,
      "learning_rate": 0.00019224104188654572,
      "loss": 4.8104,
      "mean_token_accuracy": 0.23222687542438508,
      "num_tokens": 138833219.0,
      "step": 80050
    },
    {
      "entropy": 5.950984334945678,
      "epoch": 6.228515852946898,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019221526984602717,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22631106823682784,
      "num_tokens": 138841398.0,
      "step": 80055
    },
    {
      "entropy": 5.979631519317627,
      "epoch": 6.228904882318615,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001921894990616351,
      "loss": 4.9506,
      "mean_token_accuracy": 0.21904913634061812,
      "num_tokens": 138849711.0,
      "step": 80060
    },
    {
      "entropy": 5.924089050292968,
      "epoch": 6.229293911690332,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019216372953376055,
      "loss": 4.8048,
      "mean_token_accuracy": 0.23367371559143066,
      "num_tokens": 138858317.0,
      "step": 80065
    },
    {
      "entropy": 5.956614208221436,
      "epoch": 6.22968294106205,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001921379612627946,
      "loss": 5.0095,
      "mean_token_accuracy": 0.21113065630197525,
      "num_tokens": 138866506.0,
      "step": 80070
    },
    {
      "entropy": 5.952702760696411,
      "epoch": 6.230071970433768,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019211219424912818,
      "loss": 4.9314,
      "mean_token_accuracy": 0.22341269701719285,
      "num_tokens": 138875458.0,
      "step": 80075
    },
    {
      "entropy": 6.024172163009643,
      "epoch": 6.230460999805485,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001920864284931523,
      "loss": 5.0031,
      "mean_token_accuracy": 0.2115068182349205,
      "num_tokens": 138884323.0,
      "step": 80080
    },
    {
      "entropy": 5.9845531463623045,
      "epoch": 6.230850029177203,
      "grad_norm": 1.625,
      "learning_rate": 0.00019206066399525784,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22973206490278245,
      "num_tokens": 138892758.0,
      "step": 80085
    },
    {
      "entropy": 6.008491659164429,
      "epoch": 6.231239058548921,
      "grad_norm": 1.703125,
      "learning_rate": 0.00019203490075583586,
      "loss": 4.9475,
      "mean_token_accuracy": 0.226315413415432,
      "num_tokens": 138901045.0,
      "step": 80090
    },
    {
      "entropy": 5.997810649871826,
      "epoch": 6.231628087920638,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001920091387752772,
      "loss": 4.9833,
      "mean_token_accuracy": 0.21520777195692062,
      "num_tokens": 138909677.0,
      "step": 80095
    },
    {
      "entropy": 5.938766765594482,
      "epoch": 6.232017117292355,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019198337805397282,
      "loss": 4.8538,
      "mean_token_accuracy": 0.2283240482211113,
      "num_tokens": 138918156.0,
      "step": 80100
    },
    {
      "entropy": 6.036249589920044,
      "epoch": 6.232406146664073,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019195761859231342,
      "loss": 5.0012,
      "mean_token_accuracy": 0.21367700546979904,
      "num_tokens": 138926157.0,
      "step": 80105
    },
    {
      "entropy": 5.9774253368377686,
      "epoch": 6.232795176035791,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019193186039068993,
      "loss": 5.0404,
      "mean_token_accuracy": 0.21074442714452743,
      "num_tokens": 138935240.0,
      "step": 80110
    },
    {
      "entropy": 5.975606393814087,
      "epoch": 6.233184205407508,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019190610344949337,
      "loss": 4.9877,
      "mean_token_accuracy": 0.22510451525449754,
      "num_tokens": 138944744.0,
      "step": 80115
    },
    {
      "entropy": 5.98780198097229,
      "epoch": 6.233573234779226,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019188034776911438,
      "loss": 4.887,
      "mean_token_accuracy": 0.22548603862524033,
      "num_tokens": 138952818.0,
      "step": 80120
    },
    {
      "entropy": 5.990399169921875,
      "epoch": 6.233962264150944,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019185459334994387,
      "loss": 4.9264,
      "mean_token_accuracy": 0.218938410282135,
      "num_tokens": 138961748.0,
      "step": 80125
    },
    {
      "entropy": 5.930001735687256,
      "epoch": 6.234351293522661,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00019182884019237258,
      "loss": 4.9408,
      "mean_token_accuracy": 0.2264801874756813,
      "num_tokens": 138970327.0,
      "step": 80130
    },
    {
      "entropy": 5.916275501251221,
      "epoch": 6.234740322894378,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001918030882967912,
      "loss": 4.9001,
      "mean_token_accuracy": 0.22582300007343292,
      "num_tokens": 138979061.0,
      "step": 80135
    },
    {
      "entropy": 6.026755905151367,
      "epoch": 6.235129352266096,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019177733766359057,
      "loss": 4.9544,
      "mean_token_accuracy": 0.2167467474937439,
      "num_tokens": 138987135.0,
      "step": 80140
    },
    {
      "entropy": 5.97624454498291,
      "epoch": 6.235518381637814,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019175158829316135,
      "loss": 4.9614,
      "mean_token_accuracy": 0.2231220245361328,
      "num_tokens": 138994950.0,
      "step": 80145
    },
    {
      "entropy": 5.983102035522461,
      "epoch": 6.235907411009531,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019172584018589429,
      "loss": 4.9291,
      "mean_token_accuracy": 0.21789480894804,
      "num_tokens": 139004442.0,
      "step": 80150
    },
    {
      "entropy": 5.980429697036743,
      "epoch": 6.236296440381249,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019170009334218004,
      "loss": 4.9237,
      "mean_token_accuracy": 0.23102206438779832,
      "num_tokens": 139012657.0,
      "step": 80155
    },
    {
      "entropy": 5.990257835388183,
      "epoch": 6.236685469752967,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019167434776240933,
      "loss": 4.9869,
      "mean_token_accuracy": 0.21335504055023194,
      "num_tokens": 139021365.0,
      "step": 80160
    },
    {
      "entropy": 5.975764274597168,
      "epoch": 6.237074499124684,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019164860344697272,
      "loss": 4.9609,
      "mean_token_accuracy": 0.2256344199180603,
      "num_tokens": 139030528.0,
      "step": 80165
    },
    {
      "entropy": 6.0508225440979,
      "epoch": 6.237463528496401,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019162286039626093,
      "loss": 4.9696,
      "mean_token_accuracy": 0.2183699756860733,
      "num_tokens": 139039098.0,
      "step": 80170
    },
    {
      "entropy": 5.984281206130982,
      "epoch": 6.237852557868119,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019159711861066453,
      "loss": 4.9512,
      "mean_token_accuracy": 0.2234419673681259,
      "num_tokens": 139047948.0,
      "step": 80175
    },
    {
      "entropy": 5.956870603561401,
      "epoch": 6.238241587239837,
      "grad_norm": 1.578125,
      "learning_rate": 0.000191571378090574,
      "loss": 4.9212,
      "mean_token_accuracy": 0.21947567611932756,
      "num_tokens": 139056561.0,
      "step": 80180
    },
    {
      "entropy": 5.94955415725708,
      "epoch": 6.238630616611554,
      "grad_norm": 1.609375,
      "learning_rate": 0.0001915456388363801,
      "loss": 4.9351,
      "mean_token_accuracy": 0.22104964256286622,
      "num_tokens": 139065567.0,
      "step": 80185
    },
    {
      "entropy": 6.012798261642456,
      "epoch": 6.239019645983272,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001915199008484733,
      "loss": 5.039,
      "mean_token_accuracy": 0.21409912705421447,
      "num_tokens": 139075249.0,
      "step": 80190
    },
    {
      "entropy": 5.93210277557373,
      "epoch": 6.239408675354989,
      "grad_norm": 1.671875,
      "learning_rate": 0.00019149416412724414,
      "loss": 4.9189,
      "mean_token_accuracy": 0.21581896096467973,
      "num_tokens": 139084226.0,
      "step": 80195
    },
    {
      "entropy": 5.897347450256348,
      "epoch": 6.239797704726707,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019146842867308306,
      "loss": 4.8354,
      "mean_token_accuracy": 0.23297619074583054,
      "num_tokens": 139093247.0,
      "step": 80200
    },
    {
      "entropy": 5.968873119354248,
      "epoch": 6.240186734098424,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019144269448638073,
      "loss": 4.9647,
      "mean_token_accuracy": 0.21599065661430358,
      "num_tokens": 139102771.0,
      "step": 80205
    },
    {
      "entropy": 5.982603549957275,
      "epoch": 6.240575763470142,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019141696156752748,
      "loss": 4.9077,
      "mean_token_accuracy": 0.22271136790513993,
      "num_tokens": 139111815.0,
      "step": 80210
    },
    {
      "entropy": 6.049552965164184,
      "epoch": 6.24096479284186,
      "grad_norm": 1.65625,
      "learning_rate": 0.00019139122991691378,
      "loss": 5.0083,
      "mean_token_accuracy": 0.21942714601755142,
      "num_tokens": 139119581.0,
      "step": 80215
    },
    {
      "entropy": 6.061219644546509,
      "epoch": 6.241353822213577,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0001913654995349301,
      "loss": 4.9665,
      "mean_token_accuracy": 0.2201542466878891,
      "num_tokens": 139128377.0,
      "step": 80220
    },
    {
      "entropy": 6.004542112350464,
      "epoch": 6.241742851585295,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019133977042196682,
      "loss": 4.9467,
      "mean_token_accuracy": 0.21861826181411742,
      "num_tokens": 139137158.0,
      "step": 80225
    },
    {
      "entropy": 5.936137771606445,
      "epoch": 6.242131880957012,
      "grad_norm": 1.5,
      "learning_rate": 0.0001913140425784144,
      "loss": 4.8454,
      "mean_token_accuracy": 0.22711904495954513,
      "num_tokens": 139145652.0,
      "step": 80230
    },
    {
      "entropy": 5.961682367324829,
      "epoch": 6.24252091032873,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019128831600466313,
      "loss": 4.9494,
      "mean_token_accuracy": 0.21885040402412415,
      "num_tokens": 139154670.0,
      "step": 80235
    },
    {
      "entropy": 5.932866764068604,
      "epoch": 6.242909939700447,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019126259070110347,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22698913514614105,
      "num_tokens": 139163892.0,
      "step": 80240
    },
    {
      "entropy": 5.950962066650391,
      "epoch": 6.243298969072165,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019123686666812562,
      "loss": 4.8775,
      "mean_token_accuracy": 0.22376986145973204,
      "num_tokens": 139173187.0,
      "step": 80245
    },
    {
      "entropy": 5.982334089279175,
      "epoch": 6.243687998443883,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001912111439061201,
      "loss": 4.9123,
      "mean_token_accuracy": 0.22526841461658478,
      "num_tokens": 139180970.0,
      "step": 80250
    },
    {
      "entropy": 6.029639530181885,
      "epoch": 6.2440770278156,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001911854224154771,
      "loss": 5.0085,
      "mean_token_accuracy": 0.21758198142051696,
      "num_tokens": 139189303.0,
      "step": 80255
    },
    {
      "entropy": 5.945736980438232,
      "epoch": 6.244466057187318,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0001911597021965869,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22480645030736923,
      "num_tokens": 139197953.0,
      "step": 80260
    },
    {
      "entropy": 5.980121898651123,
      "epoch": 6.244855086559035,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019113398324983972,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21194576621055602,
      "num_tokens": 139206544.0,
      "step": 80265
    },
    {
      "entropy": 5.956782865524292,
      "epoch": 6.245244115930753,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019110826557562594,
      "loss": 4.9353,
      "mean_token_accuracy": 0.21818137764930726,
      "num_tokens": 139215721.0,
      "step": 80270
    },
    {
      "entropy": 5.991478347778321,
      "epoch": 6.24563314530247,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019108254917433564,
      "loss": 4.9405,
      "mean_token_accuracy": 0.22441145032644272,
      "num_tokens": 139224441.0,
      "step": 80275
    },
    {
      "entropy": 6.020716667175293,
      "epoch": 6.246022174674188,
      "grad_norm": 1.640625,
      "learning_rate": 0.00019105683404635915,
      "loss": 4.9365,
      "mean_token_accuracy": 0.218800887465477,
      "num_tokens": 139232698.0,
      "step": 80280
    },
    {
      "entropy": 5.984579181671142,
      "epoch": 6.246411204045906,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019103112019208662,
      "loss": 4.9616,
      "mean_token_accuracy": 0.21505143493413925,
      "num_tokens": 139242190.0,
      "step": 80285
    },
    {
      "entropy": 5.966636753082275,
      "epoch": 6.246800233417623,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019100540761190815,
      "loss": 4.9271,
      "mean_token_accuracy": 0.2202674612402916,
      "num_tokens": 139251086.0,
      "step": 80290
    },
    {
      "entropy": 5.98963680267334,
      "epoch": 6.24718926278934,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019097969630621386,
      "loss": 4.8845,
      "mean_token_accuracy": 0.22730763107538224,
      "num_tokens": 139259874.0,
      "step": 80295
    },
    {
      "entropy": 5.995368528366089,
      "epoch": 6.247578292161058,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000190953986275394,
      "loss": 5.0256,
      "mean_token_accuracy": 0.21167052090167998,
      "num_tokens": 139268165.0,
      "step": 80300
    },
    {
      "entropy": 6.000151205062866,
      "epoch": 6.247967321532776,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00019092827751983866,
      "loss": 4.9962,
      "mean_token_accuracy": 0.2198484718799591,
      "num_tokens": 139277770.0,
      "step": 80305
    },
    {
      "entropy": 5.908852815628052,
      "epoch": 6.248356350904493,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001909025700399377,
      "loss": 4.7682,
      "mean_token_accuracy": 0.23624135702848434,
      "num_tokens": 139286586.0,
      "step": 80310
    },
    {
      "entropy": 5.981142663955689,
      "epoch": 6.248745380276211,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001908768638360816,
      "loss": 4.9225,
      "mean_token_accuracy": 0.21761636734008788,
      "num_tokens": 139294540.0,
      "step": 80315
    },
    {
      "entropy": 5.947405481338501,
      "epoch": 6.249134409647929,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00019085115890866018,
      "loss": 4.9059,
      "mean_token_accuracy": 0.23013843595981598,
      "num_tokens": 139302813.0,
      "step": 80320
    },
    {
      "entropy": 5.976039791107178,
      "epoch": 6.249523439019646,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019082545525806344,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21685190200805665,
      "num_tokens": 139311572.0,
      "step": 80325
    },
    {
      "entropy": 5.977466726303101,
      "epoch": 6.249912468391363,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019079975288468142,
      "loss": 4.9268,
      "mean_token_accuracy": 0.2255333036184311,
      "num_tokens": 139320653.0,
      "step": 80330
    },
    {
      "entropy": 6.0107275485992435,
      "epoch": 6.250301497763081,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019077405178890417,
      "loss": 4.9284,
      "mean_token_accuracy": 0.21743201613426208,
      "num_tokens": 139329581.0,
      "step": 80335
    },
    {
      "entropy": 6.120006847381592,
      "epoch": 6.250690527134799,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019074835197112162,
      "loss": 5.0895,
      "mean_token_accuracy": 0.20896613448858262,
      "num_tokens": 139337835.0,
      "step": 80340
    },
    {
      "entropy": 6.026638078689575,
      "epoch": 6.251079556506516,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019072265343172368,
      "loss": 5.0163,
      "mean_token_accuracy": 0.21295485645532608,
      "num_tokens": 139346967.0,
      "step": 80345
    },
    {
      "entropy": 5.976271724700927,
      "epoch": 6.251468585878234,
      "grad_norm": 1.71875,
      "learning_rate": 0.00019069695617110038,
      "loss": 4.9288,
      "mean_token_accuracy": 0.2200591653585434,
      "num_tokens": 139354980.0,
      "step": 80350
    },
    {
      "entropy": 5.974779033660889,
      "epoch": 6.251857615249952,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001906712601896416,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21681870818138121,
      "num_tokens": 139363618.0,
      "step": 80355
    },
    {
      "entropy": 5.963871240615845,
      "epoch": 6.252246644621669,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001906455654877372,
      "loss": 4.9323,
      "mean_token_accuracy": 0.22650104761123657,
      "num_tokens": 139372377.0,
      "step": 80360
    },
    {
      "entropy": 5.923278856277466,
      "epoch": 6.252635673993386,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019061987206577712,
      "loss": 4.7948,
      "mean_token_accuracy": 0.23266690522432326,
      "num_tokens": 139380842.0,
      "step": 80365
    },
    {
      "entropy": 5.877077865600586,
      "epoch": 6.253024703365104,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001905941799241511,
      "loss": 4.8105,
      "mean_token_accuracy": 0.23551322519779205,
      "num_tokens": 139390004.0,
      "step": 80370
    },
    {
      "entropy": 5.961658573150634,
      "epoch": 6.2534137327368216,
      "grad_norm": 1.71875,
      "learning_rate": 0.00019056848906324914,
      "loss": 4.8234,
      "mean_token_accuracy": 0.2260149300098419,
      "num_tokens": 139398059.0,
      "step": 80375
    },
    {
      "entropy": 5.864412641525268,
      "epoch": 6.253802762108539,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001905427994834608,
      "loss": 4.8377,
      "mean_token_accuracy": 0.234512260556221,
      "num_tokens": 139406499.0,
      "step": 80380
    },
    {
      "entropy": 5.982770729064941,
      "epoch": 6.254191791480257,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00019051711118517618,
      "loss": 4.9151,
      "mean_token_accuracy": 0.2246263101696968,
      "num_tokens": 139414629.0,
      "step": 80385
    },
    {
      "entropy": 5.924739837646484,
      "epoch": 6.254580820851975,
      "grad_norm": 1.5,
      "learning_rate": 0.00019049142416878496,
      "loss": 4.8823,
      "mean_token_accuracy": 0.21907055675983428,
      "num_tokens": 139424474.0,
      "step": 80390
    },
    {
      "entropy": 6.0668354511260985,
      "epoch": 6.2549698502236915,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019046573843467684,
      "loss": 4.9314,
      "mean_token_accuracy": 0.22173729091882705,
      "num_tokens": 139433452.0,
      "step": 80395
    },
    {
      "entropy": 5.959890079498291,
      "epoch": 6.255358879595409,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001904400539832416,
      "loss": 4.8635,
      "mean_token_accuracy": 0.22497277706861496,
      "num_tokens": 139442249.0,
      "step": 80400
    },
    {
      "entropy": 6.038123941421508,
      "epoch": 6.255747908967127,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019041437081486896,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21033526211977005,
      "num_tokens": 139451027.0,
      "step": 80405
    },
    {
      "entropy": 6.007887172698974,
      "epoch": 6.2561369383388445,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019038868892994854,
      "loss": 4.9795,
      "mean_token_accuracy": 0.21829184144735336,
      "num_tokens": 139459816.0,
      "step": 80410
    },
    {
      "entropy": 5.998119831085205,
      "epoch": 6.256525967710562,
      "grad_norm": 3.109375,
      "learning_rate": 0.00019036300832887015,
      "loss": 4.9079,
      "mean_token_accuracy": 0.22780579179525376,
      "num_tokens": 139468043.0,
      "step": 80415
    },
    {
      "entropy": 6.0031023025512695,
      "epoch": 6.25691499708228,
      "grad_norm": 1.640625,
      "learning_rate": 0.00019033732901202339,
      "loss": 4.9492,
      "mean_token_accuracy": 0.2211369425058365,
      "num_tokens": 139476561.0,
      "step": 80420
    },
    {
      "entropy": 5.965800476074219,
      "epoch": 6.257304026453998,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001903116509797979,
      "loss": 4.918,
      "mean_token_accuracy": 0.21829174011945723,
      "num_tokens": 139485179.0,
      "step": 80425
    },
    {
      "entropy": 5.984697723388672,
      "epoch": 6.2576930558257144,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001902859742325833,
      "loss": 4.9544,
      "mean_token_accuracy": 0.21962982714176177,
      "num_tokens": 139492962.0,
      "step": 80430
    },
    {
      "entropy": 5.966595983505249,
      "epoch": 6.258082085197432,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019026029877076923,
      "loss": 4.9543,
      "mean_token_accuracy": 0.21921565532684326,
      "num_tokens": 139502382.0,
      "step": 80435
    },
    {
      "entropy": 5.878050231933594,
      "epoch": 6.25847111456915,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001902346245947452,
      "loss": 4.8161,
      "mean_token_accuracy": 0.23269507735967637,
      "num_tokens": 139511693.0,
      "step": 80440
    },
    {
      "entropy": 6.020147180557251,
      "epoch": 6.2588601439408675,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019020895170490082,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21880081444978713,
      "num_tokens": 139520020.0,
      "step": 80445
    },
    {
      "entropy": 5.932745265960693,
      "epoch": 6.259249173312585,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019018328010162567,
      "loss": 4.9303,
      "mean_token_accuracy": 0.21606847196817397,
      "num_tokens": 139528291.0,
      "step": 80450
    },
    {
      "entropy": 5.948555517196655,
      "epoch": 6.259638202684303,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019015760978530926,
      "loss": 4.9502,
      "mean_token_accuracy": 0.21818622052669526,
      "num_tokens": 139536469.0,
      "step": 80455
    },
    {
      "entropy": 5.939505577087402,
      "epoch": 6.260027232056021,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019013194075634106,
      "loss": 4.8503,
      "mean_token_accuracy": 0.23321955353021623,
      "num_tokens": 139545212.0,
      "step": 80460
    },
    {
      "entropy": 5.975971126556397,
      "epoch": 6.260416261427737,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019010627301511057,
      "loss": 4.9351,
      "mean_token_accuracy": 0.21820525377988814,
      "num_tokens": 139553510.0,
      "step": 80465
    },
    {
      "entropy": 5.9626062393188475,
      "epoch": 6.260805290799455,
      "grad_norm": 1.5,
      "learning_rate": 0.00019008060656200732,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22280961871147156,
      "num_tokens": 139561553.0,
      "step": 80470
    },
    {
      "entropy": 6.006557321548462,
      "epoch": 6.261194320171173,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019005494139742068,
      "loss": 4.9933,
      "mean_token_accuracy": 0.21731957346200942,
      "num_tokens": 139571163.0,
      "step": 80475
    },
    {
      "entropy": 6.002400350570679,
      "epoch": 6.2615833495428905,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001900292775217401,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21575907766819,
      "num_tokens": 139579429.0,
      "step": 80480
    },
    {
      "entropy": 5.964996480941773,
      "epoch": 6.261972378914608,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000190003614935355,
      "loss": 4.9015,
      "mean_token_accuracy": 0.22534175962209702,
      "num_tokens": 139588851.0,
      "step": 80485
    },
    {
      "entropy": 5.954633092880249,
      "epoch": 6.262361408286326,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00018997795363865473,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22770313620567323,
      "num_tokens": 139597189.0,
      "step": 80490
    },
    {
      "entropy": 5.910151672363281,
      "epoch": 6.262750437658044,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0001899522936320287,
      "loss": 4.8283,
      "mean_token_accuracy": 0.22679278552532195,
      "num_tokens": 139605257.0,
      "step": 80495
    },
    {
      "entropy": 5.9266026496887205,
      "epoch": 6.26313946702976,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001899266349158662,
      "loss": 4.8948,
      "mean_token_accuracy": 0.2172493517398834,
      "num_tokens": 139613374.0,
      "step": 80500
    },
    {
      "entropy": 6.024798727035522,
      "epoch": 6.263528496401478,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00018990097749055669,
      "loss": 4.9969,
      "mean_token_accuracy": 0.2088245078921318,
      "num_tokens": 139622260.0,
      "step": 80505
    },
    {
      "entropy": 6.020614433288574,
      "epoch": 6.263917525773196,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00018987532135648926,
      "loss": 5.0032,
      "mean_token_accuracy": 0.21210717260837555,
      "num_tokens": 139630127.0,
      "step": 80510
    },
    {
      "entropy": 6.023227739334106,
      "epoch": 6.2643065551449135,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001898496665140535,
      "loss": 4.9775,
      "mean_token_accuracy": 0.2169090300798416,
      "num_tokens": 139638999.0,
      "step": 80515
    },
    {
      "entropy": 6.026350736618042,
      "epoch": 6.264695584516631,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018982401296363844,
      "loss": 4.9789,
      "mean_token_accuracy": 0.21807647347450257,
      "num_tokens": 139647100.0,
      "step": 80520
    },
    {
      "entropy": 5.9934406757354735,
      "epoch": 6.265084613888349,
      "grad_norm": 1.609375,
      "learning_rate": 0.00018979836070563339,
      "loss": 4.8933,
      "mean_token_accuracy": 0.22341385632753372,
      "num_tokens": 139655330.0,
      "step": 80525
    },
    {
      "entropy": 5.907823944091797,
      "epoch": 6.265473643260066,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00018977270974042764,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23568839877843856,
      "num_tokens": 139663702.0,
      "step": 80530
    },
    {
      "entropy": 5.956437063217163,
      "epoch": 6.265862672631783,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018974706006841035,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21976171135902406,
      "num_tokens": 139672302.0,
      "step": 80535
    },
    {
      "entropy": 5.925638008117676,
      "epoch": 6.266251702003501,
      "grad_norm": 1.734375,
      "learning_rate": 0.00018972141168997075,
      "loss": 4.8018,
      "mean_token_accuracy": 0.22518396973609925,
      "num_tokens": 139680881.0,
      "step": 80540
    },
    {
      "entropy": 5.976995658874512,
      "epoch": 6.266640731375219,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018969576460549798,
      "loss": 4.864,
      "mean_token_accuracy": 0.2299143761396408,
      "num_tokens": 139690134.0,
      "step": 80545
    },
    {
      "entropy": 5.9761217594146725,
      "epoch": 6.2670297607469365,
      "grad_norm": 1.703125,
      "learning_rate": 0.00018967011881538122,
      "loss": 4.9102,
      "mean_token_accuracy": 0.22536627650260926,
      "num_tokens": 139699066.0,
      "step": 80550
    },
    {
      "entropy": 5.988219261169434,
      "epoch": 6.267418790118654,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018964447432000964,
      "loss": 4.96,
      "mean_token_accuracy": 0.22141536921262742,
      "num_tokens": 139707884.0,
      "step": 80555
    },
    {
      "entropy": 5.999737787246704,
      "epoch": 6.267807819490372,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00018961883111977225,
      "loss": 4.9358,
      "mean_token_accuracy": 0.22934476435184478,
      "num_tokens": 139716848.0,
      "step": 80560
    },
    {
      "entropy": 5.919596385955811,
      "epoch": 6.268196848862089,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001895931892150582,
      "loss": 4.8115,
      "mean_token_accuracy": 0.23043763637542725,
      "num_tokens": 139725639.0,
      "step": 80565
    },
    {
      "entropy": 5.891748380661011,
      "epoch": 6.268585878233806,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018956754860625653,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2312275603413582,
      "num_tokens": 139734259.0,
      "step": 80570
    },
    {
      "entropy": 5.941840505599975,
      "epoch": 6.268974907605524,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018954190929375625,
      "loss": 4.9363,
      "mean_token_accuracy": 0.22420917004346846,
      "num_tokens": 139743049.0,
      "step": 80575
    },
    {
      "entropy": 5.969996690750122,
      "epoch": 6.269363936977242,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00018951627127794656,
      "loss": 4.9382,
      "mean_token_accuracy": 0.22064988017082215,
      "num_tokens": 139751686.0,
      "step": 80580
    },
    {
      "entropy": 5.991896772384644,
      "epoch": 6.2697529663489595,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001894906345592165,
      "loss": 4.9848,
      "mean_token_accuracy": 0.2174515664577484,
      "num_tokens": 139760268.0,
      "step": 80585
    },
    {
      "entropy": 5.95908522605896,
      "epoch": 6.270141995720677,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001894649991379549,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22710421681404114,
      "num_tokens": 139769081.0,
      "step": 80590
    },
    {
      "entropy": 6.003872632980347,
      "epoch": 6.270531025092394,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00018943936501455077,
      "loss": 5.0113,
      "mean_token_accuracy": 0.221414253115654,
      "num_tokens": 139777879.0,
      "step": 80595
    },
    {
      "entropy": 6.075345373153686,
      "epoch": 6.270920054464112,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018941373218939306,
      "loss": 5.0447,
      "mean_token_accuracy": 0.21136215776205064,
      "num_tokens": 139786105.0,
      "step": 80600
    },
    {
      "entropy": 6.046206998825073,
      "epoch": 6.271309083835829,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018938810066287078,
      "loss": 4.9539,
      "mean_token_accuracy": 0.2230125278234482,
      "num_tokens": 139795230.0,
      "step": 80605
    },
    {
      "entropy": 5.928977727890015,
      "epoch": 6.271698113207547,
      "grad_norm": 1.640625,
      "learning_rate": 0.0001893624704353728,
      "loss": 4.8199,
      "mean_token_accuracy": 0.22967091351747512,
      "num_tokens": 139803522.0,
      "step": 80610
    },
    {
      "entropy": 5.9459412574768065,
      "epoch": 6.272087142579265,
      "grad_norm": 1.59375,
      "learning_rate": 0.00018933684150728802,
      "loss": 4.8949,
      "mean_token_accuracy": 0.22807173430919647,
      "num_tokens": 139812550.0,
      "step": 80615
    },
    {
      "entropy": 5.943883752822876,
      "epoch": 6.2724761719509825,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018931121387900536,
      "loss": 4.9606,
      "mean_token_accuracy": 0.21931248605251313,
      "num_tokens": 139820725.0,
      "step": 80620
    },
    {
      "entropy": 5.896672296524048,
      "epoch": 6.2728652013227,
      "grad_norm": 1.6875,
      "learning_rate": 0.00018928558755091358,
      "loss": 4.8237,
      "mean_token_accuracy": 0.22586772739887237,
      "num_tokens": 139828754.0,
      "step": 80625
    },
    {
      "entropy": 5.97490930557251,
      "epoch": 6.273254230694418,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018925996252340167,
      "loss": 4.8992,
      "mean_token_accuracy": 0.22533331513404847,
      "num_tokens": 139837640.0,
      "step": 80630
    },
    {
      "entropy": 6.003220176696777,
      "epoch": 6.273643260066135,
      "grad_norm": 1.671875,
      "learning_rate": 0.0001892343387968583,
      "loss": 4.9563,
      "mean_token_accuracy": 0.22125212997198104,
      "num_tokens": 139846483.0,
      "step": 80635
    },
    {
      "entropy": 6.010891962051391,
      "epoch": 6.274032289437852,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018920871637167236,
      "loss": 5.0079,
      "mean_token_accuracy": 0.20958862751722335,
      "num_tokens": 139855174.0,
      "step": 80640
    },
    {
      "entropy": 5.8938984870910645,
      "epoch": 6.27442131880957,
      "grad_norm": 1.59375,
      "learning_rate": 0.00018918309524823247,
      "loss": 4.8153,
      "mean_token_accuracy": 0.224630144238472,
      "num_tokens": 139863583.0,
      "step": 80645
    },
    {
      "entropy": 5.942338323593139,
      "epoch": 6.274810348181288,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00018915747542692758,
      "loss": 4.928,
      "mean_token_accuracy": 0.2276465967297554,
      "num_tokens": 139871909.0,
      "step": 80650
    },
    {
      "entropy": 6.006707191467285,
      "epoch": 6.2751993775530055,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018913185690814637,
      "loss": 4.9748,
      "mean_token_accuracy": 0.22455118596553802,
      "num_tokens": 139880262.0,
      "step": 80655
    },
    {
      "entropy": 5.977741861343384,
      "epoch": 6.275588406924723,
      "grad_norm": 1.640625,
      "learning_rate": 0.00018910623969227764,
      "loss": 4.9528,
      "mean_token_accuracy": 0.2133171498775482,
      "num_tokens": 139889018.0,
      "step": 80660
    },
    {
      "entropy": 5.960037136077881,
      "epoch": 6.27597743629644,
      "grad_norm": 1.609375,
      "learning_rate": 0.00018908062377971,
      "loss": 4.8167,
      "mean_token_accuracy": 0.22979181110858918,
      "num_tokens": 139897230.0,
      "step": 80665
    },
    {
      "entropy": 5.977007913589477,
      "epoch": 6.276366465668158,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00018905500917083213,
      "loss": 4.9288,
      "mean_token_accuracy": 0.2252034530043602,
      "num_tokens": 139905797.0,
      "step": 80670
    },
    {
      "entropy": 6.011875200271606,
      "epoch": 6.276755495039875,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018902939586603262,
      "loss": 4.8859,
      "mean_token_accuracy": 0.22142704874277114,
      "num_tokens": 139914726.0,
      "step": 80675
    },
    {
      "entropy": 5.901184701919556,
      "epoch": 6.277144524411593,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00018900378386570027,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2379198804497719,
      "num_tokens": 139923217.0,
      "step": 80680
    },
    {
      "entropy": 5.874596977233887,
      "epoch": 6.277533553783311,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001889781731702236,
      "loss": 4.8686,
      "mean_token_accuracy": 0.22748982310295104,
      "num_tokens": 139931799.0,
      "step": 80685
    },
    {
      "entropy": 5.9368444919586185,
      "epoch": 6.2779225831550285,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00018895256377999121,
      "loss": 4.9304,
      "mean_token_accuracy": 0.21759516447782518,
      "num_tokens": 139940783.0,
      "step": 80690
    },
    {
      "entropy": 6.022189617156982,
      "epoch": 6.278311612526746,
      "grad_norm": 1.625,
      "learning_rate": 0.0001889269556953917,
      "loss": 4.8756,
      "mean_token_accuracy": 0.22712236046791076,
      "num_tokens": 139949263.0,
      "step": 80695
    },
    {
      "entropy": 5.921438884735108,
      "epoch": 6.278700641898463,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0001889013489168137,
      "loss": 4.8931,
      "mean_token_accuracy": 0.2327013149857521,
      "num_tokens": 139958012.0,
      "step": 80700
    },
    {
      "entropy": 5.920327234268188,
      "epoch": 6.279089671270181,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018887574344464563,
      "loss": 4.9046,
      "mean_token_accuracy": 0.22645654380321503,
      "num_tokens": 139966723.0,
      "step": 80705
    },
    {
      "entropy": 6.003917551040649,
      "epoch": 6.279478700641898,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018885013927927608,
      "loss": 4.9884,
      "mean_token_accuracy": 0.22453253269195556,
      "num_tokens": 139975496.0,
      "step": 80710
    },
    {
      "entropy": 6.0114177703857425,
      "epoch": 6.279867730013616,
      "grad_norm": 1.6875,
      "learning_rate": 0.00018882453642109353,
      "loss": 4.9379,
      "mean_token_accuracy": 0.2202129527926445,
      "num_tokens": 139983861.0,
      "step": 80715
    },
    {
      "entropy": 5.937944841384888,
      "epoch": 6.280256759385334,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001887989348704865,
      "loss": 4.8526,
      "mean_token_accuracy": 0.2275329202413559,
      "num_tokens": 139992565.0,
      "step": 80720
    },
    {
      "entropy": 6.00098180770874,
      "epoch": 6.2806457887570515,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018877333462784346,
      "loss": 4.955,
      "mean_token_accuracy": 0.21971245259046554,
      "num_tokens": 140001943.0,
      "step": 80725
    },
    {
      "entropy": 6.010072374343872,
      "epoch": 6.281034818128768,
      "grad_norm": 1.671875,
      "learning_rate": 0.0001887477356935528,
      "loss": 4.9867,
      "mean_token_accuracy": 0.21524942368268968,
      "num_tokens": 140010667.0,
      "step": 80730
    },
    {
      "entropy": 5.922651243209839,
      "epoch": 6.281423847500486,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018872213806800297,
      "loss": 4.8662,
      "mean_token_accuracy": 0.2303991883993149,
      "num_tokens": 140019725.0,
      "step": 80735
    },
    {
      "entropy": 5.9251086711883545,
      "epoch": 6.281812876872204,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018869654175158234,
      "loss": 4.8687,
      "mean_token_accuracy": 0.23534294217824936,
      "num_tokens": 140028789.0,
      "step": 80740
    },
    {
      "entropy": 5.931915616989135,
      "epoch": 6.282201906243921,
      "grad_norm": 1.59375,
      "learning_rate": 0.0001886709467446794,
      "loss": 4.894,
      "mean_token_accuracy": 0.2239112749695778,
      "num_tokens": 140037736.0,
      "step": 80745
    },
    {
      "entropy": 5.961803770065307,
      "epoch": 6.282590935615639,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018864535304768238,
      "loss": 4.9773,
      "mean_token_accuracy": 0.21855029165744783,
      "num_tokens": 140046120.0,
      "step": 80750
    },
    {
      "entropy": 5.97981367111206,
      "epoch": 6.282979964987357,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00018861976066097968,
      "loss": 4.9537,
      "mean_token_accuracy": 0.2174478828907013,
      "num_tokens": 140054963.0,
      "step": 80755
    },
    {
      "entropy": 5.95885443687439,
      "epoch": 6.2833689943590745,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018859416958495962,
      "loss": 4.9106,
      "mean_token_accuracy": 0.22218480408191682,
      "num_tokens": 140064215.0,
      "step": 80760
    },
    {
      "entropy": 5.943155956268311,
      "epoch": 6.283758023730791,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018856857982001051,
      "loss": 4.9024,
      "mean_token_accuracy": 0.2196154311299324,
      "num_tokens": 140074037.0,
      "step": 80765
    },
    {
      "entropy": 6.062546253204346,
      "epoch": 6.284147053102509,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001885429913665207,
      "loss": 5.0833,
      "mean_token_accuracy": 0.20941018760204316,
      "num_tokens": 140083277.0,
      "step": 80770
    },
    {
      "entropy": 6.068348836898804,
      "epoch": 6.284536082474227,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018851740422487829,
      "loss": 5.0386,
      "mean_token_accuracy": 0.21738290190696716,
      "num_tokens": 140093027.0,
      "step": 80775
    },
    {
      "entropy": 5.936559724807739,
      "epoch": 6.284925111845944,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018849181839547164,
      "loss": 4.8356,
      "mean_token_accuracy": 0.2305673837661743,
      "num_tokens": 140101154.0,
      "step": 80780
    },
    {
      "entropy": 6.027870178222656,
      "epoch": 6.285314141217662,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018846623387868894,
      "loss": 4.9984,
      "mean_token_accuracy": 0.21781247556209565,
      "num_tokens": 140109337.0,
      "step": 80785
    },
    {
      "entropy": 5.913628387451172,
      "epoch": 6.28570317058938,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018844065067491844,
      "loss": 4.8583,
      "mean_token_accuracy": 0.227262507379055,
      "num_tokens": 140117985.0,
      "step": 80790
    },
    {
      "entropy": 5.970162296295166,
      "epoch": 6.2860921999610975,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018841506878454828,
      "loss": 4.8678,
      "mean_token_accuracy": 0.2258630007505417,
      "num_tokens": 140126804.0,
      "step": 80795
    },
    {
      "entropy": 5.919177913665772,
      "epoch": 6.286481229332814,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00018838948820796664,
      "loss": 4.8507,
      "mean_token_accuracy": 0.23295949250459672,
      "num_tokens": 140136119.0,
      "step": 80800
    },
    {
      "entropy": 5.949384975433349,
      "epoch": 6.286870258704532,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018836390894556165,
      "loss": 4.8956,
      "mean_token_accuracy": 0.21734274178743362,
      "num_tokens": 140144501.0,
      "step": 80805
    },
    {
      "entropy": 5.932889604568482,
      "epoch": 6.28725928807625,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018833833099772147,
      "loss": 4.9016,
      "mean_token_accuracy": 0.2223481386899948,
      "num_tokens": 140152867.0,
      "step": 80810
    },
    {
      "entropy": 5.99184284210205,
      "epoch": 6.287648317447967,
      "grad_norm": 1.578125,
      "learning_rate": 0.00018831275436483418,
      "loss": 4.9711,
      "mean_token_accuracy": 0.2218160003423691,
      "num_tokens": 140161262.0,
      "step": 80815
    },
    {
      "entropy": 5.956709623336792,
      "epoch": 6.288037346819685,
      "grad_norm": 1.5,
      "learning_rate": 0.00018828717904728788,
      "loss": 4.848,
      "mean_token_accuracy": 0.23468939810991288,
      "num_tokens": 140170471.0,
      "step": 80820
    },
    {
      "entropy": 5.97047758102417,
      "epoch": 6.288426376191403,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00018826160504547063,
      "loss": 4.9328,
      "mean_token_accuracy": 0.22414567768573762,
      "num_tokens": 140179880.0,
      "step": 80825
    },
    {
      "entropy": 5.914414262771606,
      "epoch": 6.2888154055631205,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018823603235977043,
      "loss": 4.8364,
      "mean_token_accuracy": 0.22690450102090837,
      "num_tokens": 140188273.0,
      "step": 80830
    },
    {
      "entropy": 5.961648273468017,
      "epoch": 6.289204434934837,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018821046099057532,
      "loss": 4.9841,
      "mean_token_accuracy": 0.21643772423267366,
      "num_tokens": 140197522.0,
      "step": 80835
    },
    {
      "entropy": 5.940830087661743,
      "epoch": 6.289593464306555,
      "grad_norm": 1.578125,
      "learning_rate": 0.00018818489093827334,
      "loss": 4.8976,
      "mean_token_accuracy": 0.22867703139781953,
      "num_tokens": 140205998.0,
      "step": 80840
    },
    {
      "entropy": 5.998374032974243,
      "epoch": 6.289982493678273,
      "grad_norm": 1.671875,
      "learning_rate": 0.00018815932220325239,
      "loss": 5.006,
      "mean_token_accuracy": 0.21155470609664917,
      "num_tokens": 140214454.0,
      "step": 80845
    },
    {
      "entropy": 5.927531003952026,
      "epoch": 6.29037152304999,
      "grad_norm": 1.65625,
      "learning_rate": 0.0001881337547859006,
      "loss": 4.7898,
      "mean_token_accuracy": 0.23194326609373092,
      "num_tokens": 140222571.0,
      "step": 80850
    },
    {
      "entropy": 5.961789560317993,
      "epoch": 6.290760552421708,
      "grad_norm": 1.546875,
      "learning_rate": 0.00018810818868660585,
      "loss": 5.0013,
      "mean_token_accuracy": 0.21660682260990144,
      "num_tokens": 140231857.0,
      "step": 80855
    },
    {
      "entropy": 6.009832382202148,
      "epoch": 6.291149581793426,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018808262390575598,
      "loss": 4.9434,
      "mean_token_accuracy": 0.2226993039250374,
      "num_tokens": 140240351.0,
      "step": 80860
    },
    {
      "entropy": 5.9814859390258786,
      "epoch": 6.291538611165143,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018805706044373893,
      "loss": 4.911,
      "mean_token_accuracy": 0.22355422079563142,
      "num_tokens": 140248609.0,
      "step": 80865
    },
    {
      "entropy": 5.9399346828460695,
      "epoch": 6.29192764053686,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018803149830094268,
      "loss": 4.9285,
      "mean_token_accuracy": 0.22252807021141052,
      "num_tokens": 140257716.0,
      "step": 80870
    },
    {
      "entropy": 6.000695276260376,
      "epoch": 6.292316669908578,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00018800593747775492,
      "loss": 4.9356,
      "mean_token_accuracy": 0.22589959502220153,
      "num_tokens": 140265986.0,
      "step": 80875
    },
    {
      "entropy": 5.962469482421875,
      "epoch": 6.292705699280296,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018798037797456368,
      "loss": 4.8821,
      "mean_token_accuracy": 0.22734427601099014,
      "num_tokens": 140274633.0,
      "step": 80880
    },
    {
      "entropy": 5.894175481796265,
      "epoch": 6.293094728652013,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001879548197917566,
      "loss": 4.8859,
      "mean_token_accuracy": 0.225224170088768,
      "num_tokens": 140282962.0,
      "step": 80885
    },
    {
      "entropy": 5.9517113208770756,
      "epoch": 6.293483758023731,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00018792926292972168,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21829790771007537,
      "num_tokens": 140290945.0,
      "step": 80890
    },
    {
      "entropy": 5.998342800140381,
      "epoch": 6.293872787395449,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018790370738884653,
      "loss": 4.9283,
      "mean_token_accuracy": 0.21716422736644744,
      "num_tokens": 140299536.0,
      "step": 80895
    },
    {
      "entropy": 6.004630184173584,
      "epoch": 6.294261816767166,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018787815316951905,
      "loss": 4.9644,
      "mean_token_accuracy": 0.2159113645553589,
      "num_tokens": 140308792.0,
      "step": 80900
    },
    {
      "entropy": 5.984836721420288,
      "epoch": 6.294650846138883,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018785260027212692,
      "loss": 4.8762,
      "mean_token_accuracy": 0.22321363687515258,
      "num_tokens": 140317128.0,
      "step": 80905
    },
    {
      "entropy": 5.969964694976807,
      "epoch": 6.295039875510601,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001878270486970578,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22675697058439254,
      "num_tokens": 140326127.0,
      "step": 80910
    },
    {
      "entropy": 5.995491552352905,
      "epoch": 6.295428904882319,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018780149844469958,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21455106884241104,
      "num_tokens": 140335037.0,
      "step": 80915
    },
    {
      "entropy": 5.926424789428711,
      "epoch": 6.295817934254036,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00018777594951543974,
      "loss": 4.9324,
      "mean_token_accuracy": 0.2206193298101425,
      "num_tokens": 140343517.0,
      "step": 80920
    },
    {
      "entropy": 5.91227536201477,
      "epoch": 6.296206963625754,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018775040190966607,
      "loss": 4.9261,
      "mean_token_accuracy": 0.22226907759904863,
      "num_tokens": 140352211.0,
      "step": 80925
    },
    {
      "entropy": 5.988093471527099,
      "epoch": 6.296595992997471,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001877248556277662,
      "loss": 5.001,
      "mean_token_accuracy": 0.21614395529031755,
      "num_tokens": 140360833.0,
      "step": 80930
    },
    {
      "entropy": 5.9014129638671875,
      "epoch": 6.296985022369189,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018769931067012775,
      "loss": 4.854,
      "mean_token_accuracy": 0.22745656073093415,
      "num_tokens": 140369432.0,
      "step": 80935
    },
    {
      "entropy": 6.006623697280884,
      "epoch": 6.297374051740906,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00018767376703713828,
      "loss": 4.8647,
      "mean_token_accuracy": 0.21761720776557922,
      "num_tokens": 140378443.0,
      "step": 80940
    },
    {
      "entropy": 6.02153639793396,
      "epoch": 6.297763081112624,
      "grad_norm": 1.65625,
      "learning_rate": 0.00018764822472918535,
      "loss": 4.9854,
      "mean_token_accuracy": 0.21534970849752427,
      "num_tokens": 140386393.0,
      "step": 80945
    },
    {
      "entropy": 6.033579969406128,
      "epoch": 6.298152110484342,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018762268374665665,
      "loss": 5.0022,
      "mean_token_accuracy": 0.2141214519739151,
      "num_tokens": 140395104.0,
      "step": 80950
    },
    {
      "entropy": 5.93167929649353,
      "epoch": 6.298541139856059,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018759714408993962,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22129660099744797,
      "num_tokens": 140403363.0,
      "step": 80955
    },
    {
      "entropy": 6.032492828369141,
      "epoch": 6.298930169227777,
      "grad_norm": 1.59375,
      "learning_rate": 0.00018757160575942183,
      "loss": 5.0259,
      "mean_token_accuracy": 0.21868330538272857,
      "num_tokens": 140411803.0,
      "step": 80960
    },
    {
      "entropy": 5.929594469070435,
      "epoch": 6.299319198599494,
      "grad_norm": 1.515625,
      "learning_rate": 0.00018754606875549074,
      "loss": 4.8379,
      "mean_token_accuracy": 0.2273387908935547,
      "num_tokens": 140420906.0,
      "step": 80965
    },
    {
      "entropy": 5.994311714172364,
      "epoch": 6.2997082279712115,
      "grad_norm": 1.484375,
      "learning_rate": 0.00018752053307853384,
      "loss": 4.9432,
      "mean_token_accuracy": 0.22033000290393828,
      "num_tokens": 140429855.0,
      "step": 80970
    },
    {
      "entropy": 5.994245338439941,
      "epoch": 6.300097257342929,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001874949987289386,
      "loss": 4.9027,
      "mean_token_accuracy": 0.21899940818548203,
      "num_tokens": 140439146.0,
      "step": 80975
    },
    {
      "entropy": 5.996577739715576,
      "epoch": 6.300486286714647,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001874694657070925,
      "loss": 4.915,
      "mean_token_accuracy": 0.22597428113222123,
      "num_tokens": 140447273.0,
      "step": 80980
    },
    {
      "entropy": 6.0153467655181885,
      "epoch": 6.300875316086365,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00018744393401338294,
      "loss": 4.9764,
      "mean_token_accuracy": 0.21441719084978103,
      "num_tokens": 140456034.0,
      "step": 80985
    },
    {
      "entropy": 5.9779167652130125,
      "epoch": 6.301264345458082,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00018741840364819735,
      "loss": 4.9403,
      "mean_token_accuracy": 0.2276265099644661,
      "num_tokens": 140464175.0,
      "step": 80990
    },
    {
      "entropy": 5.889973163604736,
      "epoch": 6.3016533748298,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018739287461192307,
      "loss": 4.8545,
      "mean_token_accuracy": 0.22926456183195115,
      "num_tokens": 140472764.0,
      "step": 80995
    },
    {
      "entropy": 5.898661327362061,
      "epoch": 6.302042404201517,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00018736734690494745,
      "loss": 4.8479,
      "mean_token_accuracy": 0.22817789912223815,
      "num_tokens": 140482232.0,
      "step": 81000
    },
    {
      "epoch": 6.302042404201517,
      "eval_entropy": 5.681481358727317,
      "eval_loss": 5.041774272918701,
      "eval_mean_token_accuracy": 0.22452311488988883,
      "eval_num_tokens": 140482232.0,
      "eval_runtime": 21.6624,
      "eval_samples_per_second": 1918.443,
      "eval_steps_per_second": 239.817,
      "step": 81000
    }
  ],
  "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": 2.827511094352896e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}